ELECTRIMACS 2022

Electric autonomy, environment respect, compactness, lifespan and low maintenance costs, are real technological challenges to be taken up on the future ship propulsion system. In this paper, an experimental study of a Hardware In the Loop (HIL) based Emulation of a Naval Propulsion System associated with Supercapacitor Energy Storage System is proposed. For these purpose two electrical machines are associated in the same shaft (DC machine and PMS machine) to reproduce the behavior of the association of the electrical motor and the propeller of a ship at laboratory scale. The whole system is constituted by three subsystems. The first one is represented by a PM synchronous machine associated with a controled VSI which plays the role of the electrical propulsion motor following a realistic mission profile. The second one is constituted by a DC machine associated with a DC/DC converter which is controled in order to reproduce the hydrodynamic behavior of the propeller and the ship, and the third one is constituted by a buck-boost DC/DC converter and an ultracapacitor bank, the ultracapacitor gives to the emulated mechanical shaft the ability to reinject kinetic power into the grid in case of transients. This experimental tool can be associated to innovative hybrid energy system to test and validate hybrid system configuration and can be exploited for many kinds of propellers. In addition, it gives the designer a feedback to optimise ship design. The effectiveness of the proposed HIL platform in terms of DC grid stability and 4 hydrodynamic operation of propeller is verified by experimental results.

This paper presents a forecasting model from 1 to 10 days for the aggregated photovoltaic energy production in Spain. The model uses a convolutional neural network which inputs are meteorological forecasts, historical generation data and the location and installed power of existing plants. The model output is the hourly production of the photovoltaic energy production for the whole system for the following ten days. The results of the model can be used for generation scheduling and system operation on one side and for energy trading in the day-ahead market or in derivative markets on the other side.

As the needs for sustainable freight transport solutions grow, river freight cargo is considered to be a very promising track. Large carriage boats allow up to three times less fuel consumption than trucks per ton.km. However, the use of traditional river freight ships leads to a high level of NOx and PM (particulate matter) emissions. Late evolution in regulation, large lifetime, 60 years for hull and 20 years for engine, as well as economical tightness have slowed down technological improvements. European solutions, such as the use of alternative fuels, electric or hybrid propulsions are studied. Projects with zero emissions are favoured as well as projects with easy potential adaptation into zero emission boat: full-electric vessels and hydrogen powered hybrid ships are the main solutions to be considered. Some new generations of vessels are already launched or planned. This paper aims to expose the challenge and to present the current status and trends for improving the impact of river freight.

The nonlinearity of boost DC/DC converters is well-known in the field of power electronics control. Traditionally, the converter model can be linearized at an operating point which is the small-signal linearization. On the other hand, the converter can be regulated by using nonlinear controllers in which sliding mode control (SMC) method is commonly applied. The previous works of SMC were often based on the switching model of converter which suffer from non-uniform switching frequency and/or chattering. This paper proposes two novel average model-based SMC schemes for bidirectional boost DC/DC converters; one is a single-stage controller and the other one is a cascade control loop. The new controllers outperform the traditional ones in terms of response time and accuracy that is validated by critical simulation scenarios. The proposed approach can be extended to other complex power electronics converters.

This paper introduces a dynamic Pulse Width Modulation scheme, apply to N-level Flying Capacitor inverter. A mathematical approach is detailed to solve the expression of a linear system that model the inverter leg considered. Some degrees of freedom, that have to be set, are exhibited. The proposed technique relies on the relationship made between the degrees of freedom and the voltage unbalance of the floating capacitors. Simulation results are provided to illustrate the performance of such a modulation scheme.

Recently, electric power production using the tidal turbine system has increased in many countries. Changes in the speed of the marine current pose one of the main challenges in the planning and operation of the electrical network. This paper focuses on the analysis and simulation of Orbital O2 Tidal energy conversion (TEC) system based on Energetic Macroscopic Representation (EMR). The profile of 10 years marine current from the Pentland Firth – Orkney has been used in the EMR model system and the proposed control strategy. The simulation results show the ability of EMR model to adopt the proposed control strategy to maintain the voltage at grid side and to follow the maximum power point tracking (MPPT) at the rotor side of the O2 tidal system.

