ABSTRACT. Distributed Energy Resources (DER) are becoming widespread in most countries worldwide. In particular, at the household level, rooftop solar photovoltaics and battery energy storage systems — and soon electric vehicles — are poised to play an essential role in power system decarbonisation and more consumer-centric markets. Local energy systems with shared assets, such as community storage, are also emerging as important options. This talk will present fundamental techno-economic aspects of key aggregation approaches enabling the large-scale deployment of DER, namely virtual power plants, energy communities, and microgrids. It will discuss their main features (“what”), the conditions under which each model should be considered (“where”), and the technical and commercial opportunities and challenges for stakeholders such as aggregators, energy managers and distribution system operators (“who”). Several case studies from recent international projects will be used to illustrate the concepts presented.

ABSTRACT. Hydrogen is expected to play an important role in the energy transition, although its contribution extends far beyond electricity generation. The keynote begins with an overview of the hydrogen value chain before focusing on stationary applications and their interaction with electrical networks. Within this context, the presentation highlights the dual role of hydrogen technologies: as dispatchable power generation through fuel cells or hydrogen engines, and as controllable electrical demand through electrolyzers capable of providing flexibility and ancillary services. It then places hydrogen within the broader energy storage landscape, discussing where it complements other storage technologies and the applications for which it is best suited, particularly for long-duration and seasonal energy storage. Rather than presenting hydrogen as a standalone solution, the keynote emphasizes its integration with other flexibility resources to support resilient, renewable-based power systems.

ABSTRACT. This paper assesses the accuracy and solution quality of linearization and convexification techniques applied to key constraints of the nonlinear Hydrothermal Optimal Power Flow (HOPF) problem. These approximations are often required to enable the integration of Optimal Power Flow (OPF) models into advanced optimization frameworks that rely on convex formulations, such as bilevel and robust optimization problems. In this work, power balance constraints are convexified through a Second-Order Cone (SOC) relaxation, while the hydraulic subsystem is linearized, with particular emphasis on the hydropower production function, originally bilinear and reformulated using a four-dimensional (4D) approximation technique. Additionally, the quadratic fuel cost of thermal generating units is approximated through a piecewise linear representation to ensure compatibility with the convexified formulation. The proposed approach is evaluated through numerical experiments on hydrothermal systems with 24 and 118 buses, adapted from the IEEE Reliability Test System (RTS) and enriched with real data from Brazilian hydropower plants. The results demonstrate that the convexified formulation accurately reproduces the behavior of the nonlinear model while maintaining computational tractability for largescale problems.

ABSTRACT. This paper proposes a robust optimization formulation to calculate dynamic operating envelopes (DOEs) to safely operate unbalanced three-phase distribution systems. Unlike conventional formulations that satisfy network constraints only at the envelope bound, the robust formulation covers the entire envelope range. We formulate a robust non-linear programming (NLP) problem with the full AC power flow equations, as well as an approximate linear programming (LP) model. Numerical simulations are run with real-world data from Belgium and two different distribution test feeders. The paper compares the conventional approaches with their robust counterparts and examines the trade-off between constraint violation and envelope size as well as accuracy and solve time aspects.

ABSTRACT. European Transmission System Operators face high congestion costs and intend to make efficient use of grid resources to support the implementation of grid expansion plans. In this regard, in European balancing markets, control room operators can manually activate manual Frequency Restoration Reserves (mFRRs) for real-time congestion control. When operators are the decision-makers, their actions are partly guided by their expertise and experience. However, the benefits of these actions are hard to assess. Therefore, this work intends to better understand the motivations behind these actions and the impact of operator intervention in the decision-making process. This paper proposes a method for the direct activation of mFRR bids to resolve line congestion and formalizes control room operator decision criteria for the activation of these bids. While cost remains central, results indicate that incorporating experience-based operator heuristics in the decision-making process may lead to slightly higher activation costs, for example, to ease operation by reducing the number of activations, anticipate future risks, or limit costs related to ripple effects. Simulations conducted on the IEEE Reliability Test System-1996, using modified Security-Constrained Optimal Power Flow problems, suggest that the current dimensioning of mFRR limits their potential for congestion control.

ABSTRACT. This paper proposes a Mixed-Integer Quadratically Constrained Programming (MIQCP) formulation for the optimal allocation and sizing of Static var Compensators (SVCs) within a Preventive Security-Constrained Optimal Power Flow (P-SCOPF) framework. Unlike conventional approaches, the proposed formulation explicitly captures the local Q/V droop control behavior by incorporating the voltage set-point, droop slope, and rated reactive power capacity as decision variables.The formulation is validated on a 107-bus system representative of the Brazilian interconnected power system and compared against conventional SVC formulations with fixed or neglected droop control. Simulation results demonstrate the effectiveness of the proposed approach in improving voltage profiles under critical contingencies through a more efficient use of reactive capacity. This approach reduces the Voltage Severity Index (VSI) by 27.7\% and achieves an additional 5.1\% improvement over conventional approaches while requiring lower installed reactive power capacity.

ABSTRACT. The increasing penetration of distributed energy resources (DERs), the growing usage of power electronic devices, and the extensive replacement of overhead lines by underground cables have made the reactive power management at the transmission-distribution (T-D) interface more challenging. Alternating current optimal power flow (AC-OPF) provides physically feasible reactive power dispatch decisions for controllable resources. However, long-term horizon studies require solving the AC-OPF problem repeatedly, which becomes computationally demanding. To address this challenge, this paper proposes a graph neural network (GNN)-based approximation of AC-OPF for reactive power management. AC-OPF generates reference reactive power setpoints for a set of controllable photovoltaic (PV) units, which form the supervised-learning labels. Based on the hourly operational data collected over four months, a graph attention network (GAT) is trained to predict the target vector from graph-structured operating states. Performance is assessed using setpoint-approximation error, feasibility after applying the predicted actions and resolving load flow, and computational runtime compared with repeated AC-OPF.

ABSTRACT. Increasing electrification and remote renewable integration force power systems to operate near operational limits, often necessitating redispatch. However, high loading can compromise voltage stability even when static constraints are met. This paper proposes a Voltage Stability Constrained Optimal Power Flow (VSC-OPF) framework by extending the optimization tool AMYGO to incorporate explicit stability indicators. Applied to congestion management, the approach enhances the identification of critical states and ensures larger stability margins. Within the VSC-OPF formulation, system loadability is increased by up to 5.64% while maintaining a secure distance from the bifurcation point. In the multi-objective redispatch application, a cost increase of only 5.34% yields a 40% reduction in the average system MVSI and an 82.1% reduction in the most critical corridor, demonstrating that a modest economic surcharge enables a disproportionately large improvement in voltage stability.

