ABSTRACT. This session reviews the use of set-based methods in fault diagnosis (FD) and tolerant control (FTC). Set-based methods use a deterministic unknown-but-bounded description of noise and parametric uncertainty.

These methods aims to check the consistency between observed and predicted behaviour by using simple sets to approximate the exact set of possible behaviors (in parameter or state space). When an inconsistency is detected between the measured and predicted behaviour obtained using a faultless model of the systems, a fault can be indicated. Otherwise, nothing can be stated. The same principle can be used to identify interval models for fault detection and to develop methods for fault tolerance evaluation. Finally, some real applications will be used to illustrate the usefulness and performance of set- methods for FDI and FTC.

ABSTRACT. This work extends the Fault Detection and Isolation approach proposed by M.-A. Massoumnia by including a novel fault estimation stage. This new stage redefines the residue generator by considering a term devoted to estimate the fault magnitude. More precisely, the new term allows to simultaneously isolate the fault and compute its magnitude. The knowledge of the fault magnitude may be used in subsequent stages to determine if the system could continue with its operation or a graceful system detention should be implemented.

ABSTRACT. A common objective on literature for the system operation either on a vertical or a market environment has been to keep an acceptable level of security which means enough capacity of both generation and transmission to satisfy total load without violating system restrictions (which are usually related with the capacity and geographical location of the available equipment) and in fact, enough reserve to be prepared again any system failure. This paper shows the mathematical formulation of a reliability index that measures the level of operational security considering the availability of system’s equipment and its capacity to satisfy load demand for each studied period.

ABSTRACT. Currently fuel cells (FCs) are being taken into account to be incorporated into electrical systems. This is due to its environmental advantages and its inexhaustible service in providing you with fuel. This work presents the incorporation of a fuel cell stack to an electrical network by means of a DC/DC boost converter and a DC/AC inverter. The fuel cell stack is composed of 900 cells of the proton exchange membrane (PEM) type. The power converter uses PI linear controllers to regulate the electrical variables necessary to safely and reliably inject power into the grid. The current and voltage responses of the stack, the converter and the inverter to a change in the power injection in the system are presented.

ABSTRACT. The increase of distributed generation (DG) is now clearly significant in many countries as in Mexico. Photovoltaic systems (PV) are the most popular, due the flexibility and simple installation. As it is known, PV uses electronic power converters to adapt and get the necessary features as power source. Typically, PV converters produce certain implicit harmonic distortion current level due to the power electronics components. Therefore, additional considerations must be taken into account on typical distribution networks analysis. This paper shows the main effects of PV systems in a typical mexican radial network. Cases were adapted to approach the most symbolic and relevant technical and operational aspects. Maximum PV generation and minimum load conditions are considered as main operational states. Operational technical limits are taken also into account. Electrical parameters such as energy losses, power factor (PF), voltage regulation (VR) and voltage total harmonic distortion (THD) are analyzed in this context. Impact level is estimated in function of operational technical limits and electrical parameters.

ABSTRACT. This paper presents the design of a nonlinear speed and current control of a permanent magnet synchronous machine (PMSM). An adaptive backstepping control design for field oriented control is performed. In the proposed approach, it is assumed that all the state variables are measurable, and the controller is designed to track the reference speed despite parameter uncertainties and torque load perturbation. Since torque load is unknown, and adaptive approach for the control design is proposed. The overall control system is asymptotically stable according to the stability analysis based on Lyapunov stability theory. Simulation results on a commercial PMSM are presented to verify the performance and feasibility of the proposed control design.

ABSTRACT. This paper proposes a proportional-integral passivity-based controller (PI-PBC) for supporting voltage in linear loads integrated with \'Cuk converters. An adaptive load estimator is also employed to avoid current measurements at the load point. This estimator permits an on-line estimation of the load conductance for maintaining the output voltage as constant as possible independent on its variations. The proposed PI-PBC allows guaranteeing stability conditions in the sense of Lyapunov for a closed-loop operation by exploiting the port-Hamiltonian structure of the \'Cuk converter model. Numerical simulations evidence the advantages of using PI actions for PBC designs when compared with classical interconnection and damping (IDA-PBC) approach. All numerical simulations are conducted via MATLAB/Simulink software.