Modern electric power systems are becoming more complex due to the proliferation of distributed energy resources (DERs) and power electronic converters. The increased penetration of power electronic-based DERs gives rise to new instability issues, particularly subsynchronous oscillations (SSO) caused by control system interactions, the induction generator effect, or resonance with torsional modes of wind turbines. If poorly damped, such oscillations can compromise the stability of the power grid, resulting in damage or disconnection of equipment and grid sections. Therefore, detecting such frequency components during both planning and operation phases is of paramount importance. This paper proposes a straightforward hybrid method combining Fourier and wavelet analysis (WA) for the online detection of the subsynchronous frequency components in power system measurements. The method leverages the speed of the fast Fourier transform (FFT) and the accuracy of the WA to determine the frequency, amplitude, and damping of SSO events.

This paper presents the development of a sizing approach under reliability and energy performance indexes applied to fuel cell/battery (FC/BAT) hybrid power system. It deals with the interdependence between the components sizing process and the control strategy. The proposed approach is based on imbricated optimization loops and considers two important criteria, fuel consumption and reliability. This proposed approach involves multi-objectives optimisation to study the trade-off between reliability and energy saving to improve the optimal design relevance. Also, the implementation was performed using a low complexity process to offer compromise between computation time and results accuracy. This constitutes a very effective support tools to help design engineers in the early design stages.

Adding grid forming capability to the Wind Power Plant (WPP) allows it to carry out some functionalities that are usually performed by conventional power generation. Black start operation is one of these functions. It is necessary to assure good performance and observance of all operational and stability requirements to allow WPP to take over this role. This paper presents a compliance evaluation procedure based on a set of defined test cases. Massive simulations have been performed using a computational cluster. Among defined test cases, substations and long cable energisation entail the most challenging steps in the black start operation procedure. Results of the system performance of these cases are presented. All validation has been made through detailed PSCAD/EMTDC simulations.

Starting a fuel cell system under subfreezing temperature remains a technological challenge for a large-scale distribution of this kind of power generator, especially in embedded applications. In this paper, investigations on the cold start capabilities of a 700 W proton exchange membrane fuel cell (PEMFC) are performed. The study is based on a potentiostatic control of the load during the start-up procedure. Thanks to the proposed methodology, the fuel cell was able to start without assistance from a temperature of −10 °C in 105 s.

This paper focuses on classification methods for evaluating the lifetime consumption (LC) of power electronics modules. The generalization of power electronics devices introduces new issues concerning the reliability of equipment, especially in the transportation field. To meet these expectations, this paper discusses an approach to evaluate the percentage of lifetime of a lab-scale SiC MOSFET power module, designed for an aircraft application. This module is based on a planar technology, and presents typical failure modes concerning the SiC MOSFET chip itself and its environment. The modules have been aged on a specific instrumented test bench to trigger the expected failure modes. Thanks to it, a large database of parameters have been obtained in order to find a relevant failure signature. Once the signature obtained, a comprehensive solution is required to classify the signatures into relevant classes related to the module LC. To meet the issue, three types of classification have been tested with learning data set: Support Vector Machine, k-Nearest Neighbors and neural network. The last contribution of this paper is a discussion on the evaluation of the percentage of lifetime consumption of a new test module thanks to the most promising models obtained from the learning data set.

Wide bandgap power switching device technologies earned immense superiority in power density converters in terms of higher switching frequency and efficiency attainments. However, this becomes opposing when utilized in planar pulse transformer-based gate driver applications, where rapid switching speeds originate electromagnetic disturbances due to the passage of common mode currents through the transformer’s stray capacitances. This paper will examine the common mode transient immunity (CMTI) of a shielded pulse planar transformer whilst examining the impact of dielectric material selection on the outcome result. The simulation methodology using Altium Designer and Ansys Q3D Extractor with dynamic links will be presented, in addition to agreeable experimental verifications.