ABSTRACT. Load scheduling and flexibility play a crucial role in enabling modern power systems to integrate renewable energy sources, enhance operational stability, and respond dynamically to varying grid conditions. This paper presents a novel scalable approach for load scheduling and real-time flexibility management, validated through case studies based on the IEEE 33-Bus System. These methodologies are designed to be implemented in Energy Management System (EMS) platforms. Cost reduction is strongly dependent on both solar generation and load, including its types and profiles. Furthermore, the real-time algorithm successfully prioritizes load delivery over Battery Storage System charging (BSS) and BSS charging over energy dumping, confirming its practical viability for smart grid applications.

ABSTRACT. Retail electricity management is inherently affected by two major sources of uncertainty: demand variability and electricity market conditions. To address these challenges, this paper proposes an asynchronous smart electricity strategy for the coordinated operation of residential demand and market participation. The proposed framework is structured as a two-layer, non-cooperative architecture. The upper layer represents an external trading agent, while the lower layer governs the control of flexible residential demand. In the upper layer, an energy trading stochastic programming model determines the optimal day-ahead procurement strategy by minimizing the expected cost under uncertainty. In the lower layer, a Model Predictive Control (MPC) scheme dynamically adjusts flexible Heating, Ventilation, Air Conditioning (HVAC) based residential loads to track the energy procured in the day-ahead market. The results demonstrate that the proposed approach maintains indoor thermal comfort within acceptable bounds while reducing the mean absolute tracking error (MAE) from 10.7 kWh to 7.6–8.5 kWh per 15-minute interval.

ABSTRACT. The electrification of residential heating systems using Heat Pumps (HPs) is a key part of the European energy transition. The associated increase in winter peak demand can intensify voltage deviations and transformer loading in low-voltage grids. To reduce these risks, quantifying the power consumption of future HP usage is vital. While traditional white-box models provide high accuracy, they are computationally expensive for large-scale grid studies. Conversely, black-box models cannot guarantee the necessary representational power for thermodynamic analysis.

Grey-box Resistance-Capacitance (RC) models provide an effective balance between physical laws and data-driven parameter estimation. A critical limitation in many grid studies is the oversimplification of the heating emission system and the building envelope. Many current approaches overlook the connection among Space Heating (SH), Domestic Hot Water (DHW), and Backup Heaters (BH), or fail to separate the fast-response dynamics of radiators from the large inertia of floor-heating systems.

This study addresses these gaps by proposing an integrated simulation framework that combines grey-box envelope models for the Belgian building stock, covering 11 typologies across energy labels A-F, with a detailed, data-driven HP model. This work explicitly distinguishes between radiator and floor heating systems to address their thermodynamic differences. Moreover, to ensure a realistic grid impact assessment, the framework accounts for the stochastic nature of weather and dwelling occupancy, the non-linear Coefficient of Performance (COP) of HPs, and specific control logics such as DHW priority. By bridging the gap between parameter estimation and flexible stochastic simulation, this paper presents a scalable solution that balances accuracy and computational speed. Crucially, this framework enables Distribution System Operators (DSOs) and researchers to perform realistic grid impact assessments that capture factors often oversimplified in conventional models. Furthermore, these thermodynamically grounded models allow for more accurate evaluations of residential heating flexibility, including load-shifting potential and pre-heating strategies, providing realistic power-consumption profiles that overcome the limitations of previous HP models.

ABSTRACT. The increasing deployment of renewables, heat pumps, and electric vehicles in Germany poses significant challenges for low voltage distribution grids. This study investigates whether dynamic electricity tariffs can reduce grid stress and reinforcement costs by shifting heat pump operation to low-price periods. Three real residential low voltage grids were analyzed under low, medium, and high prosumer penetration scenarios. Heat pump time series were generated using two control strategies: a reference strategy with fixed electricity prices and a dynamic tariff strategy employing day-ahead cost optimization. Results show that dynamic tariff control substantially reduces transformer utilization in most medium and high scenarios. Grid reinforcement costs decreased by 58 to 100 % compared to the reference strategy. However, dynamic tariffs alone cannot fully eliminate grid violations under high penetration conditions, indicating that complementary measures remain necessary. The findings demonstrate that dynamic tariff-based heat pump control represents an effective strategy for reducing distribution grid investments while supporting the integration of decentralized energy resources.

ABSTRACT. The increasing electrification of heating and mobility, as well as distributed generation introduce significant uncertainty into distribution grids, complicating demand predictions and grid operation. This paper presents a Monte Carlo-based framework to quantify the impact of correlated weather and behavioural uncertainties on optimized household scheduling and grid loading. Forecast errors of temperature, solar irradiation, inflexible load, and mobility demand are modelled and propagated through a home energy management system and probabilistic power flow analysis. Results show that uncertainty in household power and grid loading is mainly driven by photovoltaic generation and electric vehicle charging, while temperature primarily affects daily energy. Dynamic tariffs reduce overall uncertainty, whereas flat tariffs lead to pronounced peaks. Despite high household-level variability, aggregation limits grid-level uncertainty, and severe congestion events remain predictable.

ABSTRACT. The growing penetration of photovoltaic (PV) generation and demand-side flexibility at the distribution level calls for coordinated local energy transaction mechanisms to ensure operational reliability. While peer-to-peer (P2P) trading frameworks enhance local renewable utilization, fully decentralized designs face challenges related to scalability, coordination, and incentive compatibility. This paper proposes a centralized, tariff-driven P2P flexibility market for a local PV energy community in which a retail electricity provider acts as aggregator. The framework relies on centralized market clearing and dynamic local tariff setting aligned with wholesale market conditions. The local market is formulated as a social welfare optimization problem that integrates PV generation, flexible load scheduling, and grid exchanges. The internal marginal energy value derived from the power balance constraint determines the dynamic local tariff. To ensure computational scalability, the problem is decomposed using the Alternating Direction Method of Multipliers (ADMM), separating prosumer scheduling from portfolio-level balancing while maintaining centralized pricing authority. A case study based on real low-voltage network data demonstrates that the proposed framework increases PV self-consumption, reduces procurement costs, and enables coordinated flexibility without violating operational constraints. The results highlight the scalability and practical viability of centralized P2P market architectures for aggregator-based energy communities.

