ABSTRACT. Growing demands for electricity and extreme weather events necessitate comprehensive resilience strategies for energy communities. The Texas power blackout in February 2021 serves as a prime example of the severe impacts of extreme weather on such systems. This paper offers a thorough exploration of this event, investigating its causes, infrastructural constraints, financial implications, and its effects on politics. An in-depth analysis reveals the limitations in generation capacity and the over-reliance on specific power sources such as natural gas, coal, wind, and nuclear power. The paper underscores the necessity for improved regulatory practices, diversification of energy sources, and enhanced preparation for extreme weather conditions to strengthen grid resilience. Financial impacts are also analyzed, including the dramatic rise in electricity costs during the crisis and potential strategies for cost reduction. The paper further delves into the political implications of the blackout, specifically its impact on electoral outcomes. The role of state-wide power infrastructure winterization to enhance resilience is emphasized. The paper concludes with a discussion on potential policy and power system management implications of the findings, contributing to the discourse on improving grid resilience against extreme events.

ABSTRACT. A high altitude electromagnetic pulse (HEMP) or other similar geomagnetic disturbance (GMD) has the potential to severely impact large-scale electric power grids. This is due to the introduction of low-frequency geomagnetically induced currents (GICs), which can saturate the magnetic core of power transformers and lead to distorted ac waveforms, increased losses, and the potential for thermal damage. This paper presents a dynamic model of a 20-bus power system which includes reduced order models of saturating transformers, coupling of HEMP/GMD generated electric fields onto transmission lines, a resistive model of earth for ground conduction of GICs, and the ability to specify conventional and non-conventional HEMP/GMD mitigation strategies via an optimal controls framework. First, a description of the power system model presented along with a comparison of the proposed dynamic model to the original static model. Subsequently, the top-down control framework for designing mitigation schemes is provided, and a solution for system control which avoids transformer saturation during a simulated HEMP/E3b insult is demonstrated. Finally, and conclusion and directions for future research are put forth.

ABSTRACT. The growing penetration of renewable energy requires upgrades to the transmission network to ensure the deliverability of renewable generation. As an efficient alternative to transmission expansion, flexible transmission technologies, whose benefits have been widely studied, can alleviate transmission system congestion and enhance renewable energy integration. However, under the current market structure, investments for these technologies only receive a regulated rate of return, providing little to no incentive for efficient operation. Additionally, a regulated rate of return creates an incentive for building more transmission lines rather than efficient utilization of the existing system. Therefore, investments in flexible transmission technologies remain rather limited. To facilitate the deployment of flexible transmission, improve system efficiency, and accommodate renewable energy integration, a proper incentive structure for flexible transmission technologies, compatible with the current market design, is vital. This paper reviews the current market-based mechanisms for various flexible transmission technologies, including impedance control, dynamic line rating, and transmission switching. This review pinpoints current challenges of the market-based operation of flexible transmission and provides insights for future endeavors in designing efficient price signals for flexible transmission operation.

ABSTRACT. The increase in the frequency of high impact low probability (HILP) events in recent years has sparked the need for addressing power system resilience. Improving a system’s operational resilience compared to the traditional infrastructure resilience is preferred because of two reasons: 1) infrastructure resilience improvement has a substantial initial investment cost associated with it; 2) due to the spatio-temporal nature of resilience metrics, the benefit-cost coefficient is not evenly dis- tributed across all customers. In this paper, an optimal scheduling strategy to improve the operational resilience is proposed using an agent-based deep reinforcement learning (DRL) framework. Specifically, the scheduling problem is formulated as a Markov decision process (MDP), considering the stochastic nature of the loads, and intermittent energy resource (solar PV system). An agent is trained, using deep Q-network (DQN), to take optimal actions for a given state that would provide the best return. Finally, the developed framework is tested during an outage scenario with the numerical results validating the effectiveness of this approach.

ABSTRACT. Power system disturbances, such as significant faults or major disruptions in generation or load, cause imbalance between supply and demand, which may result in severe frequency fluctuations. Following such disturbances, fast frequency response is needed to provide frequency support and avoid system collapse. As such, monitoring and detecting frequency events quickly and precisely is critical. This paper proposes an abnormal frequency deviation detection algorithm that uses two methods to process Phasor Measurement Unit data and declare the frequency events. The first is a signal processing based method that de-noises frequency measurements. This is followed by a second, statistics-based method that calculates rate of change of frequency, mean, variance, and standard deviation. The algorithm uses thresholds to declare frequency events. The proposed algorithm is an improvement over literature-relevant works as it has four tunable threshold parameters. Proper tuning of these threshold parameters enhances algorithm performance, thereby making the algorithm adaptable to different electric power Balancing Authority. The algorithm detection performance is evaluated using binary classification technique and evaluation metrics. The results show the effectiveness of the proposed algorithm for detecting various frequency events of different datasets.

ABSTRACT. In the last few decades, extreme weather events (EWEs) have become more frequent, especially in the southeastern part of the U.S. These EWEs affect the distribution grid greatly, resulting in long-duration power outages. To fully understand the outcomes of these catastrophic events and to correlate the impact on the reliability and resiliency of the utility company, multiple years of historical EWE-related outage data collected from a utility company are analyzed. Moreover, this study utilizes the system average interruption frequency index (SAIFI) matrix to analyze the EWEs. This study finds overgrown vegetation as one of the main reasons for outages during EWEs. Besides, loss of transmission and generation also contribute significantly to outage events resulting in a high percentage of customers being affected for long periods.

ABSTRACT. This paper presents a measurement-based electricity market structure to establish peer-to-peer (P2P) transactions along with imports from or exports to the upstream network. A key benefit of the proposed P2P market is that participants therein can fully express their proclivities by setting their individual preferences for buying and selling partners independently. Moreover, resulting P2P transactions satisfy power flow constraints of the underlying distribution system without needing an offline network model. Instead, we estimate a linear sensitivity model mapping bus voltages to injections using only online measurements collected from P2P market participants, which is then embedded as an equality constraint in an optimal power flow (OPF) problem. The OPF problem minimizes total cost of P2P transactions incurred to market participants capturing network usage fees, buying/selling preferences, net import/export cost, and operation cost. The optimal solution of the OPF problem comprises the P2P transactions (specifying partners, quantity, and price for each trade), the optimal dispatch, as well as locational marginal prices at buses where measurements are collected. Via numerical simulations involving a 22-bus test system, we demonstrate the effectiveness of the proposed method to establish P2P transactions that respect individual preferences; we also validate notable properties pertaining to the trade prices.

ABSTRACT. Phasor measurement units (PMUs) play a crucial role in monitoring and analyzing power system dynamics, and their integration into Real-time Digital Simulator (RTDS) environments enables accurate real-time simulation and testing of power systems. Therefore, this paper proposes an experimental setup and a detailed configuration of relevant software to facilitate the streaming of multiple PMUs in RTDS. An experimental setup that facilitates the execution of hundreds of fault scenarios and the acquisition of transient data, is developed with the integration of RTDS, PMU Connection Tester, openPDC, MySQL, and MATLAB. The paper examines the performance and accuracy of the implemented multiple PMU setup through a case study and experimental results of fault location detection (FLD). A convolutional neural network (CNN) is trained with the high-precision transient data gathered using GTNET PMUs for FLD. A modified IEEE-$9$ bus system integrated with PMUs is taken as a test system for demonstration. The significance of this study is to provide researchers with a detailed configuration for all the necessary software and a reliable framework to stream multiple PMUs in RTDS, store data in an SQL database, and use it for various power system applications.

ABSTRACT. Power system dynamics are generally modeled by high dimensional nonlinear differential-algebraic equations due to a large number of generators, loads, and transmission lines. Thus, its computational complexity grows exponentially with the system size. This paper demonstrates the potential use of quantum computing algorithms to model the power system dynamics. Leveraging a symbolic programming framework, we equivalently convert the power system dynamics' differential-algebraic equations (DAEs) into ordinary differential equations (ODEs), where the data of the state vector can be encoded into quantum computers via amplitude encoding. The system's nonlinearity is captured by Taylor polynomial expansion, the quantum state tensor, and Hamiltonian simulation, whereas state variables can be updated by a quantum linear equation solver. Our results show that quantum computing can simulate the dynamics of the power system with high accuracy, whereas its complexity is polynomial in the logarithm of the system dimension. Our work also illustrates the use of scientific machine learning tools for implementing scientific computing concepts, e.g., Taylor expansion, DAEs/ODEs transform, and quantum computing solver, in the field of power engineering.

ABSTRACT. The transient stability analysis (TSA) is a major numerical function in Energy Management System for large-scale power transmission system planning and evaluation. Ideally, the trajectory of system dynamics, such as bus voltage magnitude and generator phase angle, can be predicted to forecast issues and disturbances based on a time-domain solution of differential and algebraic equations (DAEs). However, the rise of system scales and complex machine models requires advanced computing techniques to achieve even faster than real-time (FTRT) criteria. For accelerating the execution, High-Performance Computing (HPC) on traditional CPU-based supercomputing clusters has been widely investigated for decades. Recently, as heterogeneous computing was introduced to power system domain areas, general-purpose computing devices such as graphics processing units (GPU) have been deployed to enhance computational performance and program portability further. This paper reviews the trials within the last 15 years on developing parallel TSA applications regarding mathematical formulations, programming approaches, and performance. We also address the future of TSA computation and its potential to improve the efficiency of FTRT executions.

ABSTRACT. The solar power plant is a large-scale photovoltaic (PV) system aimed to generate solar power for the electricity grid. It includes PV arrays (PVAs), cables, and other electrical accessories. Moreover, the builder of solar power plants has to consider various parameters and design regulations. The cable routing problem (CRP) is critically important in large-scale solar power plant design. The objective of our CRP is to minimize the installation cost of the cable by determining the partition of the photovoltaic array and the cable routing. In this study, we use the quantum computer to solve the CRP which is an NP-hard integer linear programming (ILP) problem and show its advantages over classic computers. We transfer the ILP CRP into the quadratic unconstrained binary optimization (QUBO) model and solve it by the advanced quantum annealer. Finally, we analyze the computational results and discuss the advantage of our approach to solving the CRP.

ABSTRACT. This paper proposes a stochastic optimal strategy for equitable proactive actions of distribution grids against progressive wildfires to enhance the operational resilience. The paper first models the impact of a wildfire on power distribution systems. To this end, it mathematically formulates the impact of wildfire and generated heat on powerlines’ capacity and status by employing non-steady state dynamic heat balance equations. It also models the effect of wildfire byproducts including heat, smoke, and ash on renewable power generation (i.e., solar and wind). Then, it develops a stochastic optimization problem which incorporates the impacts of wildfire on the system as well as physical, operational, and environmental constraints. An objective function is introduced to consider both grid resilience and equity criteria during system contingencies. The effectiveness of the proposed approach is tested over the standard unbalanced IEEE 37-bus benchmark, showing the effectiveness of the proposed strategy.

ABSTRACT. The aim of distribution networks is to meet their local area power demand with maximum reliability. As the electricity consumption tends to increase every year, limited line thermal capacity can lead to network congestion. Continuous development and upgradation of the distribution network is thus required to meet the energy demand, which poses a significant increase in cost. The objective of this research is to analyze distribution network topologies and introduce a topology reconfiguration scheme based on the cost and demand of electricity. Traditional electrical distribution networks are static and inefficient. To make the network active, an optimal dynamic network topology reconfiguration (DNTR) is proposed to control line switching and reconnect some loads to different substations such that the cost of electricity can be minimized. The proposed DNTR strategy was tested on a synthetic radial distribution network with three substations each connecting to an IEEE 13-bus system. Simulation results demonstrated significant cost saving in daily operations of this distribution system.