ABSTRACT. In this work the design of a distillation column actuator controlled by an interface to regulate the electrical power of the heating resistance in the column boiler is presented. The actuator is based on a AC-DC converter and a DC-DC converter, the actuator is regulated by the output voltage using a PI controller. An interface is developed in LabVIEW to monitor the system and operate the control in real time. The actuator is validated with experimental data from a distillation column boiler.

ABSTRACT. The earthquakes are natural phenomena that occur frequently in the Earth crust. Some of them can cause material losses and they can even provoke human being deaths. Their exact prediction can’t be realized because their nonlinear behavior. However, several studies have been suggested the existence of the electric precursors that might be related to earthquakes of large magnitude. Several reports of electric precursors are related to abnormal behavior of geoelectric signals. By this reason, in this work the implementation of an instrumentation system to measure geolectrical signals have been proposed. It has as a main features that it can be run continuously, it can have remote access to the data. In this paper, some experimental results are presented to evaluate its performance.

ABSTRACT. For the two dimensional canonical system, a family of boundedfinite-time stabilizing positional controls is extendedin terms of a certain parameter. We use Korobov’s controllabilityfunction, which is a Lyapunov-type function. We focus on thecase when the value of the controllability function at the initialposition is exactly the time of motion from the initial point tothe origin. As one of the consequences of such an extension, weascertain that to each bounded positional control there couldcorrespond one, two or three different times of motion from a givenposition to the origin.

ABSTRACT. This paper presents the analysis of the lifetime of a power LED based on its power waveforms. In this work the luminous degradation of the LED is studied based on the LM-80 and TM-21 Norm, comparing it with a curve fitting of the experimental data. The device under test consists of a 1W white LED, which operates for 6500 hours at a temperature of 85 ° C. The compared waveforms are the direct current and a rectified half-sine wave. The continuous wave presents minimal deterioration and its useful life corresponds to more than 50,000hrs. While the LED powered with a rectified sinusoidal half-wave presents a strong deterioration reducing its useful life to less than 7000 hrs, and showing a strong deterioration in its elements (silicone mold), in addition to losing its phosphor layer and emitting bluish light. The power waveform reduces the useful life of the LED since it affects its structure and its PN junction as shown in this article.

ABSTRACT. This article presents a robust nonlinear control scheme applied for voltage regulation in a node of an electrical distribution network, through the use of power electronic converters. This is achieved by means a series of nested control loops based on sliding mode control, which in the event of a variation in the voltage measured in a node of the electrical network, common coupling point (PCC), the proposed control scheme, drives the power converter to inject the necessary power to the grid node to perform voltage compensation. It is worth mentioning, that this control scheme can be applied to control distributed generation systems that can provide auxiliary services to the electrical network in terms of voltage regulation. Simulation results demonstrate the effectiveness of the present control scheme.

ABSTRACT. Large penetration of variable renewable sources andelectronic loads put short-term stress on microgrids. Energystorage systems account for a reliable way to mitigate theseissues. However, depending on the electro-chemistry, each onecan contribute differently to the reduction of daily energy losses.Accordingly, this work presents a benchmark of the introductionof two battery technologies into DC microgrids. The GAMSpackage was used to solve the economic dispatch problem. Resultsshow that lithium-based technologies showed higher overallperformance against lead-acid counterparts. Particularly, ironphosphate technology was not only able to reduce daily energylosses but also to reduce power losses by over 40%. The resultsof this work provide great insights for planning dc microgrids.

ABSTRACT. In this paper, the model and simulation in state space of a wind energy conversion system (WECS), using the type-4 wind turbine topology is presented. The model in state space of a wind turbine (WT) with direct-drive permanent magnet synchronous generator (PMSG) via full scale back-to-back converter integrated into the power grid using a RL filter is described. The particular interest in this work is to address the dynamic study of variable-speed wind, evaluating the performance of proposed control devices. The obtained results in the case study are validated against the obtained response with Matlab/Simulink®.