ABSTRACT. Severe disturbances highlight the limited operational flexibility of distribution networks, motivating effective operational strategies during widespread outages. This work leverages an adaptive network reconfiguration scheme to form standalone microgrids and networked microgrids under multiple simultaneous faults, enabling continued operation using available local resources. Building on these configurations, a convex lexicographic second-order cone programming scheme is developed for the energy management of the formed microgrids, incorporating relaxed battery energy storage and power flow models. The proposed lexicographic optimization scheme strictly prioritizes the minimization of energy not supplied, followed by power exchanges among interconnected microgrids and generation costs. In addition, various objective functions for energy not supplied are investigated to assess fairness in load curtailment across interconnected microgrids. The framework is evaluated on a modified IEEE 33-bus distribution network under fault scenarios, achieving up to 35% reduction in energy not supplied.

ABSTRACT. This paper presents a real-time service restoration methodology for radially operated meshed distribution networks under N-1 contingencies (line outages). The objective is to rapidly recover connectivity by identifying a new radial configuration that satisfies voltage limits, avoids thermal congestion and minimizes switching operations. The approach builds on a sensitivity-based linear model that analytically derives coefficients from network topology to estimate post-reconfiguration branch currents and bus voltages after opening any set of lines, without requiring additional load-flow executions. Service restoration is formulated as a graph-based optimization problem, where edge weights encode operational priorities and an initial feasible topology is obtained via a minimum spanning tree to restore connectivity and ensure radiality. The configuration is then refined using a Multi-Opening Branch-Exchange heuristic to capture interdependencies among switching actions. Validation on a benchmark system and a real large-scale Spanish network demonstrates effective and scalable restoration suitable for real-time operation.

ABSTRACT. The increasing penetration of inverter-based resources (IBRs) is changing fault behavior in modern power systems, thereby reducing the accuracy of conventional short-circuit calculation methods. This paper proposes a nonlinear optimization-based framework for short-circuit current evaluation that explicitly incorporates the current-limiting behavior of IBRs. To enhance computational performance, a differentiable current-limiting model for IBRs is introduced. By replacing conventional piecewise formulations with smooth transition functions, this approach enhances both numerical stability and algorithmic convergence. The proposed method is validated on a modified IEEE 39-bus system and benchmarked against the iterative short-circuit method proposed by IEEE Working Group C-24. Simulation results demonstrate numerically stable and computationally efficient performance, achieving solution times of less than three seconds for the modified IEEE 39-bus system.

ABSTRACT. Despite substantial environmental benefits, the ongoing energy transition challenges power systems operation, particularly increasing the risk of cascading failures and thereby blackouts. To prevent cascading failures, this paper proposes a new Cascade-Aware Security-Constrained Optimal Power Flow (CA-SCOPF) formulation, that augments the conventional corrective SCOPF with additional post-contingency constraints without introducing extra optimization variables. The CA-SCOPF problem is solved by coupling and iterating among a sequential sensitivity-based linear approximation (SLA) optimization algorithm, and a security assessment performed through a quasi-steady-state (QSS) cascade simulator. Assessed on the Nordic 60-bus system, the proposed CA-SCOPF approach mitigates all cascading events and maintains practical computation performance.

ABSTRACT. The demand for high voltage direct current (HVDC) transmission is increasing, as it enables efficient bulk-power transfer and supports decarbonization. In Germany, several grid expansion initiatives target 2 GW transmission capacity using ±525 kV HVDC links, with a longer-term objective of multiterminal HVDC (MT-HVDC) hub interconnections. This evolution from point-to-point links toward MT-HVDC grids raises the importance of component-level verification and insulation coordination to preserve reliability and availability under a broader range of operating and contingency conditions.

During contingencies such as DC faults, MT-HVDC grids exhibit rapid DC voltage decay due to comparatively low system impedance. The associated fault currents increase sharply due to cable discharge contributions and simultaneous feeding from multiple sources. These fault dynamics create severe transient voltages and currents for the cable systems, making fast and selective DC fault isolation a prerequisite for maintaining grid integrity. DC circuit breakers (DCCB) therefore become enabling devices as they interrupt high DC fault currents and shape the associated interruption-driven overvoltage stress seen by connected cable systems.

The paper develops a physically interpretable parameterization of DCCB interruption outputs and uses it to map system-level DC fault interruption to cable transient dielectric stress regimes in ±525 kV MT-HVDC contexts. The analysis explicitly accounts for the interaction between the cable terminal and the DCCB interruption behavior by representing the interruption outcome with four parameters: effective clamping level, commutation di/dt, fault neutralization time and arrester energy absorption. An analytical formulation based on Kirchhoff’s laws explains the dominant commutation and clamping mechanisms, and EMT simulations in a reduced ±525 kV single-pole network with a frequency-dependent cable model quantify the resulting cable voltage response. A screening design of experiments varies the stiffness of the converter-side source, clamping voltage level, fault current neutralization time, fault current limiting reactor (FCLR) value, and the V-I characteristic of the surge arrester. Cable stresses are evaluated using crest- and duration-oriented metrics to distinguish short, high-amplitude excursions from longer elevated plateaus and associated arrester-energy intake. The resulting sensitivity ranking and key interactions define the DCCB interruption behavior as a compact boundary condition to map cable stress regimes, providing traceable input for insulation coordination and representative operational stress exposure in MT-HVDC applications subject to DCCB-assisted interruption

ABSTRACT. The increasing penetration of distributed energy resources challenges conventional distance protection, as multiple infeeds can distort the apparent fault impedance seen by the relay. Conventional setting rules for distance protection rely on electrical distance in combination with predefined safety margins and therefore do not directly reflect the actual relay measurement under fault conditions. This paper proposes a mixed-integer linear programming model for the distance zone reach setting based on apparent fault impedance samples. By using these fault impedance samples as optimization input, the proposed formulation incorporates infeed effects and determines zone reach settings through a selection and assignment process rather than through purely distance-based reach equations. The method is evaluated on a modified Roy Billinton Test System comprising 17 relays and is compared with a conventional cumulative setting approach in fault simulations. The results show that the proposed model improves fault coverage and mitigates underreach effects, particularly in meshed parts of the network with multiple infeeds. At the same time, the study highlights remaining challenges regarding selectivity in scenarios where relay operations alter the network state during fault clearing.