ABSTRACT. Distributed generators (DGs), especially inverter based renewable generation sources, are becoming prevalent in power distribution grids to meet increasing demand and harness renewable energy. However, DGs of intermittent nature such as solar power can lead to significant voltage fluctuations. Traditional mechanical voltage regulating equipment such as on-load tap changers (OLTCs) and capacitor banks (CBs) are unable to handle these fast-changing fluctuations. This paper proposes a new optimal cooperative control-based method for controlling DG inverters to adjust reactive power generation or consumption of DGs. The optimization is based on local information and the information received from other nodes with or without reactive power sources. The objective function is to minimize voltage fluctuation and reactive power output since additional reactive power output could cause accelerated aging of inverters. Results based on simulation studies have shown that the proposed method can achieve smoother voltage profiles.

ABSTRACT. The electricity wholesale market is very dynamic as the supply and demand need to be matched to maintain stability in the national power grid. In recent years, the addition of renewable sources of energy, such as solar and wind energy, to the grid has introduced further instability to the system, making the prices more volatile. In such scenarios, there is a high need for accurate data-driven computational techniques to forecast the price of electricity in the wholesale market ahead of time. The forecasts can be hourly, daily, weekly, or monthly, but this paper focuses on hour ahead forecast of electricity prices, and the framework can be generalized to any time frame.

ABSTRACT. With growing interdependence among various energy forms, such as electricity, heat, and cooling, multi-energy system (MES) is playing an increasingly important role. As a focal point in MES, energy hub (EH) needs to be properly planned for increased efficiency of energy utilization. This paper presents a generic mixed-integer linear model for EH planning to minimize the overall investment and operation cost given the demand and price of each energy. Our approach uses a graph with multiple rows and layers to represent the energy conversion topology in an EH. In this work, we formulate the EH optimal planning problem with a single combined model to implement the following two steps: 1) optimize the device investments represented in system topology graph; 2) manage the energy conversion and flow to meet the end-use demands meanwhile minimizing the cost. The numerical study shows the efficacy of our proposed model.

ABSTRACT. Electric vehicles (EVs) have been ‘just around the corner’ for over 100 years, but it appears that they finally may be on the near horizon for wide scale deployment. At least, this may be said considering recent developments in California. This paper describes a state resolution in California to prohibit the sale of gasoline powered automobiles and light trucks by 2030, and the expected impact of this resolution on electric power, energy, and load factor. The approach taken for this impact assessment is statistical and probabilistic with particular attention to the statewide annual load factor. Data on expected EV deployment are given for 2023 – 2030. Since the power system load factor is a ratio of two quantities, the statistics of the load factor are characterized as a newly developed probability density function that models the system load factor as a ratio distribution. Some results, examples and preliminary conclusions are drawn.

ABSTRACT. Solar PV-storage hybrid systems are an attractive way of reducing grid dependency for project developers when collectively serving small communities. A well-structured charging schedule of the battery storage can improve the effectiveness of the hybrid system by reducing the grid dependency at the peak demand hours and improving the utilization of the battery. Studying the correlation between load and solar helps to identify periods at which short-term charging or discharging of the battery can aid to those benefits. In this paper, a dynamic charging schedule is proposed, where a nominal schedule is created based on full day forecasting of load and solar, and further adjustments are made to the schedule based on the hourly expected correlation between solar and load. The proposed method is then tested on a test system based on actual data for a period of one year. The results show that the proposed approach can reduce the maximum power import from the grid, overall cost of imports and improve the battery utilization.

ABSTRACT. Decarbonizing the power sector is key for mitigating climate change. However, as the grids integrate higher capacities of variable renewable energy, supply-demand mismatches and increased curtailment are challenging their further adoption. In this paper, two technological tools are studied to reduce these challenges: long-duration energy storage (LDES) and oxy-combustion. For modelling high renewable penetration grids,such as California, a capacity expansion model is used. For the modelled years 2030, 2035, 2040 and 2045, the impact of the availability of 100-h LDES and oxy-combustion is studied.

ABSTRACT. This research investigates the implications of light-duty electric vehicle (EV) charging patterns on energy storage requirements for California. The study examines three plausible EV charging profiles and incorporates energy storage technologies with 4-, 8-, and 12-hour durations. Assuming that light-duty EVs are charged with solar power coupled with storage, changing the plausible charging profile varies the needed storage (as reflected by the power rating, energy rating, and modeled cost) by up to 80%. Remarkably, the relative differences in the amount of the added storage among the three cases decrease as the model switches to deploy longer-duration instead of 4-hour storage, suggesting the potential to reduce the impact of different load patterns and a reduced necessity for investments in public charging station infrastructure. Specifically, 8-hour storage provides a good trade-off between the needed power and energy storage capacity and establishing extensive public charging infrastructure.

ABSTRACT. This paper introduces a block-box Frequency Spectrum Attack (FSA) effective on machine learning time-series forecasting. FSA transforms the time-series signal into frequency domain by performing Fast Fourier Transform (FFT). Next, FSA manipulates the amplitudes of specific number of dominant frequencies randomly to ensure the attack is undetectable. FSA is tested on Long short-term memory (LSTM) network to investigate the effectiveness of the proposed attack on the state-of-the-art technology in forecasting. The results showed that FSA increases the Mean Absolute Error Per Recording (MAEPR) in a time-series load forecasting three times higher than the normal condition, and by a raise of 70% in amount compared to a noise injection attack scenario. Finally, the impact of FSA on a microgrid's Energy Management System (EMS) is investigated. The microgrid's EMS includes load forecasting, EMS, and ideal microgrid units. We showed that FSA degrades the EMS's performance by increasing the utilization of the battery storage output power by 45%. The results showed that FSA is not detectable by frequency sensors of the microgrid.

ABSTRACT. The constantly evolving state of the power grid calls for some power quality-related measures to be reevaluated. Current measures in place to ensure acceptable levels of power quality only consider disturbances, such as harmonics and source unbalance, individually without taking into account their effects on one another. The work in this paper is dedicated to finding trends and correlations between source unbalance and harmonic emissions of an AC-DC rectifier by using a set of simulations. Rectifiers with two different load conditions are supplied by sources with up to 5% unbalance. The sources had their magnitudes and phase angles adjusted at random for 500 simulations per load condition. THD as well as all harmonics up to the 20th for each phase from each simulation were recorded and evaluated. Equations defining the upper and lower limits of THD and each odd harmonic as a function of source unbalance were developed. Equations like these can begin to be considered in power quality standardization related to harmonics and source unbalance.

ABSTRACT. In this paper, a highly compact, low power (<10W), high frequency (2 MHz) isolated active clamp forward converter, comprising a coreless Printed Circuit Board-based transformer is proposed. To decrease the size of converter, high switching frequency is considered which lead to decrease in inductor, capacitor and transformer size. Highly switch loss due to hard switching is an important constraint of forward topology to increase frequency. In this paper, the active clamp circuit is added to forward topology to achieve zero voltage switching and decrease switching loss drastically. Due to zero voltage switching, the proposed converter can operate in high frequency. The principle of active clamp forward converter is described in this paper. Another constraint to increase the switching frequency of forward converter is transformer core losses. In this paper, coreless PCB-based transformer is proposed and implemented to be utilized in the structure of the active clamp forward converter. Instead of classic core-based transformer, using a PCB-based transformer as the power transmitter has increased the efficiency due to elimination of core hysteresis loss. The equivalent circuit, transfer function and input impedance of PCB-based coreless transformer are presented in high frequency. Finally, an experimental prototype of the active clamp forward converter which uses a coreless transformer is implemented. The experimental results of proposed converter are presented to evaluate the theoretical analysis and performance.

ABSTRACT. This paper presents the dynamic modeling of a voltage source inverter (VSI) grid-connected system with high-performance direct power control (DPC) functions. The system is modeled in the stationary reference frame to reduce complexity and make it linear time-invariant, enhancing control flexibility. The DPC is introduced to achieve fast reference tracking and decoupling regulation of grid active and reactive power injection. Simulation results demonstrate that the DPC exhibits a fast dynamic response and adaptable to changes in grid parasitic elements, even under weak grid conditions with short-circuit ratio of 1.8. The insights and discussions provided in this paper will be valuable for researchers, engineers, and professionals in the industry working on VSI-grid integration and developing control strategies.

ABSTRACT. This paper compares a single source boosted bipolar PWM half-bridge inverter with a single source boosted multilevel half-bridge inverter. Output voltage and current rms and THD, and power efficiency of each topology are measured, analyzed, and compared under the same variable load but different modulation schemes to evaluate their performance. Both topologies employ a boost converter to boost their single DC source and regulate their output voltage using a PI controller. Simulation results show that half-bridge MLI delivers better current rms, voltage THD and power efficiency but bipolar half-bridge inverter offers better quality of output current waveform. The paper, therefore, concludes that both topologies provide benefits in suitable applications.

ABSTRACT. In many applications, wide bandgap power converters are replacing old-fashioned power converters, including aerospace, transportation, smart grid, etc., where their sensing systems require higher bandwidths for diagnostics, prognostics, and control than conventional power electronics. Considering the layout of WGB power converters, speed is not the only factor to take into account when determining the size and invasiveness of current sensors. Existing commercial current sensors lack many of the characteristics required by power converters, including size, speed, noise immunity, accuracy, linearity, high capacity, isolation, and non-invasiveness. Due to the size and cost of current probes, these converters cannot accommodate them as well. Hence, a precise, noise-immune light-size, non-invasive ultrafast sensor would be ideal for such applications in power converters. In this paper, we provide feasible guidelines to design current sensors that can be suitable candidates to be applied with desirable characteristics, utilizing new PCB-integrable methods and technologies. A final set of experiments will be conducted to evaluate their potential for such sensitive integrations.

ABSTRACT. This paper compares a single source boosted multilevel half-bridge inverter with a single source boosted multilevel H-bridge inverter. A comparative analysis is conducted to compare the performance of the half-bridge MLI with H-bridge MLI by evaluating and analyzing their respective output voltage and current rms, output voltage and current %THD, and power efficiency. Both topologies are supplied with the same PI output voltage controller to control the output voltage and share the same non-PWM square wave inverter gate drive. Simulation results reveal that half-bridge MLI outperforms H-bridge MLI by yielding better results for performance parameters in response to variable and fixed loads. The paper therefore concludes that the half-bridge MLI is more power efficient and delivers better power quality than that of the H-bridge MLI.

ABSTRACT. It is expected that the number of renewable energy sources will continue to increase offering unique challenges to the system operator for reliability evaluation. Increase in load growth along with heavy renewable penetration will make it critical to ensure the system can deliver power to the customers. Due to fundamental difference between conventional and renewable generation existing reliability evaluation approach may no longer be sufficient. This will make it necessary to include transmission line failure in reliability evaluation if there is a significant contribution from renewable energy. The paper highlights the importance of including transmission line outage in calculation of LOLE to evaluate reliability of a transmission network and the impact of lines when there is only conventional generation network is compared with conventional and wind generation

ABSTRACT. Transmission expansion planning (TEP) plays a vital role in ensuring the reliable and efficient operation of power systems, especially with the growing demand for electricity and the integration of renewable energy sources. This paper focuses on applying unconventional high surge impedance loading (HSIL) lines in transmission expansion planning and compares their outcomes with conventional line-based transmission expansion planning. Starting with a 17 bus- 500 kV power system connected by a conventional transmission line, the objective is to connect a new load located in a new bus, bus #18, to the existing 17-bus power system via two approaches: using conventional lines and incorporating unconventional HSIL lines. By comparing the number of lines required for the conventional and unconventional approaches, maintaining almost identical conductor volume per circuit, the effectiveness of unconventional HSIL lines in TEP is evaluated where using only two unconventional HSIL lines is sufficient to connect 1000 MW load demand at bus 18 while three transmission lines are required when using the conventional line design.