This paper presents a modeling and simulating methodology of transient phenomena in power converters for a large frequency bandwidth. The system under study considers parasitic elements of semiconductor switches, common mode parasitic elements of power transmission lines and their interactions in a high integrated buck converter with one switching cell. In particular, the methodology helps the modeler to chose between several types of models for a given element, i.e. T −model or Π −model of the power transmission lines. The proposed methodology finally allows to obtain a representation well-adapted to the simulation of the behavior of the power converter including parasitic oscillations along time.

This paper discusses enhanced static and dynamic modeling of a series-series compensated Inductive Power Transfer System (IPTS) using a Buck converter as a post-regulator. A First Harmonic Approximation (FHA) method is adopted to develop a static model of the analyzed Post-Regulated IPTS (PR-IPTS), highlighting operating regions in which a Buck post-regulator may exhibit controllability issues. Dynamic modeling of the proposed PR-IPTS is then performed by exploiting previous works based on a Coupled-Mode Theory (CMT). Their findings are herein extended to include a phase-shift modulation of the primary full-bridge inverter operating at whatever switching frequency. Experimental tests performed on a laboratory prototype confirm good output voltage regulation capabilities in both static and dynamic load conditions.

This paper discusses the design and optimization of a series-series compensated Inductive Power Transfer System (IPTS) followed by a post-regulator consisting in a DC-DC Buck converter. A static model of the system is developed through a first harmonic approximation method, and a procedure for compensation capacitor selection is proposed. A sensitivity analysis of the IPTS performances is carried out with respect to the variations of the primary inverter switching frequency and phase-shift angle, to achieve the efficiency maximization while ensuring the system controllability. Experimental prototype is developed able to deliver up to 29 W output power at a 12 V output voltage, yielding a 91.7% maximum efficiency. The IPTS behavior under both stable and unstable conditions is finally tested, thus confirming the static model predictions.

This paper deals with the PWM-induced current and losses in a specific segmented winding structure. The proposed segmentation process enables to split a winding into several coils. These latter are supplied independently by H-bridge converters and are wound around the same magnetic circuit. This process leads to a deeper segmentation of electric drives for enhanced modularity and reduced voltage rating. The strong magnetic coupling between each coil is described, and the control degrees of freedom are presented. This study provides a model based on an analytical method and on an equivalent electrical circuit calibrated through experimental results. A trade-off is found between the losses related either to the distribution of the fundamental component of currents or to the switching power converter supply.

In this study, a current sensor faults tolerant control method is proposed for synchronous machines. The proposed method is based on the estimation of the stator currents. A comparison algorithm between the estimated and measured currents allows detecting a possible fault in the current sensors. Once a fault is detected in the current sensors, the control system is switched to the current sensorless control. This transition to sensorless control mode is achieved quickly without stopping or slowing down the rotor speed. To validate the proposed method, simulations and experimental tests are carried out on a wound rotor synchronous machine.

The paper presents an investigation of IGBT’s zero current switching (ZCS) losses in a quasi-sinusoidal current mode series resonant converter.Theoretically, losses does not occur when the IGBT is switched at zero current. However, experiments show the opposite and switching losses are not negligible.Manufacturer’s datasheet do not indicate IGBT’s performance under ZCS condition which results in a lack of information to reduce the impact of the converter losses in order to reach high power density.A parametric model for the stored charge evacuated from the IGBT during turn off process is proposed based on experimental results.The impact of the ZCS modulation scheme and the magnetizing inductance of the transformer on zero current switching losses is discussed.

This paper shares some solutions in order to implement a state-of-the-art synchronous Interleaved Boost Converter (IBC), based on gallium nitride (GaN) power transistors. The solutions discussed have been implemented and validated on a synchronous 4-phase IBC (IBC4) prototype operating at a switching frequency of 250 kHz, specially designed to control the electric power delivered by a Proton Exchange Membrane (PEM) fuel cell module to a lithium battery pack. This paper focuses on digital control, such as PWM signal generation and the MCU requirements to reach high switching frequencies. It also discusses the issues related to the propagation delay of the sensors used and how to address them. The high switching frequency enabled by GaN transistors, combined with this DC/DC converter architecture and its phase-shifted control strategy, might heavily strain the load of the single MCU embedded. The real-time management of the different control loops is therefore exposed.