ABSTRACT. Future power systems with growing renewable energy sources and controllable loads require more flexible operating strategies. This work focuses on the utilization of soil-informed ampacity limits for more adaptive power cable utilization. Compared to the prevailing static ampacity limits, incorporating actual soil properties allows for dynamic transmission limits. In this paper, ampacity potential maps are derived to showcase existing regional thermal reserves and to support a geographical-based improvement in the utilization of underground medium voltage cables. The calculations are based on the common NA2XS2Y cable, the local soil type, and operational data such as temperature and moisture content. A case study about Bremen's metropolitan area (Germany) shows soil-dependent reserves ranging from 2 A to 109 A compared to IEC standards. These maps can help system operators to quickly identify free potentials while maintaining safe operating conditions in high-load cases.

ABSTRACT. This paper presents a stochastic bi-level generation capacity expansion planning (GCEP) framework that explicitly accounts for operating reserve provision under increasing wind penetration. The upper level determines long-term investment decisions in conventional and wind generation by minimizing total system cost, while the lower level represents short-term operation through sequential reserve and energy market clearing. Wind uncertainty is modeled via a two-stage scenario-based formulation. A key feature is the endogenous determination of system-wide upward and downward reserve requirements, enabling a cost-aware representation of operational security. The bi-level problem is reformulated as a single-level mixed-integer linear program by replacing the lower-level problems with their KKT optimality conditions and linearizing complementarity constraints. The approach is evaluated on a modified IEEE 118-bus test system, showing that higher wind penetration increases reserve needs and may not reduce total costs when curtailment and reserve provision become significant. Endogenous reserve sizing yields more realistic and economically efficient expansion outcomes than exogenous reserve requirements.

ABSTRACT. The increasing penetration of renewable energy sources and the displacement of conventional generation pose significant challenges for ensuring an uninterrupted power supply in modern power systems. Accurate resource adequacy (RA) assessments are therefore essential for maintaining security of supply and supporting system operation planning. This work proposes an RA assessment framework by developing two mid-term Unit Commitment (UC) models to calculate RA metrics. A detailed UC model is formulated as a mixed-integer linear program (MILP) that minimizes load shedding while considering operational constraints and hourly dispatch detail to enhance solution accuracy. To reduce computational complexity, a linear program (LP) is also developed by relaxing the binary variables of the UC model. These models are applied within a rolling-horizon framework, enabling fast and high-quality solutions for seasonal and mid-term planning horizons. Both mid-term UC models are applied to the Cyprus power system using real data. Simulation results based on 900 annual scenarios show that the MILP-based UC model improves solution accuracy by 11% compared to a base-case approach, while the LP-based model generates solutions approximately four times faster with a small optimality gap of 0.25%. The results also show the probability distribution of RA metrics across the scenarios, enabling the assessment of system reliability and resilience through their mean, Value-at-Risk, and Conditional Value-at-Risk indicators.

ABSTRACT. Achieving accurate long-term capacity expansion planning is intrinsically constrained by the computational complexity induced by the high-frequency temporal variability of renewable energy sources. While time-series clustering is widely utilized to ensure the tractability of Mixed-Integer Linear Programming (MILP) models, traditional aggregation techniques inherently distort chronological sequence and scarcity boundaries, leading to sub-optimal investment trajectories. This paper proposes a novel Two-Phase Clustering (2PC) framework that structurally integrates two distinct hierarchical aggregation paradigms: one optimizing for representative diurnal cycles (Representative Days) and another preserving hourly continuity (Chronological Hours). The 2PC methodology links these approaches through an optimization-driven cascade, using the capacity investment decisions derived from the first phase as strict lower bounds for the chronological second phase. We apply the framework to an eight-node European power system model, demonstrating exceptional computational efficiency and investment fidelity. Empirical results indicate that the hybrid 2PC method achieves cost convergence with the Full Time Resolution (FTR) baseline using as few as 120 representative hours, significantly outperforming the two clustering approaches when applied separately (i.e., one-phase clustering followed by optimization), which require up to 1440 hours to reach comparable accuracy. Consequently, this study establishes a robust, bounding-constrained architecture for integrating multiple temporal aggregation methods in large-scale energy system optimization.

ABSTRACT. The goal of this paper is to determine line-specific loading limits for N-1 secure medium-voltage distribution grid planning. Current approaches to achieve N-1 security rely on either static loading limits or independent per-contingency analyses. However, the former neglects backup-path diversity, while the latter fails to capture interdependencies among alternative restoration paths, leaving a critical gap in comprehensive contingency analysis. This paper proposes a mixed-integer linear programming formulation that embeds all contingencies into a single model and simultaneously determines an optimal backup topology for each contingency, with restoration restricted to switching operations. The result is a line-specific, topologydependent loading limit that triggers reinforcement decisions and reveals N-1 bottlenecks. Additionally, the associated nodal spare capacity quantifies the additional load or generation connectable at each bus without violating N-1 security. The formulation uses LinDistFlow with reactive power, a polygon-based linearization of apparent power, and voltage-dependent thermal capacity. It is applied to the 179-bus MV Oberrhein network under contrasting high-load and high-generation scenarios, and validated against nonlinear AC power flow.

ABSTRACT. The increasing integration of distributed energy resources, in particular electric vehicles and heat pumps, changes the stochastic and time-dependent load behavior of distribution grids. Conventional planning methods rely on static simultaneity factors and therefore cannot adequately capture heterogeneous and temporally varying load compositions. This paper presents an analytical probabilistic method for deriving supply taskspecific simultaneity factors. Probability density functions are obtained from pools of power time series, aggregated by convolution, and used to determine design-relevant power levels at a specified quantile. The design-relevant power is then allocated to the individual grid user groups using conditional expectations, enabling the derivation of group-specific simultaneity factors. The results show that the proposed approach captures time shifts in peak loading conditions and avoids artificial peak-load increases caused by finer group disaggregation.

ABSTRACT. The electric power grid is undergoing the largest transformation in its history, one that involves the decentralization of generation, transportation, and dynamic utilization. Distributed energy resources (DER) play a crucial role in this transformation. However, one type of DER has been under-utilized: the "smart building." While it is broadly accepted that smart buildings can provide various grid services such as demand response and load shaping, they are usually treated separately from other DERs, hindering larger-scale adoption. This paper introduces a masking approach for smart buildings which presents their capabilities as conventional DER. This allows rapid integration into existing DER management tools which boosts their accessibility as dynamic grid resource. The conducted case study using a prototypical medium office building revealed that such smart building can be masked as up to 17.2 kW of distributed generation and up to 137.6 kWh of distributed energy storage, while following the same system integration as conventional DERs.