ABSTRACT. This paper presents a 17-bus 500 kV test system for transmission expansion planning (TEP) studies. An actual 500 kV transmission line geometry was used for the overhead lines of this system. Although many test systems have been introduced for different types of power system analysis, those especially for TEP studies at a transmission voltage level, not distribution voltage level, are few. To the best of our knowledge, the introduced test systems for TEP studies, either those combined with electricity market problems or those used to connect a new load or generation to an existing power grid, consider the studies under only normal condition. However, for TEP studies it is needed that a test system meets voltage drop and line loading limits criteria under normal condition as well as all single contingencies, and in this regard, addressing the latter, all single contingencies, is challenging. This paper addresses this technical gap, introducing a 17-bus test system at a transmission voltage level, 500 kV, that meets requirements under normal condition as well as all single contingencies. In addition to presenting all details of this new test system, load flow results under normal condition as well as the worst single contigency are presented. For studies on the TEP, this test system can be an invaluable resource.

ABSTRACT. This paper develops a new concept that we call transmission expansion planning (TEP)-based unconventional high surge impedance loading (HSIL) line design. To date, these two areas (TEP and transmission line design) have been conducted separately. For TEP, planners typically use the electrical parameters of a few standard conventional line designs to study planning scenarios, and then, the final candidate line is constructed. In such a sequence, cost-effective scenarios often do not meet the technical criteria of load flow. In this paper, we will study whether this sequence can be overturned; namely, can a transmission expansion planner get optimal line parameter values that lead to the most cost-effective scenario, and then have a transmission line with those parameters be designed? Although this cannot currently be realized through conventional designs, in this paper, we demonstrate that it is a possibility if breakthrough designs for transmission lines are used by shifting phase configurations and subconductors into unconventional HSIL arrangements, leading to the optimal line parameters determined by TEP.

ABSTRACT. Transmission expansion planning (TEP) is crucial for maintaining the reliable and efficient operation of the power systems, particularly in the face of increasing electricity demand and the integration of renewable energy sources. This paper aims to investigate the application of unconventional high surge impedance loading (HSIL) lines in TEP and presents a comparative analysis of their outcomes against conventional line-based TEP approaches. Starting with a 17-bus 500 kV test system, which can operate well under normal operating condition as well as all single contingency conditions, the objective is to connect a new load located in a new bus, bus #18, to the existing test system via two approaches: using conventional lines and incorporating unconventional HSIL lines. By comparing the number of lines required for the conventional and unconventional approaches, maintaining identical conductor weight per circuit, the effectiveness of unconventional HSIL lines in TEP is evaluated where using only two unconventional HSIL lines is sufficient to connect 1250 MW load demand at bus 18 while three transmission lines are required when using the conventional line. Finally, a thorough economic analysis has been conducted on both TEP scenarios, revealing that implementing unconventional HSIL lines leads to remarkable cost savings and thus can be considered a promising option for TEP studies.

ABSTRACT. In conventional lines, subconductors are located symmetrically on an identical circle in each phase. The number of subconductors in a bundle, the radius of the bundle circle, and the radius of each subconductor, are chosen so that the maximum electric field strength on the subconductors, Emax, is limited to the permissible field strength on the conductor surface, Epr, (Emax <= Epr), which is determined by the corona discharge limitation requirement. In this paper, we show that by shifting phase configurations and subconductors into unusual/unconventional arrangements that are geometrically optimized within the space, high power density designs can be achieved. A novel bundle and phase arrangement of a 500 kV transmission line is presented in this paper, resulting in higher natural power than conventional design.

ABSTRACT. A 30kW laboratory scale model electric power grid was re-engineered to use more modern distributed data acquisition in the form of a Multi Agent System (MAS) with a laboratory-wide timing pulse to serve for measurement synchronization similar to GPS signals used in industrial synchrophasor equipment. The scale model power grid is useful in an electric power academic environment for teaching concepts, providing a hands-on tool for students, and to serve as a testbed for researching new ideas in the ever evolving “smart grid”. Various elements of the design of the MAS Smart grid are presented highlighting unique circuit designs, a state machine control algorithm, distributed timing pulse synchronization, and Smart Meter protection. Three test cases of the grid are presented highlighting autonomous generator connections, synchronized measurements, and power flows.

ABSTRACT. To ensure safe, reliable operation of the electrical grid, we must be able to predict and mitigate likely failures, leading to the classic security-constrained AC optimal power flow (SCOPF) problem. A common solution method for SCOPF is adversarial optimization, where the dispatcher and an adversary take turns optimizing a robust dispatch and adversarial attack, respectively. We show that adversarial optimization is liable to severely overestimate the robustness of the optimized dispatch, leading the operator to falsely believe that their dispatch is secure.

To prevent this overconfidence, we develop a novel adversarial sampling approach that prioritizes diversity in the predicted attacks. We find that our method not only substantially improves the robustness of the optimized dispatch but also avoids overconfidence, accurately characterizing the likelihood of voltage collapse under a given threat model. We demonstrate a proof-of-concept on small-scale transmission systems with 14 and 57 nodes.

ABSTRACT. Geomagnetic disturbances (GMDs) threaten the grid through geomagnetically induced currents (GICs), which saturate transformers, causing operational effects such as voltage stability issues, potentially leading to load curtailment and, in extreme cases, grid blackout. GMDs pose a severe threat to system reliability, and it is imperative to model the effects of GMDs in the reliability assessment. Hence, this paper proposes an integrated reliability assessment framework wherein the GMD effects are included by adding a GMD reliability module to the generally accepted reliability assessment framework. The paper addresses the first subprocess in the integrated framework - reliability modeling of GMDs. A way to characterize the GMD storms in the context of reliability analysis is shown by introducing three parameters (TTGMD, GMDT, and GMDC) that model the storms' frequency, duration, and intensity. Over 90 years of historical geomagnetic data were processed, and historical observations for TTGMD, GMDT, and GMDC were obtained. An automatic fitter procedure then fits the historical data to probability distributions culminating in the initial steps for developing a GMD-integrated reliability assessment framework.

ABSTRACT. AC optimal power flow (AC OPF) is a fundamental problem in power system operation and control. Accurately modeling the network physics via the AC power flow equations makes AC OPF a challenging nonconvex problem that results in significant computational challenges. To search for global optima, recent research has developed a variety of convex relaxations to bound the optimal objective values of AC OPF problems. However, the quality of these bounds varies for different test cases, suggesting that OPF problems exhibit a range of difficulties. Understanding this range of difficulty is helpful for improving relaxation algorithms. Power grids are naturally represented as graphs, with buses as nodes and power lines as edges. Graph theory offers various methods to measure power grid graphs, enabling researchers to characterize system structure and optimize algorithms. Leveraging graph theory-based algorithms, this paper presents an empirical study aiming to find correlations between optimality gaps and local structures in the underlying test case’s graph. Network graphlets, which are induced subgraphs of a network, are used to investigate the correlation between power system topology and OPF relaxation tightness. Specifically, this paper examines how the existence of particular graphlets that are either too frequent or infrequent in the power system graph affects the tightness of the OPF convex relaxation. Numerous test cases are analyzed from a local structural perspective to establish a correlation between their topology and their OPF convex relaxation tightness.

ABSTRACT. A high altitude electromagnetic pulse (HEMP) or other high intensity geomagnetic disturbance (GMD) has the potential to severely impact the operation of large-scale electric power grids. This is due to the introduction of low-frequency geomagnetically induced currents (GICs), which can saturate the magnetic core of large power transformers and lead to distorted ac waveforms, increased losses, and the potential for cascading blackouts. This paper presents a controls-based strategy for protecting transformers against such threats. Analysis of two different HEMP/GMD protection approaches are evaluated: first, a neutral-path blocking approach, and second, a magnetizing-path flux cancellation approach. The relative trade-offs between the two techniques are provided, and a demonstration of their application in a representative 4-transformer grid scenario is considered. Finally, conclusions and directions for future research are proposed. 

ABSTRACT. Due to thermal, electrical, mechanical, and chemical stresses, line-post insulators in the power system may degrade over time. The degradation process continuously gets exacerbated by the above-mentioned factors. Therefore, condition monitoring of line insulators must be frequently carried out. Optical cameras are considered the most accurate among existing technologies for detecting such defects. Computer vision techniques aided by optical cameras could automate faulty insulator identification. However, there is a limited size of the training data set obtained from real-world optical camera images. In this paper, we propose a generative approach to creating a massive amount of line-post insulator fault images through Deep Convolutional Generative Adversarial Networks (DCGAN). The additional training data obtained from DCGAN-based approach is shown to improve the accuracy of the insulator fault classification. In the case study, we show that with an increasing number of synthetic images created by DCGAN, the accuracy of the fault classification continuously improves. The ability to classify true faulty insulators has increased from 56\% to 94\%. The performance of the DCGAN-based approach is also compared with the random oversampling approach. The numerical results suggest that the DCGAN-based approach has the advantage of detection accuracy and a lower false positive rate.

ABSTRACT. This paper introduces Artificial intelligence (AI)-enabled anomaly-aware occupancy prediction methods to improve energy efficiency and demand response capability in grid-interactive efficient buildings (GEBs), while ensuring occupant comfort. The paper first introduces an Artificial intelligence (AI) method based on Long Short-Term Memory Auto-Encode (LSTM AE) for anomaly detection in time series data obtained from Internet of Things (IoT) sensors in GEBs. The AI-enabled method helps the building management system with detecting anomalies early on, reducing the risk of cascading failures. Isolating anomalous sensors and managing the on-site resources based on accurate data can increase building’s reliability. The anomaly detection can also improve the accuracy of occupant activity prediction in GEBs. LSTM and graph convolution networks (GCN)-LSTM models are used and compared to forecast occupant behavior on both raw data with anomalies and accurate data. Validation and assessment are conducted using data gathered from more than 1,100 IoT sensors located across a large academic building. The power consumption of the sensors used as a metric for anomaly detection.

ABSTRACT. The availability of fine granularity electricity consumption data is critical for managing building energy usage and designing efficient electric retrofits. Methods have been developed for producing year-long hourly electric load profiles of residential and commercial buildings without the need for smart meter measurements, which however may incur privacy, data inaccuracy, and security concerns. Many of these techniques are built upon monthly utility bills, some leveraging multiple time-of-use intervals. This work proposed an adaptive building electric load profiling technique, which improves upon the limitations of existing work by introducing a transition period that is not always included in the utility bills. It also considers the impacts of seasonal weather changes, which are especially non-neglectable to fully electric buildings. Specifically, profiling for solar photovoltaic (PV) dominated buildings is also studied by leveraging accurate PV system energy production profiles. The proposed profiling method is tested on a gas-heated building and a fully electric building in the State of New Jersey. Results show that the gas-heated building exhibits better profiling performance compared to the fully electric building, whose electric load is more sensitive to environmental temperature changes, resulting in errors outside of the acceptable error thresholds during shoulder seasons. However, this may be acceptable as shoulder seasons do not meaningfully impact electric retrofits.

ABSTRACT. Utilities analyze outage data to improve the reliability of their power distribution networks. Historical outage data support investment planning, developing mitigation strategies, and reporting of reliability metrics. However, outage data often exhibit dependencies and correlations between different outage causes and grid configurations that are challenging to understand. This paper demonstrates an innovative approach to comprehensive reliability analysis through modeling a power distribution circuit that has historically impacted reliability in Synergi Electric. The reliability analysis tool in Synergi Electric facilitates reporting of reliability metrics, root cause analysis, and visual mapping of outage data. Engineering solutions are analyzed and proposed as recommended mitigation measures to enhance reliability performance. Simulation case studies on one substation with multiple feeders have demonstrated the effectiveness and efficiency of the proposed mitigation methods.