This paper presents the Electromagnetic Transient (EMT) modeling and simulation of power electronics in Modelica, a declarative equation-based language. In this paper, modeling of switching components such as diodes, insulated-gate bipolar transistors (IGBT) and multi-level converters using ideal and nonideal components are investigated. A three-phase three-level and a single-phase two-level converter with an open-loop controller are simulated in Modelica and EMTP®. The accuracy and performance of simulations are compared using the variable and fixed-step solvers. Analytical solutions are used for verification of results as well.

In this paper, a first concept of monolithic integration of a fuse on a silicon PiN diode is realized and experimentally characterized. An integrated fuse on PiN diode allows fast cut-off, with low I2T (less than 2 A2.s) and short pre-arcing times (4–6 μs). These fuse-on-diode components are intended for fail-safe topologies power converter, aiming for more compact and reliable applications. The fuses were electrothermally designed using Comsol Multiphysics™ and TCAD Sentaurus™ simulations were carried out to study their integration on PiN diodes. Characterization and experimental tests were carried out after components realization.

In this paper, a distributed control is proposed for Distributed Energy Storage Systems (DESSs) and Renewable Energy Sources (RESs) power management in islanded Microgrid (MG). The power management strategy is designed to maintain generation/consumption balance, to ensure State of Charge (SoC) balancing of the DESSs and MG frequency/voltage (f & V) regulation. A fully distributed control without leader-follower strategy is used to manage the power flow between renewable generators, energy storage and consumption (critical and non-critical loads), to balance the SoC of the DESSs and to restore the frequency and voltage to their nominal value only thanks to low bandwidth communication. The strategy framework of the power management set the islanded MG in 04 operations modes (normal mode, PV active power curtailment mode and load shedding and reconnection mode) in order to provide a high quality and reliable power source in the islanded MG. A MATLAB/Simulink simulation is performed with a system of two Batteries Energy Storage Systems (BESSs), three loads (a critical/variable load and two non-critical/constant loads) and photovoltaic (PV) generator, in order to verify the effectiveness and the resilience of the proposed power management method in several operation modes.

This paper deals with the behavioural modelling of a multi-MW PV/Diesel hybrid power plant based on long term monitoring over years. The approach is based on the analysis of production data correlated with solar resources and fuel consumption. The links between PV power and irradiance at different points of energy conversion chain up to the AC point of common coupling are carried out. The adopted methodology is to proceed to a formulation of the behavioural model of PV production considering trends and statistics observations. The first step was data classification targeting causalities and consequences of main disturbances. The second step validates the established models with operating data. The aim is to provide relevant set for scenarios simulation that allows optimal design and energy management for such hybrid plant.

Changes will be required to handle the increased power flow in the network as the distribution infrastructure ages and the number of EVs and renewable grows. Designing and operating an intelligent network that reacts to changing power flows to ensure the optimal operation is the most economical option than fortifying the network with heavier cables. This work aims to build a model of a 13 bus medium voltage distribution network with high penetration of distributed energy resources and use it to analyze network conditions. The MATLAB Simulink software is used to model and evaluate the network. The outcomes suggest the model is promising and valid even when renewable generation is at low levels.

This paper proposes a strategy for the resources management and power allocation in an energy community. Especially, the fairness of the benefit sharing is assessed thanks to a metric introduced as a monthly net energy price (in c€/kWh) from the viewpoint of each individual and computed as the individual bill over the consumed energy. The community management decouples the operational (i.e., power dispatch) from the settlement phase (i.e., monthly community billing). In particular, the investigated billing approach is based on an optimization process with an additional constraint to limit the gap between the maximum and minimum identified prices over all the community members. This study then provides a new method to better address individual’s need in the community. The results show a narrow range of the individual energy price and 11.5% collective bill reduction compared to a case where the members act individually.