ABSTRACT. Optimizing energy efficiency and occupant comfort remains a major challenge in building operation, calling for adaptive strategies that are not only effective but also explainable. By making system decisions transparent and linking actions to their comfort and energy impacts, users are encouraged to understand, trust, and actively engage in optimizing their environment. This paper presents a multi-agent recommendation framework that leverages forecast-driven decision-making to improve both energy efficiency and comfort. The system employs an AutoML-based forecasting engine to predict building energy loads, solar generation, and indoor environmental conditions. Coordinating agents integrate these forecasts with day ahead energy prices and weather data to produce actionable, context-aware recommendations. Additionally, the system introduces the Volatility Leverage Index (IVL) metric, an index that quantifies the maximum achievable potential profit across the Day-Ahead horizon, which is used to prioritize agent actions based on market volatility and building’s net load. The framework has been evaluated through realistic building scenarios, demonstrating its ability to reduce energy costs by 13.8% and increase the duration of comfortable indoor conditions by 5.6% relative to the baseline over the one-month evaluation period.

ABSTRACT. Adaptive facades can significantly reduce building energy and peak consumption. Yet, their multi-objective control to balance thermal comfort, visual comfort, and energy efficiency remains an open challenge for large scale adoption. This paper introduces a reward shaping framework within FRADS-GYM, an open-source simulation environment coupling building simulation (EnergyPlus) and daylighting (Radiance) tools to establish and train Deep Reinforcement Learning (DRL) control agents. The agents are evaluated across three climate zones on a simulated prototypical medium office building. The DRL agents achieved up to 40 % energy savings in comparison to static glazing baselines, while maintaining occupant comfort constraints at all times.

ABSTRACT. Nonintrusive Load Monitoring is a research field that aims to disaggregate electricity consumption or label loads using the signals/data acquired from the residence's main panel. However, in practice, it is common to use meters with low sampling rates, which are only capable of delivering average power. In this sense, it is difficult to define a feature engineering stage to treat the average power. Thus, the present paper proposes a discretization procedure of average power to enable the transformation of temporal windows into bidimensional color images, considering five different techniques. From the generated images, a pre-trained MobileNet architecture was used to extract features. Next, shallow learning algorithms were employed to perform the labeling task, reaching up to 97.4% of F1-score.

ABSTRACT. Voltage regulation in distribution networks has become increasingly difficult as PV systems spread across the grid and summer cooling loads increase. This paper addresses the problem through a Decentralized Multi-Objective Optimization Problem (DMOOP) strategy that takes advantage of the flexibility available in split-system air conditioners (HVAC) and PV inverters. The thermal behavior of a split-system air conditioner is captured by a state-space model The comfort of occupants is managed through a Weibull-distributed set-point temperature, which defines the acceptable temperature band and the range over which the HVAC power consumption can be adjusted. PV systems provide the reactive power flexibility that the distribution system operator can rely on to improve grid operation, while compensating customers for this service. The overall optimization problem couples technical objectives, voltage deviation and network losses, with economic ones, and is decomposed into local sub-problems using the Alternating Direction Method of Multipliers (ADMM). The framework is tested on an IEEE MV benchmark network and compared against two solutions: a voltage optimization problem (VOP) using only PV reactive flexibility and without optimization.

ABSTRACT. Micro-grids often use a combination of a battery and a grid-forming inverter to reduce the dependency of a grid segment on a constant power supply. Because these systems offer a high resilience against blackouts, we can expect their number to grow in the coming years. In this paper, we present the Resilience-Monitor, a software that allows to use the rising battery capacity of these microgrids optimally. It can find the optimal operation mode between following economic incentives and retaining battery capacity for emergencies. A Hardware-in-the-Loop environment was used to simulate a generic medium-voltage grid. Data from this grid simulation was used to verify and benchmark the ResilienceMonitor. The software was able to identify grid states with a low resilience across several different scenarios based on a KPI calculation. Furthermore, in critical situations the ResilienceMonitor could recommend preventive measures and dispatch control signals accordingly.

ABSTRACT. Extreme weather events disrupt power and transportation infrastructures simultaneously, affecting the effectiveness of Mobile Energy Resources (MERs) for during and post-disaster support. While MERs are widely studied for power grid resilience, planning often assumes uninterrupted road access, treating transportation as a routing medium rather than an infrastructure subject to hazard-induced disruptions. This paper develops a transportation-aware optimization framework for MER depot siting when substations become electrically islanded from the main grid. Using spatiotemporal hazard impacts on power lines and road segments, a mixed-integer model determines depot locations such that MERs can reach islanded substations while minimizing Expected Energy Not Served (ENS). Demonstration on the Cyprus power and transportation network shows that this approach improves MER accessibility and reduces ENS compared to assuming intact roads.

ABSTRACT. Emissions reduction and resilience to outages motivate the adoption of renewable microgrids. Surprisingly, research integrating both probabilistic grid outages and electric vehicle (EV) charging requirements remains limited. This paper addresses this gap by developing a scenario-based stochastic model predictive controller (SMPC) for a microgrid energy hub comprising solar generation, battery storage, diesel backup, and an EV fleet connected to a weak grid. Grid outage and campus load scenarios are generated from a continuous-time Markov chain and a Gaussian Process, respectively. Using 2023 operational data from the Ashesi University Energy Hub in Ghana, we demonstrate that the SMPC achieves performance within 1\% of a perfect-forecast benchmark. In contrast, a naive MPC that assumes continuous grid availability offers no economic or sustainability advantage over rule-based control, with both incurring significantly higher costs and emissions than the SMPC. These results highlight that outage anticipation is essential for economic viability. Finally, we show that incorporating a deterministic buffer against EV consumption uncertainty eliminates over 90\% of state-of-charge violations with negligible impact on total operating costs.