ABSTRACT. Power system operation is largely dependent on the substations operation. Substation operation requires fast monitoring due to a large amount of data, complexity of the data flow and anticipated control action to be taken in case of any abnormalities. Therefore, there is a need for Substation Automation System (SAS) since it is not possible to track the high- speed data manually and take a fast control action. SAS uses protocols to communicate between the physical device and Supervisory Control and Data Acquisition (SCADA). Data gathering is also performed by these protocols algorithms. As the utility industries facility are becoming more and more complex, the need for more sophisticated automation is realized. The substation automation protocol most widely used in Europe is IEC-61850 whereas American utilities use DNP-3. The purpose of this paper is to emphasize the differences and similarities between these two most widely used protocols in the utilities in the modern world.

ABSTRACT. This paper proposes a TSO-DSO framework including DER set-point dispatch using sensitivity analysis. We address all three steps of TSO-DSO integration stating by estimating the capability curve of an unbalanced distribution system and then modeling of capability curve on the transmission side grid and estimating the reactive power needed from DSO during emergencies, and ending by dispatching the requested reactive power from TSO. The proposed interaction framework is tested on an IEEE 37-node distribution network connected to IEEE 9 bus transmission system. The simulation shows that using sensitivity-aware dispatch would reduce the number of communication signals between DERs and DSO and that can decrease the risk of communication contingency.

ABSTRACT. In this research paper, an optimized fuzzy adaptive distributed secondary controller is proposed to address frequency deviations in Micro Grids (MG) caused by the primary controller. The performance of the secondary controller has been improved by optimizing the parameters of the fuzzy logic controller (FLC) using a combined genetic algorithm (GA) and particle swarm optimization (PSO). The proposed FLC uses the frequency variation and its derivative as inputs and the parameter of the secondary controller as the output. In this paper, the range of changes in fuzzy rules and fuzzy membership is defined as an optimization problem and optimized by the proposed algorithm. The effectiveness of the suggested controller is evaluated by conducting simulations with Simulink. Multiple scenarios are simulated, and the results indicate that the optimized secondary controller successfully reduces both the frequency overshoot and response time

ABSTRACT. The offshore oil and natural gas platforms, mostly powered by diesel or gas generators, consume approximately 16TWh of electricity worldwide per year, which emits large amount of CO2. To limit their contribution to climate change, a proposed solution is to replace the traditional fossil fuel based energy resources with offshore clean energy. One of the main challenges in designing such a system is to ensure that energy demand is met while minimizing cost and reducing environmental impact. To address this challenge, several strategies including microgrid systems consisting of offshore wind turbines, wave energy converters, tidal energy converters, floating photovoltaic systems and battery energy storage systems are being proposed. In this paper, cost optimization for sizing these renewable energy sources are investigated. A cost optimization renewable sizing (CORS) model is proposed to optimize the sizes of the generation and storage resources. The proposed CORS model considers the variability of the power outputs of various renewable energy sources and load, as well as the cost of different generation technologies and the energy storage system. Simulations conducted on three test systems show the proposed resource sizing method significantly reduces the total lifetime cost of energy while maintaining a high level of reliability and sustainability.

ABSTRACT. In this paper, an Extended Kalman Filter (EKF) based control algorithm is proposed for a telecom microgrid. The system loads are instantaneous constant power loads (CPLs). The proposed controller minimizes the number of current sensors in the system. The only state variable that is measured is the microgrid voltage. Without using full state feedback, the proposed controller attains the following goals. Damping of limit cycle oscillations due to CPL in the converter currents and voltages. Current sharing among the parallel connected sources. Regulation of load voltage in the event of source voltage or load power change. Since many renewable sources require a current source interface, parallel connected boost topology is chosen. The EKF based controller design is proposed and expressions for equilibrium points are derived. Stability analysis of equilibrium points is performed. Using simulation results, perfect tracking of state variables is shown both in open loop and closed loop. Further, simulation results also show the validity of the proposed controller.

ABSTRACT. Accurate prediction of solar irradiance is crucial for efficient energy use in building heating/cooling systems, especially when coupled with rooftop solar panels. In areas with high variability in solar irradiance, such as islands like Puerto Rico, it is challenging to plan electricity usage in the face of outages, such as those caused by natural disasters like hurricanes. The goal of this project is to design computationally accelerated methods to downscale nationally available climate data, specifically solar irradiance, to forecast micro-climate data for building-local areas in western Puerto Rico. Downscaling is a method that relates low-resolution global climate models, which are complex and computationally expensive, to local-scale data. In this case, spatiotemporal data from multiple nearby weather stations and the global weather forecast are used to design highly accurate building-local models in Puerto Rico, which is a hurricane-prone region. Specifically, the project uses low-power sensors, small local computing resources, and an empirical downscaling method to derive quantitative relationships between the global and local data. This is achieved by using a computationally efficient (and parallelizable) nearest neighbor Gaussian process (NNGP). The graphics processing unit (GPU) acceleration is then used to speed up the NNGP. The results of this project come from different models created from local and global climate data, with each model pertaining to different seasons in Puerto Rico. Using the known location, times, and dates, each model predicts the solar irradiance, resulting in micro-climate data with higher accuracy due to more data points. The applicability of the NNGP on embedded computers was also validated on an NVIDIA Jetson Nano. Predicting micro-climate data can have various other applications, such as improving the performance of individual plants and other organisms in agriculture.

ABSTRACT. Restoring power distribution and communication networks after a natural disaster poses a significant challenge for utilities and the research community. The interdependence between the power system and communication infrastructure adds complexity, as communication networks are vital for power system monitoring, control and operation. Existing restoration strategies have primarily focused on power distribution infrastructure, overlooking the interdependence with communication systems. This paper introduces a novel three-dimensional Markov Chain model that can be used to understand the effects of restoration strategies. The proposed approach considers all possible failure and recovery scenarios, explicitly addressing the interdependence between power distribution and communication networks (PDCM). The results demonstrate the time dynamics of load restoration by showing the importance of restoring distributed generators during microgrid operations, and the need for restoring communications so microgrids can be recombined once the distribution feeder is energized.

ABSTRACT. This paper models the irradiance beneath solar photovoltaic (PV) panels, which is essential in agrivoltaics applications for selecting the appropriate crop types. To this end, this paper presents a model-approach to calculating the irradiance on a virtual surface beneath a PV panel at several heights. The model employs view factor techniques to address the shading due to the solar panel and reflection from the ground. The modelled results are in agreeable match with the PVsyst results, verifying the utility of the proposed modelling approach. Furthermore, results regarding the dependence of the irradiance profile on

ABSTRACT. Rising temperatures from climate change will have significant impacts on users’ energy consumption behaviors and the resulting electricity demand. As New York State (NYS) is moving towards a 100% decarbonized electricity system by 2040, the long-term load demand needs to be forecasted for generation and transmission planning. To this end, this study analyzes the historical meteorological data to quantify the geo-spatially and seasonably rising temperature at each load zone across NYS. The rising temperatures together with NYS zonal load data, demographic distribution data, and economic trends, are used to train and develop a linear regression model to forecast the long-term load demand. Study results show that due to climate change, 1) NYS experienced an average temperature increase of 0.6234 degrees Fahrenheit/decade from 1980-2020, 2) 0.82% mean load demand increase is expected from 2021 to 2050, and 3) 2,773MW summer peak load demand increase is expected from 2021 to 2050.

ABSTRACT. As the global trend towards decarbonization and decentralization of power systems accelerates, the adoption of solar photovoltaic (PV) systems has seen a significant increase at both utility and residential levels. However, the reliability of power grids in colder regions, particularly those with heavy snowfall, could be at risk due to substantial reductions in solar power generation caused by snow cover. The performance of PV systems can be adversely affected by snow for prolonged periods, and the rate of recovery is primarily dependent on post-snow climatic conditions. This research aims to quantify the impact of snow on PV generation considering meteorological factors in the days following a snow event. A data-driven analysis is presented, with empirical data employed to evaluate the influence of snowfall on PV system performance. Additionally, a multilinear regression model is proposed to predict PV solar power generation under snow-affected conditions. This study not only provides a better understanding of the interaction between snowfall and PV systems but also offers a novel forecasting tool that aids in enhancing the resilience and reliability of power grids in cold regions.

ABSTRACT. Novel methods are required to properly evaluate energy resiliency of military bases. Previous approaches have investigated either quantitative monetary metrics or subjective metrics based on qualitative observations. This research looks to bridge the gap and combine the two by developing a quantitative resiliency metric that can be applied to various types of noncommercial installations. The resiliency metric in this work involves a three-tiered load priority scheme and a quantitative cost to value of lost load suitable for military installations. The approach was applied to a Vermont Air National Guard (VTANG) base. The base demand and energy generation was modeled, a collection of resiliency solutions proposed and simulated, and the resiliency metrics evaluated for each solution. Initial results on a 10-day power grid outage showed that diesel generation is the most resilient. However, a battery energy storage system coupled with the existing photovoltaic infrastructure can achieve comparable resiliency for shorter outages.

ABSTRACT. This paper introduces a cyber digital twin (CDT) for supporting analysis of power system substation network security and validating solutions utilizing Machine learning and Software Defined Networking (SDN). This work focuses on targeted attacks against the Generic Object Oriented Substation Event (GOOSE) protocol. The developed CDT mirrors a realistic substation network consisting of a real-time digital simulator (RTDS), industry-graded hardware relays, merging units, SDN switches, and related communication networks. The CDT consists of a virtual network to which external devices can connect, an open-source SDN flow controller to specify network packet destination, OpenFlow protocol to configure packet flows, a custom-built GOOSE parser and builder, and an open-source library to facilitate GOOSE message exchange between external hosts. Two use cases of GOOSE spoofing attacks were conducted, which were successfully detected using novel feature engineering and a machine learning approach, yielding an accuracy of 95%, precision of 96%, and F-1 score of 96%. Finally, SDN was introduced as a potential solution to mitigate such GOOSE attacks from the bottom up. The overall study underlines the proposed CDT for substation network design's efficacy in improving overall network resilience against GOOSE integrity attacks, as well as its potential for developing and validating solutions for cyber attack defense and mitigation for real-world systems.

ABSTRACT. Geomagnetically induced currents (GICs) have the potential to be highly disruptive to regular power grid operations. Despite the study of geomagnetic disturbances (GMDs) since the 1940s, comprehensive methods for accurate and precise analysis of their impact, especially in individual grid components, remain limited. This paper describes a method based on the synergistic use of different data-driven and physics-based computational tools for detailed simulation of GMDs in power grids. The proposed methodology allows the effective integration between three main aspects required for accurate determination of GICs and their effects on power systems: (1) data-driven and location-specific estimation of Earth’s magnetic fields at ground level, (2) detailed physics-based ground impedance modeling considering specific geological environment, and (3) power grid modeling considering detailed network and grounding topology. The goal is to generate an end-to-end methodology for GIC calculation in power grids from the magnetic field environment during a geomagnetic event, the soil conductivity profile at the site of interest, and the grid topology data. The resulting methodology was evaluated by means of a sample case that utilizes magnetic field data from an actual event.

ABSTRACT. Distributed Energy Resources (DERs) are a growing source of electricity within the United States. These systems, often found on the rooftops of residential homes or commercial businesses, are a useful way to reduce demand on transmission lines and create additional revenue for the owners, but they can create vulnerabilities in the security of the grid, due to both a lower regulatory threshold, and their more decentralized nature. Their risk only grows when. This paper will analyze the current cybersecurity systems in pace for DERs, as well as potential weaknesses in them that could pose risks to the grid as a whole when compared to attacks upon larger systems that could have more significant impacts on the system. Therefore, the proposed solution of this game is a three-stage game, wherein the first stage of the game requires the DERs to work together to find the ideal rate of change of frequency for the sake of finding the most stability in a short term basis through the forming of coalitions. After that stage is complete, there will be an attacker, who will have the opportunity to disrupt the coalition in the most damaging way possible. Finally, the utility who helped coordinate the coalitions will have the opportunity to find which DERs were compromised by the Attacker, and either remove them from the grid, or recover their benefit to put them back in the coalition. This game theory model is tested on a distribution network, wherein This will include analyses of the simulated results of an attack upon a wide variety of DERs when compared to a single significant target in terms of their impact upon voltage stability, frequency stability, and potential loss of load events.