Since the electrical grid creation, assessing the electricity demand is essential as we need to match the energy production/demand at all times. Load analysis is essential in improving the reliability and efficiency of the grid. Beside regular human activities, the main impact factor which explains consumption variations is the outside temperature. But there are still unpredictable variations that are mainly coming from arising social events. To build a better understanding of these variations, this work will focus on how to detect these events from social media and how to quantify their impact on residential and professional typical profiles for energy demand.

The supply of electricity to remote areas such as islands or rural areas presents many challenges. To reduce costs, the use of renewable energy resources is recommended. In these circumstances, it is always important to improve the power quality in terms of waveform and power sharing between different distributed generators. In this paper, we focus on the power sharing between different distributed generators, and for this purpose, an improved decoupled control, proposed in our previous work, is studied to prove its effectiveness in providing wide range of stability. The internal control has been assimilated to a second-order filter and then, the improved decoupled control is studied to prove its effectiveness to ensure a wide range of stability. For this purpose, a microgrid composed of two distributed generators is studied. Its small-signal model including the distributed generators, the loads and the droop control laws that ensure the interconnection between the generators is revealed. A stability study in the sense of the indirect Lyapunov theory based on the evaluation of the eigenvalues of the system is performed to show the “local stability” in presence of different types of loads, the impact of the system and the control parameters on the eigenvalue is studied using the modal analysis technique. The validation of these results is proven by simulation

In this paper, a novel load-flow method for unbalanced bipolar dc microgrids (BDCMGs) is presented. The principles of this method are based on the modified augmented nodal analysis (MANA) formulation, which is generic and simple to formulate. An unbalanced BDCMG is also used to verify the validity of the proposed MANA-based formulation. The findings also substantiate that in BDCMGs, different connections of DGs can highly affect the bipolar voltage profiles, in presence of different line resistances and droop coefficients.

The growth of residential rooftop solar PV has given rise to new operating concepts such as collective solar self-consumption where several prosumers come together to form a microgrid with its distributed PV generations, its storage systems, its local loads with an energy management system (EMS) to optimize the operation modes according to desired criteria. Such microgrid working is relatively complex and adapting the EMS of a commercial microgrid to meet national standards and regulations or to perform deep investigations requires first analysis and assessments. This paper deals with a commercial microgrid laboratory platform. Tests and data analysis are performed to establish the flowchart of its central EMS, and then, in a future work, to develop an accurate model of the platform to test new investigated EMS.

The increasing penetration of intermittent Renewable Energy Sources (RES) induces more instability of the grid and constraints on the Energy Management (EM). Microgrids (MG) are more and more experimented to better implement local flexibilities for dynamically balancing the production and load demand inside a specific area as districts of a city, as example. New solutions like Artificial Intelligence (AI) and Artificial Neural Networks (ANN) are being developed in order to improve the real-time energy management. Specifically this paper deals with the operational management of energy resources via the tuning of the frequency control parameters to satisfy the load demand. A non-exhaustive review of ANN techniques for enhancing the frequency control in microgrids is proposed. ANN techniques are shown to be performing better than other AI techniques on the specific cases reported here.

The fault diagnosis of Permanent Magnet Synchronous Motors (PMSMs) has been the subject of much research in recent days. This is due to the growing safety and reliability requirements for drive systems. This paper concerns detection and classification of the PMSM stator interturn short-circuits (ITSC) by using selected machine learning algorithms. The spectral analysis of symmetrical current components is applied for ITSC symptom extraction. The utilized and compared algorithms are K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Naive Bayes (NB) and Multilayer Perceptron (MLP). Experimental results confirm that the use of the KNN, SVM and MLP classifiers allows for ITSC detection with high effectiveness. The most effective is KNN, which is simple to implement and not computationally complex.