ABSTRACT. The increasing penetration of converter-based distributed energy resources (DERs) introduces critical frequency stability challenges during post-contingency islanded operation. Conventional restoration planning models neglect rate-of-change-of-frequency (ROCOF) and frequency nadir limits, producing islanded configurations that are electrically feasible but could be dynamically unstable. A two-stage stochastic mixed-integer linear programming model is proposed for restoration planning in low-inertia distribution systems, jointly optimizing DER siting and sizing, grid-forming capability enablement, microgrid formation, and post-contingency load restoration under uncertainty. Frequency security is enforced through analytical surrogates derived from the swing equation, embedded directly into the optimization. Applied to the IEEE 33-bus system, the proposed model eliminates all frequency violations observed in the baseline, where ROCOF values exceed the admissible limit by more than six times in 22% of scenario-fault pairs, at a 5.7% increase in total expected cost. Sensitivity analysis confirms robustness against variations in virtual inertia and droop parameters.

ABSTRACT. Wildfires pose a growing threat to energy infrastructure, yet spatially resolved risk attribution across heterogeneous asset classes remains limited. This paper presents a data-driven wildfire risk assessment framework for power-system infrastructure and demonstrates its application to the Cyprus power grid. The framework uses a two-stage hazard model that separates ignition probability, estimated with a Random Forest classifier, from conditional severity potential, derived from Principal Component analysis of structural wildfire drivers, and combines them into a normalized risk index ranging from zero to one. Results identify photovoltaic parks as the most exposed asset class, with mean risk values of 0.31 and 0.24 during fire and nonfire seasons, respectively, and up to 21.5% of their infrastructure located in the high-risk class. The framework provides actionable, spatially resolved metrics to support risk mitigation strategies, including vegetation management and infrastructure hardening.

ABSTRACT. Winter weather poses a challenge to the reliable operation of electricity distribution networks, yet examining hazards in isolation can underestimate network impacts. This paper develops a negative binomial generalised linear model with interaction terms to decompose individual and compound meteorological contributions to overhead line fault occurrence, validated using historical outage data from the Scottish Highlands. Whilst wind gusts are the dominant driver of faults, compound effects capture risk that single predictor models miss: the wind--precipitation interaction contributes a 6\% increase in expected faults per standard deviation, and the wind--snow interaction a further 8\%. Precipitation was non-significant independently yet significant only through its interaction with wind, illustrating how isolated hazard models can underestimate fault risk. Antecedent 30-day precipitation was non-significant across all specifications. These findings highlight the importance of modelling compound weather effects for distribution network resilience.

ABSTRACT. In this paper, a new strategy is proposed for controlling the on-load tap changer (OLTC) of medium-/low-voltage (MV/LV) transformers. The primary objective is the real-time voltage regulation of unbalanced, low-voltage (LV) distribution grids. Emphasis is placed on the mitigation of undervoltages that may occur during the charging of electric vehicles (EVs). Core of the proposed methodology is the identification of a critical bus that acts as a control reference for the proposed OLTC control scheme to determine the tap position that effectively alleviates undervoltages. The performance of the proposed control scheme is evaluated by performing quasi-static simulations on the IEEE European LV test feeder, assuming a high EV penetration scenario. In addition, a parametric analysis is conducted to identify the critical parameters that affect the ability of the proposed method to regulate the network voltage with reduced number of tap changes.

ABSTRACT. The increasing integration of photovoltaic generation in Active Distribution Networks (ADNs) challenges conventional voltage regulation, necessitating autonomous local control strategies to overcome the scalability limits of centralized dispatch. Yet, standard local Volt-VAR rules often suffer from poor coordination and does not guarantee technical safety nor include stability limits. To bridge this gap, this paper proposes a Data Driven Joint Chance Constrained optimization framework to design adaptive Volt-VAR droop curves. This methodology coordinates inverters to minimize network losses while ensuring voltage and thermal constraints are met with a user-defined probability. Results demonstrate that a tuneable risk threshold achieve near optimal power flow performance without requiring frequent communication, facilitating a better operation of ADNs.

ABSTRACT. The increasing penetration of photovoltaic (PV) systems in low-voltage (LV) networks often causes voltage rise, leading to active power curtailment and thereby limiting local energy utilization within energy communities. This paper proposes a two-stage optimization framework for coordinated reactive power support to mitigate voltage violations and enable increased redistribution of PV generation within an LV feeder. A sensitivity-based analysis is used to quantify the contribution of individual users, revealing a saturation effect in reactive power provision and demonstrating that a subset of electrically influential participants can unlock most of the achievable export increase. Furthermore, the impact of bidirectional reactive power capability is evaluated, showing additional improvements compared to absorption-only control. The economic value of reactive power is assessed through a yearly analysis of community energy costs, and an impact-based pricing mechanism is proposed to allocate benefits according to user contribution. Results show that coordinated reactive power support can significantly reduce curtailment and improve the economic performance of LV energy communities, enabling up to 40% higher PV export and achieving annual community cost savings of 4.85% with a reactive power value of 0.28 €/kVArh.

ABSTRACT. The widespread adoption of distributed generation (DG), driven by its economic and environmental benefits, has significantly increased the need for voltage regulation in distribution networks, with growing impacts even on low-voltage (LV) grids. Voltage regulation devices must ensure high power quality while meeting key practical requirements, including fast response, cost-effectiveness, ease of installation, compactness, lightweight design, and reliable operation for both end users and the network. Maintaining appropriate voltage levels is particularly critical for prosumers, who simultaneously act as energy consumers and generators. To mitigate long-term voltage variations, this study further develops a recently proposed prototype of a single-phase LV voltage regulation device intended for installation at prosumer premises. The device offers a simple, compact and cost-effective solution, using a two-winding transformer with static switches to reconfigure its operation dynamically. This paper provides a detailed analysis of the device architecture and operating principles, with particular emphasis on the tests carried out for its validation. Extensive laboratory testing under diverse conditions, highlights the device's performance improvements and shows the latest advancements in its development.

ABSTRACT. The increasing penetration of distributed photovoltaic (PV) systems in residential low-voltage distribution grids leads to frequent overvoltage violations, posing challenges for conventional control approaches. Data-Enabled Predictive Control (DeePC) offers a model-free alternative to classical model predictive control by leveraging historical input–output measurement data to implicitly represent system dynamics via Hankel matrices. In this work, a regularized DeePC controller incorporating singular value decomposition (SVD) is developed for voltage control using household-level PV curtailment, requiring neither a network model nor explicit system identification. The robustness of the proposed approach to measurement noise, communication delays, communication failures, and limited PV controllability is evaluated on a 14-bus low-voltage rural grid from SimBench.