ABSTRACT. Distributed Network Protocol (DNP3) is a popular protocol used in Industrial Control Systems (ICS) and Critical infrastructure (CI) landscape, and is predominantly used in electrical sector. Its protocol stack is a slightly modified version of the Open Systems Interconnection model and allows reliable communication via TCP/IP making it optimal for Supervisory Control and Data Acquisition (SCADA) systems for remote device monitoring of various electrical equipment & devices, real time monitoring, as well as fault detection. DNP3 however is a plain text application protocol making it vulnerable to Man-in-the-Middle, message replays, message manipulation, and Denial of service attacks. With increased attacks on critical infrastructure, DNP3 protocol security is gaining importance as this SCADA protocol was traditionally designed with flexible communication as requirement but not with security requirements. In this work we test protection features offered by Transport Layer Security protocol, that can be used to secure data and communication channel from attackers by using asymmetric cryptographic suites such as RSA for generating secure session keys to symmetrically encrypt the data in transit; to prevent eavesdropping, ensure message integrity and authentication. This research paper will test security in different attack scenarios and demonstrate that TLS can mitigate some of the more common types of cyber attacks implemented on critical infrastructure.

ABSTRACT. In this paper, a supervised artificial intelligence (SAI)-based control strategy is proposed to protect the DC link in a single-phase dual active bridge (SP-DAB) converter from cyber-attacks. The proposed approach involves developing an automated system for training and fine-tuning the SAI model to detect and remove the cyber-attacks based on the errors of the converter's DC link voltage and output power. The proposed SAI-based DC-link voltage estimation along with an adjusted Proportional-Integral (PI) controller are incorporated by the DC-link voltage under cyberattack to appropriately modify the dynamic part of the converter’s phase shift angle. To further inspect the abilities of proposed cyberattack strategy, the different forms of malicious data including constant data, triangular wave, and sinusoidal wave are applied. The proposed method is evaluated using simulations in MATLAB/SIMULINK environment to show that the method is effective in detecting and removing the cyber-attacks on the DC link of the DAB converter.

ABSTRACT. It takes power to generate power; this is particularly important after a blackout. For a restoration to be successful following partial or total system outages, a restoration plan must be organized, put into action, and tested as required by reliability organizations. High impact events, such as natural disasters and cyber-attacks, are known to cause outages lasting anywhere from minutes to several days, affecting the electric grid's infrastructure and delaying restoration efforts. %old sentence: affecting sustainability of the grid. Restoring load after a prolonged interruption---generally referred to as cold load pickup---requires additional power which can exceed equipment's rating and restricts grid operators from simultaneously re-energizing the affected area. This paper implements a load model that considers the effects of cold load for synthetic cases to improve the level of detail in modeling black start procedures. A blackstart scenario where cold load effects are considered is compared with a case (baseline) where cold load is not present.

ABSTRACT. Accurate time aggregation of renewable energy and electricity demand data is essential for effective planning, forecasting, and decision-making of modern power systems. However, two critical assumptions in time aggregation have been overlooked in the existing literature, underpinning the accuracy and validity of aggregated data. This paper aims to address these gaps by shedding light on the cyclical behavior of wind power data and evaluating its year-to-year consistency. To achieve this, two algorithms are designed to explore the cyclic patterns inherent in wind power data and their robustness from one year to the next. These algorithms are then applied to the comprehensive dataset provided by the Electric Reliability Council of Texas (ERCOT). By analyzing the results of our investigation, valuable insights are gained, opening up new possibilities and avenues for future research in this domain.

ABSTRACT. This paper introduces the messaging environment and the geographical visualization tool of the CURENT Large-scale Testbed (LTB) that can be used for large-scale power system closed-loop simulation. First, Distributed Messaging Environment (DiME) implements an asynchronous shared workspace to enable high-concurrent data exchange. Second, Another Grid Visualizer (AGVis) is presented as a geovisualization tool that facilitates the visualization of real-time power system simulation. Third, case studies show the use of DiME and AGVis. The results demonstrate that, with the modular structure, the LTB is capable of not only federal use for real-time, large-scale power system simulation, but also independent use for customized power system research.

ABSTRACT. Cascading failures, although not very common, are remaining concerns in power systems, which can result in enormous electricity service interruption with massive costs for society. Extensive research has been conducted to understand these complex phenomena and mitigate their effects. With the availability of large energy data, data-driven and machine learning (ML)-based techniques have emerged to support the efforts towards cascade resilient power systems. Predicting the risk and scale of cascading failures is one of the key areas that can support other essential functions for controlling and mitigating such events. In this work, predicting cascade size after the initial triggers are modeled and analyzed through data-driven ML-based frameworks. The prediction performance of various ML techniques including Random Forest, Decision Tree, k-Nearest Neighbor, and Artificial Neural Network are evaluated at different stages of the cascade. It has been observed that the power flow data in the early stages of the cascade can be enough to achieve promising performance in predicting the risk of large cascades.

ABSTRACT. This work presents the implementation of deep reinforcement learning (DRL) agents as supplementary primary frequency controllers. To achieve this, the primary frequency regulation problem is formulated in a DRL framework; where an actor-critic algorithm, for continuous actions space, is used to change the frequency reference of traditional governors. By modifying this reference, the DRL agent effectively reduces the magnitude of the frequency nadir and rate of change of frequency, thereby enhancing the power grid frequency response. Two DRL algorithms including Deep Deterministic Policy Gradient (DDPG) and Twin Delayed DDPG (TD3) are employed for the frequency regulation. The supplementary control using these two algorithms is tested on the IEEE14 bus system. Moreover, results show that the frequency stability of the grid can be improved by using DRL algorithms as supplementary controllers in the primary frequency regulation.

ABSTRACT. This work utilized a Reinforcement Learning (RL) agent to track for the Maximum Power Point (MPP) of a 200 kW Photovoltaic (PV) Power System to supply power to a load of 150 kVA with a 0.8 lagging power factor at 120 V_rms where the solar irradiance was changing randomly to a gaussian distribution in the range of 0 to 1000 W/m^2. The implemented RL agent was trained using a Proximal Policy Optimization (PPO) algorithm that used a Fuzzy Logic reward system that was compared with the Perturb & Observe (P&O) algorithm, and another PPO agent trained with a simpler reward system. The implemented RL agent was able to outperform the P&O algorithm and PPO agent utilizing a simple reward system by outputting the highest mean PV power of 132.0 kW and regulated voltage of 117.6 V_rms with little to no oscillations in MATLAB Simulink simulated for 10 seconds.

ABSTRACT. With increasing penetration of inverter based resources (IBR) in distribution systems, the distinction between fault and load current is lost due to the current limiting feature of inverters. This may render the fault undetectable to traditional overcurrent relays. Voltage assisted overcurrent relays and machine learning-based relays have been shown to be promising in this regard. This paper compares the performance of the two relays in a distribution system with 100\% IBR penetration using the voltage and current measurements obtained at inverter terminals.

ABSTRACT. Protective relaying plays a crucial role in ensuring the reliable operation of substations and their equipment. As power systems become more complex, advanced techniques are needed to improve the effectiveness of protective relaying. This paper explores the potential role of machine learning in improv- ing protective relaying mechanisms within substations. Various aspects of machine learning, including deep learning algorithms and data-driven approaches, are examined in the context of their application to protective relaying. The benefits and challenges associated with integrating machine learning techniques are discussed, highlighting their potential to enhance fault detection and prevent false tripping. Through this investigation, valuable insights are gained into the utilization of machine learning as a promising avenue for enhancing the reliability and efficiency of substation protection systems.

ABSTRACT. This paper proposes a method for optimal voltage regulation in a distribution grid by optimally coordinating real and reactive power injections at each node. This enables smart inverters to provide grid support functionality of voltage regulation by optimally managing real and reactive power. This is realized by simultaneously enabling both Volt-Var and Volt-Watt functions and online adjusting real and reactive power settings according to the R/X ratio of the Thevenin impedance seen by a smart inverter. The proposed method considers recommendations of IEEE 1547 standard and requirements of California’s Rule 21 developed under California Public Utilities Commission. Moreover, this enables each DER to have extra impact by providing voltage regulation service while also exchanging real power to the grid. The proposed technique also allows harnessing Battery Energy Storage System (BESS) capacity for voltage regulation service later in the day when the consumer is interested in selling power to the grid during high prices/no PV generation scenario while at the same time utility is interested in achieving maximum voltage regulation. The proposed method also increases observability in the distribution grid as it keeps track of the R/X ratio of the grid. The robustness of this technique under rapidly changing impedance is also shown.

ABSTRACT. Accurate near real time monitoring is needed for Distribution Management Systems (DMS), and power flow-based methods are commonly used in practice for this purpose. However, near real time power flow results are not always accurate because of errors in load estimation, and which may result in poor voltage estimates on distribution feeders. This paper focuses on this issue and proposes a data analytics-based method to parse through data and identify factors/parameters that are unique and relevant to poor near real time power flow performance. The proposed method uses a clustering method (K-means clustering) to determine if we can separate the save cases that perform well from those that do not based on the selected parameters. In the second step, we propose to use a binary logistic regression method to identify unique features for the clusters having distribution feeders with a poor voltage estimate. Test results show that the proposed method can help us to identify unique features for the feeders having poor near real time power flow performance.

ABSTRACT. The paper compares the mitigation strategy applied to eliminate the voltage unbalance in an active distribution network from a utility-scale PVs and Behind the Meter (BTM) PVs, while taking into account the constraints imposed by single-phase center-tapped secondary transformers (common in the North America distribution systems). An AC Optimal Power Flow formulation for the integrated primary and secondary distribution network is introduced to determine the reactive power absorption/injection as a control variable to minimize system-wide voltage unbalance. The analysis treats all nodes as critical, which corresponds to real-world scenarios. As most of the single-phase PV systems in the Low-Voltage systems is owned by a residential customer, the service transformers must be taken into account to evaluate the participation of single phase PV in mitigating the voltage unbalance. The findings of this study contribute to understanding the limitations and effectiveness of voltage unbalance mitigation strategies, considering the impact of secondary transformers. The results can assist distribution system operators and planners in making informed decisions regarding the integration and control of PV systems for voltage management in active distribution networks.

ABSTRACT. The continuous increase in load demand and high penetration of renewable energy sources have driven electric distribution systems to operate close to their voltage stability boundaries. Hence monitoring voltage stability limits in distribution networks is essential to determine any required preventive or corrective remedial action. This paper presents a new method for voltage stability assessment of distribution networks using local measurements. In contrast to existing methods in the literature, which rely on phasor measurement units (PMUs) that are not usually available in distribution systems, to estimate Thevenin Equivalent (TE) for the power system, the proposed method uses the available measurements from smart meters without exclusively relying on PMUs to determine TE. The system equivalent is then used to derive a voltage stability index (VSI). The operating and security constraints at the node of interest are represented in the complex voltage plane which also allows the consideration of additional constraints such as the maximum and minimum voltages. The proposed method is implemented to a 17-Bus test system. The proposed indicator shows a more linear behavior than a previously developed indicator making it easier to estimate the distance to voltage instability and the magnitude of the required remedial action. Margins to maximum loadability can be easily estimated. The required reactive support to keep a specified level of voltage stability can also be determined with high accuracy.