ABSTRACT. This paper explores the design of the controller for changing batteries for electric vehicle applications using the direct power representation of the system. The design of these controllers is made via passivity-based control (PBC) theory by considering the open-loop port-Hamiltonian representation of the converter. The usage of PBC theory allows designing controllers for closed-loop operation guaranteeing stability operation in the sense of Lyapunov. Two different PBC methods are explored in this contribution, these are \emph{i)} interconnection and damping assignment PBC, and \emph{ii)} proportional-integral design. These methods work over the incremental model of the system for reaching a control law that ensures asymptotic stability. Numerical validations show that both controllers allow controlling active and reactive power independently in four-quadrants. This is important since allows using batteries as dynamic energy compensators if is needed. All the simulations are conducted in MATLAB simulink via SymPowerSystems library.

ABSTRACT. Frequently, various energy sources are used simultaneously in electric vehicles, smart grids, wind farms, among others, and boost power converters are interconnected to match impedance. However, the average output voltage level and the output voltage ripple are factors that directly impact the whole system’s efficiency, useful life, component degradation, power quality, among others. Such analysis is complicated by the number of stages, synchronization and conduction-mode, and related research has been limited to interconnection issues, voltage gain, maximum power point tracking, among others. In this paper, formulations for the average output voltage and the output voltage ripple are presented. These formulas can be used if all boost stages operate either in continuous or discontinuous conduction mode, and their triggering is either equally phase-shifted or simultaneous. The formulations are validated by specialized software.

ABSTRACT. This paper presents a voltage controller based on an Artificial Neural Network (ANN). Its online learning back-propagation algorithm allows the output voltage regulation according to a desired reference output voltage. To verify the effectiveness of the proposed algorithm, a simulation results are presented in three basic DC-DC converters: buck, boost and buck-boost topologies.

ABSTRACT. Experience shows that solar resource prediction is a difficult task. The available solar irradiance where a photovoltaic plant is located or is planned to be installed depends mainly on the cloud incidence at the site. This incidence of clouds depends on the region's climate system, which is well known to be a non-linear, chaotic, and extremely complex, for which there is no exact mathematical model. The chaos level has been determined for various time series of wind and solar irradiance, and it turns out that the chaos level of the solar time series is higher than that of the wind series. A higher level of chaos indicates that the complexity of solar irradiance prediction is considerable. In previous works of solar irradiance prediction, using Artificial Neural Networks, it has been observed that the trained models fail to predict irradiance spikes in conditions of intermittent cloudiness. By conducting a study in this area, we have found that, for a given date, there exists a model to determine the ideal solar irradiance in any geographical location of the planet. These models, so-called clear-sky models, have been taken as a reference to predict not the solar irradiance, but the amount of irradiance occluded by the clouds. That is the difference between ideal irradiance and that measured by the weather station. The proposed model is called SolarDiff, which predicts this difference using Artificial Neural Networks. This article empirically demonstrates that the SolarDiff model exhibits better behavior than models based on direct data. The performance, as in most forecast models, is measured by quantifying the forecast error.

ABSTRACT. Clean-energy generation in smart grids is limited by the availability of the energy to be transformed and advanced energy management strategies occupies solid and anticipated information about its dynamic behavior. This includes multi- variable prediction of meteorological and user consumption data simultaneously in time series. The selection of a predicting model, from long short-term memory (LSTM), convolutional neural networks (CNN), gated recurrent units (GRU), or their hybrid models, is a very complex task. In this paper, mean absolute error (MAE), absolute percentage error (MAPE), and root mean square error (RMSE) comparisons of the CNN, LSTM, GRU, CNN- LSTM, and CNN-GRU hybrids, are presented for prediction of energy consumption, and solar and onshore wind generation. A three-day prediction-horizon is used, with four-year hourly training data from Madrid to show that the CNN-GRU hybrid model shows minor normalized errors, with 1.7102, 1.5631, and 1.5397 for RMSE, MAPE, and MAE respectively. Some training- time and coding appreciations are also presented.