ABSTRACT. The increasing demand for power generation and decarbonization goals has involved the advancement and rapid growth in photovoltaic (PV) integration. Despite many advantages, excessive integration of large-scale PV into electrical networks can cause a variety of problems. Accounting for the uncertainty and variability of load demand and PV generation is essential for reliable planning and operation of PV-rich distribution systems. This paper presents a probabilistic, time-series hosting capacity (HC) evaluation and enhancement framework that explicitly accounts for uncertainties in PV generation and load demand. Day-ahead uncertainty is modelled using machine-learning–based quantile forecasting, and realistic time-coupled variability is propagated through scenario generation via a calibration-based residual bootstrap, producing multiple joint 24-hour PV–load trajectories. These scenarios are simulated on IEEE 33-bus distribution system to compute dynamic HC considering operational constraints, and a probabilistic acceptance criterion is used to obtain a robust HC profile. To enhance HC, smart-inverter Volt/Var control is integrated at PV buses according to the dynamic grid parameters, providing local reactive power support subject to inverter capability limits. Results demonstrate that Volt/Var control enhances robust dynamic HC compared with the baseline case, highlighting the effectiveness of Volt/Var control for hosting capacity enhancement of distribution system under high uncertainties.

ABSTRACT. Historically, electricity and heat networks have been planned and operated independently since heat demand was mostly supplied by natural gas. However, the proliferation of power-to-heat technologies increases the interactions between these networks, making their operations more intertwined [1]. For example, heat pumps could alleviate or aggravate congestion in the electricity distribution network depending on how they are operated. Therefore, it is essential to plan and operate these networks as a system of systems, rather than as individual subsystems [2]. Prior research has mostly used centralized optimization to jointly optimize investment and operational decisions across both networks [3,4]. However, this approach has several limitations. First, it assumes that both infrastructures are owned and operated by a single entity, leading to solutions that are impractical to implement in real-world; second, it makes simplifying assumptions to model both the electric and thermal domains using a standalone monolithic software, and usually disregards the physics governing network flows. Such assumptions can result in misleading conclusions, such as under-/over-dimensioning network infrastructures, leading to a more costly energy transition.

To address the need for coordinated operation and planning between electricity and heat networks, we propose a soft-linking framework between a techno-economic optimization model of a district heating network and a power flow model of an electricity distribution network. The motivation for soft-linking is that no single tool can capture all aspects of electricity and heat networks in high detail [5]. Leveraging the strengths of domain-specific tools increases model fidelity since each system can be simulated in its native tool with dedicated solvers, and the output of one model can be used as the input of another, rather than assumed to be exogenous. Moreover, soft-linking reduces computational complexity (due to modularity), and makes it possible for network operators to simulate the interactions between their systems without sharing any confidential information.

The proposed approach is applied to a South-Holland case study to investigate the impacts of different low-carbon heating technologies on its local electricity distribution network. First, the soft-linking framework is validated with historical data (2018 to 2022) such as heat prices and load profiles from both network operators. Next, future scenarios (2030 to 2050) for both networks are simulated. These scenarios incorporate multiple potential heat transition pathways elicited from the district heating network operator, and future infrastructure expansion plans provided by the electricity distribution network operator.

Our preliminary results show that nearly half of the heat demand in the district heating network is met by power-to-heat units (electric boilers and heat pumps) and geothermal sources, consequently causing about 660 network limit violations in the electricity distribution network by 2030. Interestingly, a strategic placement of power-to-heat units alleviates electricity network congestion by absorbing surplus generation from local distributed energy resources such as solar panels. These findings highlight the need for coordinated planning and operation between district heating and electricity distribution networks to achieve a cost-efficient and reliable energy transition.

ABSTRACT. The energy transition is bringing about a significant change in the energy system. In order to achieve the climate targets across all sectors, photovoltaic (PV) systems, charging stations for electric vehicles, and heat pumps are being installed, particularly in the low-voltage (LV) distribution grid. High PV penetration in the LV level causes reverse power flows when solar radiation is strong and local load is low. This paper presents a novel relief strategy that combines the advantages of an electrolyzer and a grid coupling. The dimensioning concept is based on the design of a combined heat and power plant with a peak load boiler. The electrolyzer is deliberately undersized in order to achieve a high number of full-load hours, while the grid coupling is used to transfer any load peaks that occur. A prerequisite for the effective use of grid coupling is complementary consumer areas, so that surplus power that cannot be absorbed by the electrolyzer is used in neighboring grids. The results show that the approach presented is an effective way to relieve low-voltage grids with high PV penetration.

ABSTRACT. Decarbonizing the heating sector is central to achieving climate neutrality by 2050. Municipalities play a key role in developing district-level target scenarios for the future supply of heat. However, this depends on decentralized decisions by a variety of stakeholders, whose individual incentives are rarely considered in existing planning frameworks. This paper therefore proposes a multi-level, stakeholder-oriented optimization framework for generating feasible and diverse district-level target scenarios. At the building level, multi-objective optimization identifies Pareto-optimal solutions for each building based on owners, landlords and tenants costs, as well as CO₂ emissions. A heat grid expansion model determines whether buildings are suitable for a district heating connection. At the district level, these results are aggregated and used in a secondary multi-objective optimization to derive target scenarios that balance CO₂ emissions, total costs, and rental costs. A case study on a German urban residential district comprising 1,314 buildings revealed that, while CO₂ emissions can be reduced significantly across all target scenarios, district targets differ in terms of stakeholder burden, technology usage and renovation intensity.

ABSTRACT. Multi-Energy Systems (MES) require coordinated control strategies capable of handling the heterogeneous dynamics of different energy domains. This work presents the implementation and experimental validation of a hierarchical Model Predictive Control (MPC) architecture applied to a real multi-energy facility. The proposed approach uses a centralized high-level MPC to optimize energy exchanges and storage management, while decentralized local MPC controllers regulate subsystem dynamics on separate virtual machines using MQTT-based communication. Experiments conducted on the RSE's DERTF laboratory show that the hierarchical strategy significantly improves overall system performance compared to independent local control. In particular, coordinated operation enables more effective use of CHP generation and thermal storage, achieving up to 24% reduction in operating costs over a six‑hour test window. These results demonstrate the practical feasibility of hierarchical MPC for real-world multi-energy coordination and its potential to enhance efficiency in integrated energy systems.