ABSTRACT. In the dynamic analyses of transmission systems (TSs), distribution grids are treated as lumped static net loads. However, with the increasing penetration of distributed energy resources (DERs), the distribution systems (DSs) are transforming into active distribution networks (ADNs) which exhibit significant dynamics. For transmission system dynamic analyses, solving the dynamic model of TS along with the detailed dynamic model of multiple DSs can be computationally involving. Therefore, this paper proposes dynamic equivalent model (DEM) of the ADN based on recurrent neural networks (RNNs). The DEM is created using nonlinear auto-regressive with external input (NARX) type RNN, and employed in the dynamic simulation of the TS, replacing the detailed model of ADNs. The performance of the proposed DEM is compared with that of the detailed model of ADNs, which demonstrates high accuracy with mean errors of 0.36% for active power and 0.04% for reactive power responses. Moreover, the NARX-RNN-based DEM exhibited faster computation speed, achieving a unit solve time acceleration of more than 4 times compared to the detailed model.

ABSTRACT. The operating point of a power system may change due to slow enough variations of the power injections. Rotating machines in the bulk system can absorb smooth changes in the dynamic states of the system. In this context, we present a novel reservoir computing (RC) method for estimating power system quasi-steady states. By exploiting the behavior of an RC-based recurrent neural network, the proposed method can capture the inherent nonlinearities in the power flow equations. Our approach is compared with traditional methods, including least squares, Kalman filtering, and particle filtering. We demonstrate the estimation performance for all the methods under normal operation and sudden load change. Extensive experiments tested on the standard IEEE 14-bus case corroborate the merit of the proposed approach.

ABSTRACT. The distribution grid is a critical infrastructure that provides electricity to end-users, including residential, commercial, and industrial customers. However, the distribution grid is prone to failures, resulting in economic losses and blackouts in daily activities, due to various events such as equipment aging, natural disasters, and human errors. On the other hand, the increasing number of distributed energy resources (DERs) and the aging of the existing infrastructure have posed the challenges to distribution grids reliability. One way to enhance the reliability of the distribution grid is to optimally locate reclosers. Reclosers are automatic, protective switches that isolate faults and restore power electricity within seconds, which minimizes the impact of outages to end-users. Therefore, this paper aims to present a comprehensive study in reclosers placement strategies for enhancing distribution grid reliability. In this study, the System Average Interruption Frequency Index (SAIFI) and System Average Interruption Duration Index (SAIDI) reliability indices have been evaluated in a simulation-based approach to investigate the effect of reclosers placement on distribution grid reliability with a focus on reducing the damages caused by electricity undersupply. The results, carried out in Electrical Transient Analyzer Program (ETAP) software, show that reclosers placement can significantly improve the reliability of a real-world 10 kV distribution grid.

ABSTRACT. Perovskite solar cells have reached efficiencies ~10% more effective than their silicon counterparts. However, the methylammonium lead iodide compound in perovskite raises concerns regarding potential environmental lead leaching. This paper investigates a novel solution by testing a bioplastic layer bonded to natural chelators derived from kelp, cilantro, chlorella, and spirulina against the chemical chelator, tetrasodium EDTA, on a damaged perovskite solar cell's production of lead. The layers’ efficacies were determined by testing if they affected the cell’s electrical output and measuring the quantity of lead they trapped. The spirulina chelator performed the best, reducing lead production by 95.31%. The next most efficient was cilantro by 90.64%, tetrasodium by 84.38%, kelp by 74.6%, and chlorella by 64%. Additionally, the paper investigates anti-reflective properties of corn and lotus leaf surface structures imprinted on the bioplastic layer which resulted in a 16.22% and 23.9% efficiency amelioration respectively.

ABSTRACT. A comprehensive review of the integration and optimization of Vehicle-to-Grid (V2G) into the power grid is provided in this paper. The concept of V2G, its different components and requirements, applications, challenges, and optimization approaches are discussed. V2G has the potential to provide significant benefits to the grid such as frequency regulation or renewable energy support, but several challenges need to be addressed before it can be widely adopted. These challenges include limited charging/discharging infrastructures, battery degradation, and the need for more sophisticated optimization models that consider practical constraints and multiple objectives.

ABSTRACT. Motivated by the rapid expansion of utility-scale Battery Energy Storage Systems (BESS) in the U.S. energy market, this study presents a model aimed at mitigating the high investment costs associated with grid-scale BESS technologies. This model, formulated as Mixed Integer Linear Programming (MILP), optimizes the power rating and energy capacity of BESS for stacked grid services. It does this while taking into account the BESS’s life cycle and longevity considerations, using real- world historical data to manage the many constraints inherent in BESS scheduling. The objective is to lower BESS investment costs while ensuring profitable revenue and maintaining grid reliability. This model takes into account the energy market, the reserve market, and the ’pay for performance’ frequency regulation market within the PJM Interconnection

ABSTRACT. For effective grid management and the integration of renewable energy sources, accurate forecasting of photovoltaic (PV) power output is essential. This paper presents a novel approach to PV power forecasting using deep learning models with Transductive Transfer Learning (TTL) assistance. The proposed models, namely LSTM, Bi-LSTM, and Feed Forward Neural Networks, are trained and tested on a dynamic PV sites dataset that captures the diverse and dynamic nature of PV power generation. The efficiency of day-ahead power forecasting is evaluated on Winter and Summer days to account for seasonal variations. A detailed comparative analysis is conducted, comparing the performance of the proposed models with and without transfer learning assistance, as well as other conventional deep learning models. Additionally, comprehensive statistical analysis is performed to assess the sensitivity and reliability of the proposed models. The results demonstrate that TTL significantly improves the accuracy and reliability of the forecasting models, leading to more effective PV power management. The findings of this research contribute to the field of PV power forecasting and provide valuable insights for practical implementation in renewable energy systems.

ABSTRACT. As variable renewable energy penetration in the power system increases, there is a growing need for flexible resources to balance the resulting variability. Although many systems are transitioning away from fossil fuels, open-cycle gas turbines are likely to play an important balancing role. Accordingly, accurate modeling of the operational parameters of gas turbines will be increasingly crucial as these units are relied on more heavily for flexibility. This paper explores the impact of inclusion of three operational parameters - start-up costs, run-up rates, and forced outage rates – in the production cost model of a system as it adopts higher levels of wind and solar. Using PLEXOS simulations of the NREL-118 bus test system, the study examines how higher the detailed parameter modeling affects outcomes such as the number of start-ups and shut-downs, ramping and total generation costs for open-cycle gas turbines, as renewable energy levels increase. The results suggest the importance of detailed parameter modeling and continued research on combustion turbines’ ability to provide flexibility.

ABSTRACT. This article presents the performance evaluation of ratio consensus-based distributed power apportioning engine along with centralized net load management (NLM) engine that ensures viable and stable operation of an islanded microgrid. Managing net load variability in a microgrid with high penetrations of uncertain renewable generation and ever-changing load demands is a crucial need in order to ensure viable and stable operation of the microgrid. Centralized “dispatch-rule”- based and/or multi-agent-based distributed control of distributed energy resources (DERs) in microgrid are well accepted for microgrid by adapting ANSI/ISA-95-based hierarchical control architecture. In the application where microgrid network has large geographical span with multiple DERs dispersed in the network, high penetration of uncertain renewable energy resources, and ever-changing load demands, a judicious selection of techniques/solutions for managing net-load resources for maintaining viability and stability is required. With this motivation, this article proposes a novel solution to mitigate the challenges by incorporating a mixed centralized NLM engine and distributed power apportioning control of DERs and loads. A power hardware- in-the-loop (PHIL) -based experiment is conducted with the centralized NLM engine and the distributed power apportioning engine along with two commercial inverters. The experimental results validates the efficacy of the proposed method in ensuring viability and stability of a microgrid.

ABSTRACT. High-impact, low-probability weather events have become more prevalent in recent decades due to anthropogenic climate change. As a result, power system resilience has become a vital concept to understand with regards to not only customer satisfaction, but also supplying and upholding critical loads throughout electrical networks. Considering a previous resilience assessment of the IEEE 30-bus test system using a Monte Carlo simulation, the effect of shiftable loads on various resilience metrics was investigated. Through modeling a test system with shiftable load capability, data was gathered and analyzed on the effect load shifting might have on power system resilience during extreme weather events. Conclusions drawn from this data, as well as future directions for research were also discussed (these suggestions include research into the impact of a load shifting methodology on the design and implementation of micro grids).

ABSTRACT. As India embarks on a decarbonisation trajectory following its net-zero commitments at Glasgow in 2021, this paper evaluates the impact of the declining cost of solar and battery storage technologies as well as the cost of capital (CoC) on the future net-zero portfolio of the Indian power sector. The analysis uses an hourly-resolved, technology-explicit, bottom-up TIMES-based cost optimisation and modelling framework incorporating various power generation, energy storage, and green hydrogen (GH2) production technologies with thermal power plant flexibility constraints. Results indicate that in an optimistic cost scenario, a future 100 % non-fossil fuel-based (FFB) power sector complimented by energy storage will be able to meet India’s round-the-clock electricity and GH2 demand. Achieving a smooth net-zero power sector transition requires the CO2 emissions to peak by 2035 and then gradually decline and taper off after 2060. In this case, the crossover between the generation share of FFB and non-FFB resources will occur early next decade. The output from energy storage is likely to meet 20 % of India’s end-use electricity demand by 2070. The transition cost will be highly capital-intensive, which will increase significantly with a higher CoC.

ABSTRACT. The transition to renewable resources has been hindered by the inherent intermittence of power production from those resources. This has necessitated the use of battery storage systems (BSS) to be able to meet the load constantly by charging when there is an excess generation and discharging its power when there is a lack of power to follow the electric demand. While currently essential to operating a fully clean energy system, the BSS is prohibitively expensive. The high cost of the BSS has severely discouraged the transition to completely clean energy systems. Recent advancements in the small modular reactor (SMR) technology have allowed it to increase and decrease its power output to act as a load follower and become a possible replacement for the far more expensive BSS. The goal of this research is to investigate the difference in SMR size and cost when using two and five-percent ramp rates for the NuScale Small Modular Reactor (SMR) using a genetic algorithm for optimization. This was done to understand a relationship between the ramping rate of the SMR and its effects on levelized cost. Four case study locations were chosen, and tests were run with combinations of SMR and BSS where a model was created to find the globally optimized battery and SMR size.

ABSTRACT. Electric grids worldwide are changing and evolving as the grids are modernized and new technologies are introduced and adopted. Solar and wind energy are expected to dramatically increase by 2030, consumption of electricity will likely increase with the addition of electric vehicles and large loads such as bitcoin mining. Accordingly, it is helpful to create open-source grid planning models to reflect these transformations. This paper focuses on updates to a network planning for a Texas 2030 synthetic power system model. Transmission lines of an existing Texas 2016 synthetic power grid are modified to operate under future load and renewable generation scenarios for 2030. The process entails modifying a network model with DC power flow then performing AC Reactive Power Planning (RPP) for AC power flow convergence. In this work, various algorithms are utilized in making alterations to the 2016 grid, resulting in a well-functioning synthetic grid for 2030. Contributions are made towards the transition to clean energy and valuable power flow algorithms are added to the power systems community.

ABSTRACT. The city of Boston has set a goal for all buildings to achieve net zero carbon emissions by 2050. To meet this goal, urban universities could take the following steps, such as increasing renewable energy generation, adding energy storage, converting heating and cooling systems to electric power, and reducing energy consumption. This project aims to analyze the feasibility of incorporating photovoltaic (PV) systems on the campus of Wentworth Institute of Technology, located in an urban area of Boston. Though the university currently emits less greenhouse gas (GHG) than the target for 2025, GHG emissions would need to be reduced by 30% to meet its 2030 goals. PV systems could contribute to this reduction. Rooftop PV systems are suitable for urban universities as the building structures cover much of the available land. The design and analysis were done using industry-standard software packages of Helioscope and National Renewable Energy Laboratory (NREL) System Advisor Model (SAM). The analysis included considerations for shade, weather, module and equipment selection, and campus energy consumption. Two PV designs were made for the campus. The first design resulted in a maximum power production of 1 MW and an energy output of 1,396 MWh/year. The second design yielded a power production of 0.887 MW and energy output of 1,135 MWh/year.