ABSTRACT. This paper proposes the analysis and design of a three-phase back to back converter and a controller whose objective is to operate as a harmonic emulator, in which the main objective is to generate the principal low order harmonics found in the electrical grid. It was considered that the harmonics emulator is part of a wind energy conversion system based on a double fed induction machine. A control law is proposed; the objective is to obtain three-phase voltage with the principal low order harmonics, according to the reference signals. These references set parameters as harmonic order and amplitude. Apart from the back to back analysis, low pass filter analysis is shown to couple the machine and the converter. Finally, simulation results are presented with satisfactory results.

ABSTRACT. This paper shows a comparison in terms of the Total Harmonic Distortion between two techniques to obtain the switching angles of a five-level Cascaded H-bridge Multilevel Inverter, the Selective Harmonic Elimination method, and the Particle Swarm Optimization algorithm. The procedure for both techniques is shown and a theoretical Total Harmonic Distortion value is obtained for each one. Theoretical results of the obtained switching angles for both techniques were verified in a laboratory prototype, where a Total Harmonic Distortion comparison was made with a Power Quality Analyzer.

ABSTRACT. Overhead transmission lines are probably the electrical equipment most vulnerable to lightning, short circuits, human errors, overloads and aging due to the long trajectories they have. In this paper, the operation of the distance relay (21) as primary or main protection as well as backup protection for overhead transmission lines is analyzed. The relay was modeled by using the ATP/EMTP program in order to visualize its response to different fault conditions and impedance trajectories in the R-X complex diagram. The MODELS programming language is used, which allows to implement algorithms that are used in the structuring of the functions and the operation logics of the relay. The relay behavior in the event of faults at different locations on the line is validated by comparing it with the fault records from a real event in the utility of Mexico.

ABSTRACT. Steady-state voltage stability analysis has been approached through different tools based on the polar coordinates power flow model, such as eigenvalues, minimum single value, PV and QV curves and linear sensitivities. In this paper, it is presented a discussion about the application of different sensitivity models, in order to analyze the problem about voltage behavior with respect to reactive power load increases in electric power systems. For this purpose, three different models used to obtain voltage linear sensitivities are formulated using the complete Jacobian matrix, reduced Jacobian matrix, and B’’ matrix from the fast decoupled power flow method. Results obtained from the application of these models to a test system, as well as the advantages and disadvantages of each model are discussed.

ABSTRACT. One of the interests in integrating renewable energy through power converters is to maintain the functionality provided by synchronous generators in the utility grid. This paper presents the nonlinear optimal control of a three-phase inverter to obtain a similar behavior and properties of a synchronous generator, providing energy from renewable resources, inertia and robust operation to the utility grid. The main contribution of this work is the usage of the similarities between the synchronous machine and inverter models, for the design of the inverter-based nonlinear controller to regulate the active and reactive power. This model and control approach can be used, a posteriori, for frequency control in isolated grids and voltage regulation in electrical power systems. The controller is based on the State- Dependent Coefficient Factorization (SDCF) representation of a nonlinear system. The effectiveness of the proposed control scheme and the inertia system is verified via simulations.

ABSTRACT. Between problems in the power grid expansions, distributed generation, energy market customers migrations, commercial losses and other problems, power distribution companies seek to improve both energy quality and costs reductions, enhancing profitability. A basic and crucial point for any energy company is how much energy to buy. That is, determining the amount of energy to be purchased as close as possible to that needed to serve its customers, avoiding financial losses by consuming more or less than they have. However, this is not a trivial problem, as energy consumption depends on several exogenous and endogenous factors, such as all the problems previously mentioned, in addition to economic, social, climatic, political and cultural aspects, among others. Thus, energy forecasts are realized with aid of both statistical analyzes and computational techniques. This article exposes a very short and short term energy forecast model using Neural Networks and feedback, applied in the new global context: the new coronavirus pandemic and its implications for energy consumption. The method was implemented with a real consumption dataset provided by the Brazilian energy company Equatorial from Para State and from Maranhao State. Very short term energy forecasts results reached a MAPE of around 1.2% in a 15-day window for both States, Para and Maranhao. For short term energy forecasts, results for both States were 3 possible scenarios in a window from June to December 2020, due to the unpredictability of the pandemic, especially in Brazil, which so far has shown no signs of reducing the contagion curve.