ABSTRACT. In this work, we tackle the problem of a Multi-Utility (MU) company producing heat and electricity that seeks to schedule the production of its facilities in the day ahead maximizing the profits. The MU owns renewable power plants, such as wind and solar photovoltaic (PV), combined heat-and-power (CHP) plants, and hydrogen facilities, such as an electrolyzer, fuel cells, and a hydrogen-fired turbine. The electrolyzer produces green hydrogen using the renewable power coming from the wind and solar PV plants. Then, the hydrogen may be stored in tanks. In addition, the waste heat generated during the electrolisis process is used for district heating. Moreover, the MU needs to satisfy the heat and electricity demand of its clients while it can participate in the electricity market to maximize the profit. The heat production of the MU is only used to satisfy the demand of the clients, whereas part of the electricity production can be sold on the market. The planning horizon considers two decision stages, namely the day-ahead scheduling and the real-time operation. Uncertain sources are represented based on stochastic optimization.

ABSTRACT. Gas market-clearing models often rely on simplified network representations with fixed pipeline capacity limits, neglecting the nonlinear relationship between gas flow and pressure. However, physical gas flows are governed by the Weymouth equation, introducing nonlinear and nonconvex constraints.

This paper presents a systematic comparison of two gas market-clearing formulations with identical economic structures but different network representations. The first model uses fixed capacity limits, while the second incorporates nonlinear Weymouth equations and pressure bounds. The analysis evaluates differences in prices, flows, and social welfare under varying demand levels.

The results show that both models yield similar aggregate social welfare, indicating that simplified formulations can approximate overall market efficiency. However, significant differences arise in nodal prices and congestion patterns under higher network stress.

These findings suggest that simplified models are suitable for large-scale economic analysis, while detailed physical modeling is necessary to accurately capture spatial price signals and congestion effects.

ABSTRACT. The growing demand for the electrification of heavy-duty vehicles has raised concerns due to current congestion challenges in the existing electrical infrastructure in the Netherlands. This motivates the investigation of solutions that enable the connection of fast-charging infrastructures to the grid without overloading the network. This paper explores the potential of energy hubs to integrate charging stations into the electrical grid. Therefore, a quasi-dynamic modeling approach is proposed to simulate the operation of a grid-connected charging energy hub. The model enables the analysis of the impacts of integrating these systems into the electrical distribution system. The results illustrate that the control algorithm effectively manages the charging scheme and the energy storage system, ensuring that the energy hub remains within its rated limits while imposing a minor impact on the distribution network.

ABSTRACT. European distribution grids are increasingly constrained by congestion and long connection queues. This paper assesses practical options to operate fast-charging stations with a reduced grid connection while keeping service impacts acceptable. Using high-resolution field data, we compare (i) curtailment-based load management, (ii) battery peak shaving, and (iii) a sequential strategy combining both. Grid import is capped to a fixed share of the installed charging capacity. Curtailment follows a fair power-sharing rule across active charging points, and the share of curtailed sessions is summarized with a logistic model to support comparison across sites. Battery peak shaving is evaluated with a simple real-time control scheme. Preliminary results indicate that moderate connection reductions have limited impact on charging service, while strict limits met with batteries alone can require large, lightly used storage. A sequential approach: small curtailment followed by battery peak shaving, can significantly reduce the required battery size, especially for power-limited batteries. This suggests that accepting a small amount of unmet demand can enable substantial storage downsizing.

ABSTRACT. The increasing demand for electric vehicle charging poses planning challenges across the power system, particularly in distribution grids. The stochastic nature of driver behavior and charging locations creates uncertainty in peak demand, complicating grid expansion planning. This paper derives planning-relevant peak-concurrency metrics for residential EV charging in distribution grids, spanning scales from local groups of charging points to system-level aggregation. We analyze one year of pseudonymized 15-minute smart-meter data from a Swiss DSO supply area. EV charging activity is identified in aggregated household loads using a non-intrusive load disaggregation pipeline with noise suppression, event detection, and boundingbox refinement. A statistical analysis of the identified charging activity shows that system-scale worst-case concurrency peaks at about 14% and that smaller groups of charging points can experience significantly higher concurrency than larger groups. These behavior-based metrics support scenario modelling for long-term strategic grid planning, at both local and system scale. Additionally, they can help DSOs assess tariff designs with respect to their potential to avoid or exacerbate synchronized charging peaks.

ABSTRACT. The rapid adoption of electric vehicles (EVs) is accelerating the expansion of public electric vehicle charging infrastructure (EVCI). However, existing planning approaches often assume sufficient grid capacity or neglect spatial constraints and economic heterogeneity across candidate locations. In practice, private EVCI operators face uncertain EV demand, grid connection limits, land-use constraints, and spatial differences in land costs and electricity selling prices. To address these challenges, a two-stage stochastic programming model is proposed for the optimal planning of public EVCI with local energy flexibility based on photovoltaic generation and battery energy storage systems; the installation of EVCI, distributed energy resources, and land utilization under EV adoption scenarios is jointly optimized. Results demonstrate that local energy flexibility enables the expansion of effective service capacity in grid-constrained nodes, reducing congestion while simultaneously improving economic and operational performance under uncertainty. Furthermore, economically efficient locational strategies considering heterogeneous electrical, spatial, and commercial constraints are identified.

ABSTRACT. The increasing penetration of electric vehicles (EVs) poses significant operational challenges for distribution networks due to spatial concentration and temporal variability of charging demand. Existing studies often treat the spatiotemporal evolution of EV adoption separately from network-constrained operation, or rely on single representative operating conditions, limiting the assessment of cumulative network impacts. This paper presents a common spatiotemporal optimization framework to assess the progressive impact of EV integration under different charging strategies, including uncontrolled charging, smart charging, and vehicle-to-grid operation. The proposed approach integrates dynamic zonal EV adoption modeling with a convex network-constrained power flow formulation, enabling scalable multi-period analysis. EV adoption is represented through heterogeneous zonal growth patterns driven by socioeconomic factors, capturing both spatial disparities and temporal evolution of charging demand. The framework is evaluated on the IEEE 70-node radial distribution feeder under increasing EV penetration levels. Results show that uncontrolled charging quickly activates network constraints, while coordinated strategies restore feasibility. Moreover, vehicle-to-grid operation further reduces peak loading and overall network stress, leading to operating conditions closer to the base case.

ABSTRACT. This paper investigates the impact of electric heavy-duty electric vehicle charging stations on the small signal stability of medium voltage grids. It develops analytical state space models for the truck chargers, the charging stations, and the local grid. The dynamic effects on small signal stability are assessed by examining the trajectories of the system modes in the complex plane, using the complete dynamic model obtained by combining the charger, charging station, and grid sub models.