ABSTRACT. The decreasing cost of solar PVs is making them a viable non-wire alternative for rural areas considering infrastructure improvements. However, the intermittency and non-dispatchability of solar generation limits this possibility. Using real data from a small farming town in Kansas, this work analyzes the use of a combined community-scale solar farm and battery energy storage system as a non-wire alternative. In this work, various battery energy storage system operating strategies – such as fixed time discharging, outage mitigation, peak shaving, and market event mitigation – are considered. These operating strategies are evaluated based on financial viability for the utility, impact on battery life, and reliability of power supplied to the community served. The results and analysis provide useful insight regarding economic and reliability impacts with respect to the functionality and complexity of battery energy storage operating strategies.

ABSTRACT. This paper addresses the power flow problem for electric power networks that are based on inverter-interfaced resources. We consider inverter-based resources operating in grid-forming and grid-feeding modes, and formulate a power flow model that accounts for the terminal relations of these resources. Then, we present Newton’s solution method for computing the unknowns in such model, and leverage the structure of the associated iterations to formulate a Quasi-Newton algorithm. We show that such algorithm is closely related to the conventional power flow solution method that is widely adopted in the power systems literature. Numerical results are presented to compare the performance of our proposed Quasi-Newton method to that of Newton’s Method.

ABSTRACT. Power systems research requires realistic test cases to demonstrate and benchmark algorithmic innovations. However, to keep power grid information secure, much of the data for actual systems cannot be publicly released. This motivates the creation of synthetic test cases designed to match the characteristics of actual power grids. By only using publicly available data, these synthetic test cases can be freely released for research and educational purposes. Motivated by the ongoing conflict in Ukraine, this paper presents a synthetic test case for the Ukrainian electric transmission system. With the ability to run power flow and other analyses, this test case is relevant to emerging research and educational activities related to power grid security. To develop the test case, we leveraged publicly available data on population centers, generators, and transmission topology and voltage levels along with synthetic test case creation methodologies from recent literature. After describing the process used to develop this test case, this paper presents validation results using several widely accepted criteria for assessing the test case's realism. As an illustrative application, we demonstrate one use case by solving a bilevel linearization of the $N-k$ interdiction problem to identify critical sets of line failures which cause the most load shedding.

ABSTRACT. Power systems often include a specific set of lines that are crucial for the regular operations of the grid. Identifying the reasons behind the criticality of these lines is an important challenge in power system studies. When a line fails, the line outage distribution factor (LODF) quantifies the changes in power flow on the remaining lines. This paper proposes a network analysis from a local structural perspective to investigate the impact of local structural patterns in the underlying graph of power systems on the LODF of individual lines. In particular, we focus on graphlet analysis to determine the local structural properties of each line. This research analyzes potential connections between specific graphlets and the most critical lines based on their LODF. In this regard, we investigate N-1 and N-2 contingency analysis for various test cases and identifies the lines that have the greatest impact on the LODFs of other lines. We then determine which subgraphs contain the most significant lines. Our findings reveal that the most critical lines often belong to subgraphs with a less meshed but more radial structure. These findings are further validated through various test cases. Particularly, it is observed that networks with a higher percentage of ring or meshed subgraphs on their most important line (based on LODF) experience a lower LODF when that critical line is subject to an outage. Additionally, we investigate how the LODF of the most critical line varies among different test cases and examine the subgraph characteristics of those critical lines.

ABSTRACT. In this paper, we develop an efficient method for performing parameter identifiability analysis for the generator dynamic model by utilizing the parameter sensitivity matrix, which includes all derivatives of the model outputs with respect to the parameters. Then, we present a parameter ranking technique based on the rank-revealing QR decomposition of the right singular vectors associated with the non-zero singular values of the sensitivity matrix. This technique ensures that unidentifiable parameters are consistently positioned at the end of the reordered parameter list. To validate the effectiveness of the proposed approach, we conduct experiments on a hydro generator model. The simulation results demonstrate that the sensitivity matrix based approach can accurately and efficiently assess parameter identifiability, facilitating the identification of candidate parameters for calibration.

ABSTRACT. Synthetic grids are fictitious power system models which statistically resemble actual grids, but do not contain non-public data and hence can be freely shared. This paper focuses on how to tune exciters and governors for realistic synthetic grids. In order to create stable and realistic systems, the parameters of these controllers are sampled from a realistic distribution to satisfy the stability indexes. Of the various types of stability index, three types are used for this paper: the damping ratio, the phase margin, and the gain margin. These indexes can be obtained from frequency response of open-loop system and eigenvalue analysis of closed-loop system. Statistics from actual grids are used to determine machine dynamics and some acceptable parameter ranges for governor models. The introduced synthetic dynamics of generators are applied to publicly available, 37-bus, Hawaiian synthetic power systems and its performance is verified through several disturbances (3-phase balance fault, the change of the exciter set point, and generator outage).

ABSTRACT. In this paper, we propose a comprehensive algorithm for identifying alternative solutions in power flow analysis. Our approach considers various power system conditions and grid sizes, from small to large-scale cases. By analyzing the Jacobian matrix and its singularity, we accurately detect the proximity to voltage collapse and identify alternative solutions. The results demonstrate the effectiveness of the algorithm in handling diverse scenarios and showcasing its capability in identifying alternative solutions in power flow analysis.

ABSTRACT. The occurrence of cyber and physical disturbances in power systems is increasing, resulting in public attention on cyber-physical architectures. It is observed that disturbances can propagate between cyber and physical systems, emphasizing the need to study interdependencies. In this work, we present an approach toward improving cyber-physical interdependency characterization through modeling techniques. The improved assessment of these dependencies can then aid system design optimization to improve functional resilience. To achieve this, we transform the cyber-physical architecture to a graph and apply a bio-inspired network analysis using bipartite network methods to characterize the system during the disturbances. Moreover, a DeepWalk method is applied to cluster the component based on their interdependencies. In this paper, a WSCC-9 bus system is used for numerical study and quantification.

ABSTRACT. The increasing availability of measurement data has resulted in the growing adoption of data-driven state estimation techniques in smart grids. However, the large volume of measurement data has raised concerns about data communication and processing, especially in smart grids with highly stochastic dynamics and stringent reliability requirements. To address these concerns, researchers are exploring new approaches in the field of distributed state estimation. This paper specifically focuses on investigating data-driven state estimation in smart grids using a temporal graph convolutional network framework within a multi-area setup. The performance of the distributed state estimation is evaluated by considering different modes of information sharing and graph structures in the multi-area model. A comparison is also made with the case of isolated localized models, where no information is shared among the regions. The presented analyses aim to provide insights into the role of the multi-area model in improving the performance of state estimation.

ABSTRACT. In power system applications, accessing real data is one of the main challenges. Load modeling is an example that can be hampered by a lack of real data. This paper proposes synthetic data generation as a solution to deal with insufficient datasets. To this end, Timeseries Generative Adversarial Networks (TimeGANs) are proposed to create synthetic electricity data for an agricultural load. Center pivot irrigation system data is targeted, which does not exhibit the same trends as residential or commercial utility data, and information is presented in the lower dimensions. Obtained results show that TimeGANs can leverage the lower dimensional information to produce synthetic data with the same characteristics as the actual data. The proposed method can be utilized for modeling different electric loads.

ABSTRACT. As the integration of renewable energy sources (RES) is increasing in the electric grid, the level of inertia is decreasing which makes frequency stability in the power grid more challenging. To address this issue, this paper proposes a reinforcement learning (RL)-based approach for power system fast frequency response (FFR). The proposed method uses a neural network controller based on soft actor-critic (SAC) to learn an optimal control policy. The SAC RL-based FFR is trained using a reduced order power system frequency dynamics model in Simulink. To analyze the effectiveness of the proposed method, it is compared with another FFR approach, Model Predictive Control (MPC). The results show that the proposed model-free method can efficiently provide FFR in power systems and outperforms the model-based MPC with almost 24% more reduction in frequency deviation and almost 5 times faster computation time.

ABSTRACT. This paper proposes a cyber-resilient consensus control scheme for multi-agent microgrid, which reduces network connections using two-hop communication links. First, it utilizes a method, known as Weighted-Mean-Subsequence-Reduced (W-MSR), where each agent eliminates state values of neighbors that are significantly higher than its own state value. Subsequently, it presents the multi-hop weighted MSR method, where sending agents transmit not only their own state value but also the data of their neighbors to the receiver agents. This approach, referred to as the two-hop communication method, aims to reduce network connections. Our results demonstrate that employing the two-hop technique leads to a reduction in network connections. The effectiveness of the proposed method is verified via simulation results conducted in MATLAB/Simulink

ABSTRACT. This paper presents a comprehensive review of the existing methodologies of data-driven power flow analysis. It begins by discussing the fundamental concepts of power flow analysis and highlighting the challenges faced by conventional methods in solving power-flow. Subsequently, it presents the key principles and techniques underlying data-driven approaches. Then, It overviews and classifies different machine learning models that have been employed in the literature to solve the power flow problem . Further, the challenges faced by these approaches in solving the power flow are explored. Finally, the paper concludes by discussing future research directions and potential advancements in data-driven approaches in power flow solution.

ABSTRACT. Recent advances in power electronic equipments and increased penetration of non-linear loads in modern power systems have presented numerous challenges, requiring immediate and effective solutions to enhance the reliability and quality of power supply. Accurate estimation of harmonics is crucial in mitigating the detrimental effects of non-linear loads. Currently, expensive power quality analyzers are employed to analyze and estimate harmonics present in electrical systems. In this research work, a novel General Algebraic Modeling System (GAMS) based harmonics estimation technique is proposed that can effectively estimate the harmonics present in electrical signals without the need of expensive power quality analyzers. The effectiveness of the proposed technique is demonstrated by estimating the harmonics present in real-time electrical signals obtained from an Axial Flux Permanent Magnet Generator (AFPMG), Uninterruptible Power Supply (UPS) and non-linear current drawn by Light Emitting Diodes (LEDs). Several test cases are also validated considering the harmonics present in power electronics based industrial load and in VFD waveform. Results indicate the effectiveness of the proposed technique compared to that in the literature.

ABSTRACT. In this paper harmonic currents that are introduced in a system due to geomagnetic induced current (GIC) are studied for a 20-bus system. When a transformer is subjected to a DC bias during a geomagnetic disturbance event (GMD) it leads to an effect known as half-cycle saturation. Due to this, reactive power consumption increases and the transformer also starts to draw exciting current rich in harmonic contents.

It is well known that excess harmonic currents can cause issues such as increased losses, misoperation of power system protection, equipment overheating, and damage to capacitor banks in the power network. An important consideration while doing harmonic analysis during GIC is the detailed core modeling of a transformer, since characteristics of harmonic current depend on the nonlinear effects of core saturation. Time-domain simulations are conducted in ATPDraw of ATP-EMTP to analyze harmonics in 20-bus system. This is useful in understanding how the harmonics behave with respect to time which can determine what corrective actions can be taken to reduce the detrimental effects of GIC on a system.

ABSTRACT. The rapidly growing number of distributed energy resources is increasing uncertainty and variability in distribution system operations. At the same time, better access to measurement data is enabling new, data-driven methods for state estimation and analysis. Unfortunately, topology changes pose a significant challenge to existing methods, which are geared towards approximating the power flow (PF) equations in a continuous manner. To address this gap, this paper proposes a data-driven topology-adaptive piecewise PF linearization approach which inherently adapts to topology changes in the distribution system. The approach leverages measurements of a topology-identifying variable that helps cluster the data according to the system topology, without requiring explicit information about the topology status. Specifically, when we fit a three-phase piecewise linear PF model using a K-plane regression algorithm while integrating the topology-identifying variables, we observe that each linear piece only incorporates samples from a single topology. Numerical tests demonstrate that the resulting piecewise linear PF model enables high accuracy PF calculations, even under system topology changes.