ABSTRACT. Photovoltaic solar energy has been widely accepted in recent times as a complement to traditional energy sources. DC-DC converters play an important role in transferring photovoltaic energy to a certain load. In applications where a high DC bus is required, it is necessary to use high voltage gain converters. This article presents an analysis based on switching functions of three high gain DC-DC converters including parasitic elements, these converters operate in continuous conduction mode, it is presented the switched model, the averaged model, and the static model. The results obtained during the analysis will be used to obtain the voltage stress on the switches, design high gain DC-DC converters to operate the photovoltaic module at its maximum power point, and raise the voltage from the module to 380 V by simulation on PSIM. The flyback converter presented an efficiency concerning output power and input power of 67.6%, cascaded boost converter of 71.1%, and Proposed-I converter of 84.8%.

ABSTRACT. In this paper a method to perform maximum power tracking on isolated PV systems without energy storage is proposed. Initially, numerical simulations are performed for a PV system composed of an array of panels, a DC/DC converter and a single-phase power inverter feeding a resistive load. Results under load variations show the need of performing power tracking to achieve the demand supply. Moreover, in order to keep a proper operation of the power inverter a DC-link voltage control is also performed by an improved power tracking method combining a P&O algorithm and a PI regulation loop. The proposed algorithm shows an acceptable performance to keep power delivery under environmental variations. This method seems to be effective to PV systems in stand-alone applications. Ongoing work is currently devoted to perform experimental validation of the numerical results presented.

ABSTRACT. Common mode voltage is an important issue in motor drive systems since its dynamics could produce leakage ground current reducing the machine lifetime. In order to contribute to reduce the effects of the leakage ground current, threephase power inverters can be designed using dedicated pulse width modulation strategies. Space vector modulation technique is a commonly used continuous method due to its flexibility to implement different vector sequences. In this paper it is proposed a space vector based modulation technique to reduce common mode voltage in a three-phase ANPC inverter. An analysis and comparison regarding conventional space vector modulation techniques is performed. Numerical results are obtained to prove the feasibility of the proposed method.

ABSTRACT. This paper presents a system for tracking the maximum power point (MPP) and maintain regulation of output voltage supply by a photovoltaic panel. Were interconnected two converters SEPIC. The P\&O MPPT algorithm (maximum power point tracking) was applied in the first stage is to obtain the maximum power of PV independently of irradiance and temperature that depend on environmental conditions. The passivity based-control to obtains regulation of output voltage is applied in the second stage. The objective of this study is to operate the photovoltaic panel at the point of maximum power in variable environmental conditions to increase efficiency and also provide appropriate current and voltage to load applied. The design of the system was developed using the mathematical model of the panel and the controller. Later the whole system was simulated in PSIM. Simulations results confirm the effectiveness of the proposed system.

ABSTRACT. An orthogonal vector analysis of windowed dynamical electrical signals with a Kalman filter tracking is proposed in this paper. Based on The Fourier Series Theory, the correlation of the signal coefficient that minimizes the error energy is obtained to extract the sinusoid component of the dynamical electrical signal. Further, The Kalman Filter is applied to track the hidden time-frequency behavior. The proposed algorithm is validated here comparing the response of the Kalman Filter of the non-treated signal and the approximated one with the discrete Fourier transform. The results suggest that the proposed method is able to improve The Kalman filter response when treating with signals with non-linear dynamical properties.

ABSTRACT. Every day the devices connected to IoT systems networks are increasing, so it is necessary to have a server to handle these connections. Therefore, in this work proposed implementing a multi-platform connectivity node for IoT, which uses LoRa net-works to guarantee connectivity in WAN (Wide Area Networks). It also introduces a general methodology for the construction of IoT systems that allows the validation of the hardware, thus, which guarantees interoperability between different platforms. By being able to implement this multi- platform node in WAN, it is possible to use it in several areas such as intelligent cities, public services, agriculture, health services, and government.

ABSTRACT. The monitoring of air pollutants levels in urban environments is of vital importance because their potential negative effects on human health. In particular, high ambient concentrations of particulate matter (PM) have been associated with high occurrence of respiratory and cardiovascular diseases. Current air quality monitoring systems provide high quality data but with limited spatial coverage. Low-cost sensors represent a potential tool to increase spatial and temporal measurements coverage by complementing the existing networks. Furthermore, the remote access to real-time data may allow to obtain detailed information relevant for instance, for protecting vulnerable populations health. This study describes a portable PM monitoring system with remote access via IoT. The proposed system is based on low-cost sensors of the Plantower type and has Internet access via WiFi through the ESP32 microcontroller. We describe a methodology to fit the PM sensor with wireless access to store real-time information in databases located on both local and external servers using an open-source code. Statistical tests conducted for preliminary monitoring data are also described to verify the PM sensors accuracy and performance.

ABSTRACT. Analyzing data on smart meters is a trend increasingly used by utility companies as it allows a better understanding of data directly from the source of origin. New distributed computing architectures like edge computing have given advance to improve data analytics. Generally, the capacity of such devices, including smart meters, is quite limited, so the use of specialized auxiliary hardware has begun to be used in these devices. The present work shows the results of using a neural stick compute for forecasting and data classification processes within smart metering systems. The results show that the processing times can be remarkably improved with the use of stick computers having a suitable model for artificial neural networks.

ABSTRACT. This article is about a remote monitoring system of internal and external temperature, internal and external humidity, luminosity and wind speed in a bee hive to generate a basic information system for beekeepers. Next, the development of an intelligent hive for the monitoring of health in Apis Mellifera bees is disclosed through remote monitoring of internal and external temperature, internal and external humidity, luminosity and wind speed in the hive. The project consists of three stages: electronic instrumentation from digital and analog sensors that provide the information to the system, radio frequency communication in order to send the hive information to an IoT application for data analysis. This project is directed to the beekeeping sector of the municipality of Miraflores Boyacá (Colombia).

ABSTRACT. Maturity of LoRAWAN based solutions has led to the integration of innovative services in IoT systems, being geolocation a remarkable feature that enables functionalities as asset tracking. The before mentioned capability, combined with Real-time responsiveness and data analysis could derive to the possibility of smarter systems that significantly add value to applied technology. The current work contains the implementation of a technique for geolocation based on LoRaWAN which has been integrated into an IoT end to end (e2e) solution that monitors the probability of an existing fire. By adding this characteristic, the system can locate a fire incident at the right moment and prevent material, economic and environmental affectations.

ABSTRACT. The Internet of Things (IoT) is a technological revolutionary paradigm which has enabled to communicate a wide number of devices through of Internet. In this paper we present the design of a terrestrial drone which uses an architecture of Internet of Things to be controlled remotely. This drone include a camera which is used to obtain landscapes images from the terrestrial drone. A web server allow to store the commands to control the movements of the terrestrial drone in a MySQL database as well as the transmission of commands between the server and terrestrial drone. using web pages with PHP and Android APP. Our electronic system is based on WiFi Module ESP8266 which offers advantages to incorporated open software architectures. The landscapes images are processed in a computer using python and machine learning libraries. The machine learning algorithms recognise in the landscape two type of classes: agave and stones.