ABSTRACT. Deciding setpoints for distributed energy resources (DERs) via local control rules rather than centralized optimization offers significant autonomy. The IEEE Standard 1547 recommends deciding DER setpoints using Volt/VAR rules. Although such rules are specified as non-increasing piecewise-affine functions, their exact shape is left for the utility operators to decide and possibly customize per bus and grid conditions. To address this need, this work optimally designs Volt/VAR rules to minimize ohmic losses on lines while maintaining voltages within allowable limits. This is practically relevant as excessive reactive injections could reduce equipment’s lifetime due to overloading. Optimal rule design (ORD) is technically challenging as Volt/VAR rules entail mixed-integer models, stability implications, and uncertainties in grid conditions. Uncertainty is handled by minimizing average losses under voltage chance constraints, both replaced by smooth sample approximations. To cope with the piecewise-affine shape of the rules, we build upon our previous reformulation of ORD as a deep learning task. A recursive neural network (RNN) surrogates Volt/VAR dynamics and thanks to back-propagation, we can expedite ORD. The RNN weights coincide with rule parameters, and are trained using primal-dual decomposition. Numerical tests corroborate the efficacy of the proposed ORD formulation and solution methodology.

ABSTRACT. Increased penetration of Distributed Energy Resources (DERs) in distribution feeders may have reduced dependency to meet demand with power from substation, but has also lead to a rapid rise in several power quality issues. An effective DER scheduling model is required to combat issues related to system requirements and power quality. This paper proposes a DER scheduling tool using a robust AC optimal power flow (ACOPF) model for an unbalanced distribution system with Watt-VAr curves modeled. The Watt-VAr curves are mathematically modeled using a mixed-integer formulation based on the IEEE 1547-2018 standard. This ensures accurate real-time reactive power dispatch to support local and system-wide fast voltage fluctuation issues. The proposed tool is tested on a snapshot of a distribution feeder in Arizona. The proposed tool is also compared with the Volt-VAr control mode, generally deployed among DERs. The potential of the tool for feeder-wide deployment is illustrated by the results.

ABSTRACT. This work documents the implementation of an electrical network co-simulation between two simulation platforms with different time steps. Stability issues that arise from partitioning a distribution network across two simulators are described. A theoretical stability analysis is presented, and experimental results validate the findings. This analysis provides guidelines for how to divide a network model between simulation platforms.

ABSTRACT. This paper proposes an advanced method using fault tree analysis (FTA) to evaluate the reliability of autonomous electric vehicles (AEVs). Due to an increase in the use of AEVs for transportation, their reliability has drawn further attention. The reliability objective of AEV in this paper is to maintain safe operation by analyzing the effects of important subsystems such as the steering system, automation system, and powertrain system on the AEVs. Reliability is evaluated using a level 3 driving conditional automation vehicle. Relyence software and Monte-Carlo simulation are used to develop the FT of the proposed AEV in order to find some of its reliability indices. Manufacturers and reliability engineers can foresee the overall system's reliability and make any necessary adjustments to improve system design by identifying the possible sources of failure in these systems that could cause the reliability objective to fail. Using the criticality importance (CRIT), risk achievement worth (RAW), risk reduction worth (RRW), and time interval method, the importance of AEV components influencing the overall system reliability is calculated in this study in order to improve the overall reliability at the design stage by finding vulnerabilities in system components. The proposed method is applied to AEV level 3 to show its effectiveness in reliability evaluation.

ABSTRACT. With full-voltage starts on induction motors, locked rotor current and asymmetric inrush current occur. Using reduced-voltage solid-state soft-starters the AC phase current is limited to a value less than the locked rotor value. This extends the start time in favor of lessening the impact on the grid and mechanical stress of the motor. In this paper, the torque per Ampere versus speed are shown for several starting methods. These fixed frequency starting methods are compared to the adjustable frequency drive approach. A presentation of torque per ampere versus angular velocity along with a traditional RMS phase current versus time plot is provided for all methods.

ABSTRACT. This paper compares a single source boosted bipolar PWM half-bridge inverter with a single source boosted bipolar PWM H-bridge inverter. Using the H-bridge inverter as a benchmark, the paper draws a comparison between the two topologies by evaluating the system performance and providing a comparative analysis. Both inverters are fed by a single DC source and employ a boost converter to maintain constant output voltage level over a variable load range. PSIM simulation software is used to examine the system performance of half-bridge inverter and compares it against its H-bridge counterpart for various performance parameters. Simulations demonstrate that half-bridge inverter is more power efficient and yields lower THD due to its smaller number of switches in the inverter’s bridge.

ABSTRACT. Power electronics converters, as an enabler for future power and energy system, are heavily used in military ground vehicle powertrain systems to provide reliable and efficient power conversion. Virtual prototyping of power electronics converters using computer-aided modeling and simulation to create and test power electronics circuits and systems before they are physically constructed can help engineers to optimize designs, reduce development time, and lower costs. This paper demonstrates the effectiveness of developing a High-Performance Computing (HPC)-based power electronics modeling and simulation approach to speed up the entirety of simulation time to support high-speed power electronics-enabled power architecture for demanding ground vehicle powertrain applications. First, a multiple power electronics building block (PEBBs)-based model for the virtual prototyping of power electronics converters is developed in MATLAB. This is then followed by a parallel implementation of the same model in Julia utilizing its highperformance compiled programming language’s fast computing capability and a variety of supported parallel computing programming interfaces. Later, more PEBBs are added to the multi-PEBB model to test the scalability of the parallel simulation. It is observed that the Message Passing Interface-based parallel multi-PEBBs simulation in Julia is both fast and scalable without a noticeable increase in execution times when more PEBBs are added to the system.

ABSTRACT. This paper introduces a novel design for an interior permanent magnet (IPM) motor to reduce cogging torque and torque ripple. The design incorporates a higher number of stator slots and a V-shaped rotor configuration. Cogging torque and torque ripple are significant issues in highperformance IPM motors. The proposed model utilizes two types of rectangular stator slots to accommodate a rectangular stator winding, resulting in a more sinusoidal radial flux distribution at the airgap, improving the overall torque performance. To evaluate the effectiveness of the proposed design, the new stator-slot approach is compared to a conventional motor while maintaining the same rotor configuration for both models. Simulations were conducted using finite-element analysis under identical electrical and magnetic loading conditions. The maximum torque and per ampere (MTPA) control strategy was employed to optimize the output power for both models. Various aspects, including radial flux density at the airgap, static torque, back electromotive force (EMF), cogging torque, d- and q-axis inductances, saliency ratio, iron loss, output torque, and torque ripple, were analyzed using 2-D finite element analysis. The results confirm that the proposed model achieves higher output torque, improved torque density, and reduced torque ripple compared to conventional designs.

ABSTRACT. Multi-port DC-DC converters are gaining more significance in modern power system environments by enabling the connection of multiple renewable energy sources, so the efficient operation of these converters is paramount. Soft switching methods increase efficiency in DC-DC converters and increase the reliability and lifespan of devices by relieving stress on components. This paper proposes a method for soft-switching of a dual-output step-down current-fed full-bridge push-pull DC-DC converter. The converter enables two independent outputs to supply different loads. The topology achieves zero-current switching on the primary side and zero-voltage switching on the secondary side, eliminating the need for active-clamp circuits and passive snubbers to absorb surge voltage. This reduces switching losses and lower voltage and current stresses on power electronic devices. The paper thoroughly investigates the proposed converter's operation principle, control strategy, and characteristics. Equations for the voltage and current of all components are derived, and the conditions for achieving soft switching are calculated. Simulation results in EMTDC/PSCAD software validate the accuracy of the proposed method.

ABSTRACT. In monitoring the power grid, an ability to differentiate between fault types is essential to ensuring electrical safety. Accordingly, this study introduces a fault detection and classification method by considering different machine learning (ML) and feature extraction (FE) methods combinations. Specifically, the proposed method is established in two classification layers; the first layer determines the fault, and the second layer distinguishes the type of fault. Based on the proposed system model, this study seeks to determine the influential data attributes in a power grid signal using FE methods, including fast Fourier transform, power spectral density (PSD), auto-correlation, and wavelet transform (WT). A cross-comparison of the effectiveness of the Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) is also performed to accomplish the classification layers of the proposed method. The designed algorithm is analyzed under the various combinations of FE and ML methods, and outcomes are presented by considering the trade-off between computational complexity and prediction accuracy. The results reveal that the RF-based ML algorithm shows the most accurate classification performance with PSD, and the most time-saving of the models is the DT WT. Also, SVM emerges superior on a subsequent test of the simulated models on real-world signals.

ABSTRACT. Transmission line outage detection plays an important role in maintaining the reliability of electric power systems. Most existing methods rely on optimization models to estimate the outage of transmission lines, and the process is computationally burdensome. In this study, we propose a transmission line outage detection method using machine learning. Using this method, we could monitor the power flow of one line and estimate whether another line is in service or not, despite the load fluctuations in the system. The study also investigates the principles for observation point selection and the effectiveness of this method in detecting the outage of transmission lines with different levels of power flows. The method was implemented on an IEEE 118-bus system, and results show that the method is effective for transmission lines with all levels of power flows, and line outage distribution factors (LODF) are good indicators in observation point selection.

ABSTRACT. Large volumes of distributed energy resources (DERs), such as solar photovoltaic (PV) plants are integrated into the power distribution system due to increased awareness of climate change. These DERs introduce variable and uncertain generation sources due to changing weather conditions. This makes operations and controls challenging and complex. To better understand and manage the dynamic nature of solar PV power plants, digital twins (DTs) will be needed. DTs based on artificial intelligence (AI) methods can be applied to replicate the dynamics of PV plants. This study utilizes a popular paradigm of AI - neural networks to create a variety of data-driven DT (DD-DT) prediction models for a 1 MW solar PV plant located at Clemson University in South Carolina, USA. State-of-the-art internet of things (IoT) based real-time measurements are used to develop the DD-DTs. Typical results for short-term PV power prediction for DTs implemented using multilayer perceptron neural networks (MLPNNs) and Elman recurrent neural networks (ERNNs) are presented in this paper.

ABSTRACT. A feasible and viable industrial microgrid necessitates the efficient integration of diverse Distributed Energy Resources (DERs) including Wind Turbines, Photovoltaics (PV), Energy Storage Systems (ESS), natural gas generators, hydrokinetics, and boilers. To ensure the optimal selection and efficient integration of DERs, effective planning for Renewable Energy Sources (RES) becomes crucial. This paper focuses on improving power engineering education at the undergraduate level by analyzing multiple case studies using the HOMER Pro software to assess economic and sustainability factors associated with microgrid systems. Each case study incorporates unique combinations of DERs for microgrid models that are situated in different locations of the United States, such as in the western region of South Carolina (case study 1), the northern part of California (case study 2), and the northern part of New York (case study 3). The analysis and comparison of these three case studies provide insights into the economic and sustainable aspects of DER integration into industrial microgrids.

ABSTRACT. This article presents the application of numerical integration methods, such as forward Euler (FE), backward Euler (BE) and trapezoidal rule (TR), based on discrete Norton equivalent models (DNEM) through companion-circuit branches RL, for a representation of systems of linear equations for the solution of power systems of any scale in the time-domain (TD).This approach allows obtaining a matrix relationship from the companion-circuit analysis (CCA), which is mainly composed of a symmetric in structure and particularly sparse conductance matrix. The solution of this matrix is exploited using two sparse matrix LU and LDU decomposition processes. The resulting unified methods consist of CCA-LU and CCA-LDU, which are applied to the solution and analysis of the 2383-bus modified power system under fault conditions. The performance of the method is compared in terms of CPU time.