ABSTRACT. A time series is a sequence of observations, measured at certain moments in time, ordered chronologically and evenly spaced, so that the data are usually dependent on each other. Currently, time series are used to estimate wind gusts, which are highly non-linear, unknown, and at times unpredictable. A good estimation of wind gusts implies correct planning on the generation of clean wind energy. In this work, we use Artificial Intelligence (AI) techniques such as the use of convolutional neural networks for wind gust estimation. One of the best models for dealing with this type of information is the Large Short Term Memory (LSTM) network because it is a type of recurrent network that specializes in sequence information. In this work, an LSTM prediction model is implemented for five different wind speed data sets using different multi-step forecasting strategies. The strategies used are Recursive, Direct, MIMO (multiple-input to multiple-output), DIRMO (Combination of direct strategy and MIMO), and DirREC (Combination of direct and recursive strategy).

ABSTRACT. This paper presents a procedure suitable to design an ANFIS automatic voltage regulator of a synchronous generator. Several aspects related to the design of this intelligent AVR are explored and compared, in order to determine the best procedure to perform the design. The performance of the AVR is tested and compared to the one of conventional AVRS at the 9-bus 3-machine IEEE test power system, showing an important improving of the system dynamic response.

ABSTRACT. Currently, deep learning methods are being used and proposed to deal with the problem of wind speed time series forecasting. This is since they have good forecast accuracy; however, they also have greater complexity and there is an increase in the computational effort used in comparison with the conventional forecasting methods. This paper reviews the deep learning methods most widely used in time series forecasting, such as convolutional neural networks, long short-term memory networks, and hybrid methods. The results are compared against the autoregressive integrated moving average (ARIMA) method, which is typically used, due to its simplicity and high precision. A benchmark was generated based on a wind speed time series obtained from a meteorological station, obtaining hourly forecasts one step ahead and subsequently obtaining forecasts of several steps ahead. The results show the improvement in the accuracy in the forecast obtained when using the methods based on deep learning, as compared with the ARIMA method.

ABSTRACT. Monitoring of power quality disturbances (PQD) in power systems is crucial in determining their causes and avoid equipment damages. In this work, a Matlab algorithm was implemented to detect and extract the distinctive features of seven simple power quality disturbances (sag, swell, interruption, flicker, harmonics, oscillatory transient, and notch). Firstly, a database was generated with the seven types of disturbances designed from their mathematical models. The decomposition of the signals was subsequently performed using the Discrete Wavelet Transform (DWT) through the Multi-Resolution Analysis (MRA) with six levels of details. The sampling frequency was varied to identify the useful features that, with energy distribution and RMS extraction methods, serve as input to classifiers to distinguish disturbances. Three classifiers were considered to demonstrate the effectiveness of the algorithm to identify the useful features, Probabilistic Neural Network (PNN), k-nearest neighbors (k-nn) and Multilayer Feed Forward Neural Network (MLFF). With the proposed method, it was possible to detect the seven simple disturbances analyzed, maintaining a balance between simplicity, robustness, and efficiency, which will have an impact on guaranteeing a lower processing cost and can be used in real-time applications.

ABSTRACT. Obtaining clean energy through wind farms is considered viable because of its low operating cost. However, the behavior of the wind speed is not constant, it has a chaotic behavior and it is highly data dependent. The aim of this work is to carry out a comparison of several shortterm wind forecasts using Artificial Intelligence models such as Artificial Neural Networks (ANN), Multilayer Perceptron (MLP), Convolutional Neural Networks (CNN), and Large Short-Term Memory Networks (LSTM). We present several scenarios where the models are contrasted in order to analyze the advantages and disadvantages of using these strategies to decide the viability to build a wind farm in the analyzed place.

ABSTRACT. Energy storage is considered by many as the Holy Grail of future decarbonized systems, characterized by nearly 100% share of modern renewables, much more distributed and less controllable. This keynote speech first discusses the main reasons why massive amounts of energy storage capacity will be needed beyond this decade. Then, a brief overview is provided of those technologies which are most promising to fill the existing gap in stationary applications, with special attention to pumped storage and batteries. Finally, the main applications in power systems, from generators to prosumers are illustrated, and some prospective data regarding future deployments of energy storage assets are provided.