As a consequence of the enormous growth being experienced by renewable energy systems (RES), conventional technologies such as coal or gas, which unlike RES are dispatchable, are reducing their participation in energy markets, increasing the instability of the electric energy systems. Therefore, there is a need for RES or other converter-based technologies to replace the traditional ancillary service providers. In this sense, battery energy storage systems (BESS) are considered the best candidates. This paper defines a model to simulate the Spanish Automatic Generation Control (AGC). This simulator is used to provide inputs to the operation of a BESS that participates in the secondary frequency regulation market and in the continuous intraday energy market. Subsequently, the paper introduces an economic feasibility study to determine the best BESS size to operate simultaneously in both markets. The results obtained show that BESS with energy capacities of 2 h are the best option (from both a technical and an economic point of view) to be part of a regulation zone.

This work is inserted in the context of second life Li-ion batteries: for such storage devices, their first life characteristics are unknown and a simple capacity measurement might not be sufficient to fully characterize and get it ready for its second life. The Incremental Capacity Analysis (ICA) was used in this study to give a more intimate diagnosis of the batteries’ Degradation Modes (DMs), providing a link with physical degradation phenomena. This method was applied to a lithium-ion battery module (NMC/Graphite) which was used in an electrical vehicle and to a single cell from a similar module in order to verify its potential use in this context. Both IC curves were then compared to a DM simulation using the ′Alawa software, capable of simulating different ageing phenomena and their effects on the IC curves. Moreover, this work gives an intrinsic view and explanation of the IC signature for the mentioned battery technology.

The development of a battery monitoring system is one of main tasks for applications needing an efficient and well-designed battery storage system. In this framework, a fast, on-board, non-invasive and low-cost diagnosis system has a primary importance. Among the large number of diagnosis techniques, the Electrochemical Impedance Spectroscopy (EIS) is one of the most powerful. It allows one to extract information about the overall state of an electrochemical cell by stimulating it with current or voltage signals with appropriate shapes and frequency. In this work, we present the changes made to a commercial Lithium-ion battery charger to implement a system for the generation of EIS stimuli, preserving large part of the native functions of the battery charger. The stimulation functions are implemented using a field-programmable gate array (FPGA) board, which ensures a good voltage resolution and an optimal frequency range for this kind of applications.

Electrochemical storage systems play an increasingly central role in microgrids, providing several services which allow for more flexible and reliable operation. Lifetime of battery storage systems is a critical aspect to consider for their sustainable and cost-effective employment. In this paper a survey of energy management systems (EMSs) designed to contribute to battery lifetime extension is presented. To pursue this objective, the design of EMSs must rely on suitable battery degradation models, the most significant of which have been retrieved from the technical literature and described as well.

The work provides early results obtained with a multichannel EIS system, which were used to identify an equivalent circuit of an Industrial Scale Vanadium Redox Flow Battery (IS-VRFB) stack with a rated power/energy of 9 kW/27 kWh. The single cell impedance is represented with three different models, including a series resistance and an RC loop (RRC model), or a constant phase element (CPE) loop (a ZARC element), or a ZARC element including also a Warburg impedance. The inclusion of the CPE constitutes a substantial improvement in the fit. Conversely, the addition of the Warburg element, which aims to model the mass transfer in the electrochemical process, does not produce significant effects for the frequencies at which we have experimental data. This numerical results are validated against EIS measurements taken on IS-VRFB. Very few analyses of this type are reported in the literature for such batteries. This study set the stage for developing advanced online State of Health (SOH) management for IS-VRFB.

Lithium-ion batteries have the advantages of high energy density, high charge-discharge efficiency, low self-discharge effect and long cycle life that make them suitable in both stationary and mobile applications. They are the most widely used solution in the field of electric vehicles and are increasing their application for stationary applications. Both the life-time and performances are negatively affected by high temperatures so the prevision of the thermal behaviour is a crucial step in the battery modelling.Based on an experimental setup, a simplified thermal model was developed to estimate the surface temperatures of a lithium titanate cell from current and voltage measurements.The model was implemented in the COMSOL Multiphysics® Finite Element code. Charge and discharge cycles of the cell were performed and the predicted heat generation used as input of the thermal model. The calibrated model was lastly used to assess two thermal battery management (TBM) cooling systems, in this case applied to a single cell: a passive phase change material (PCM) system and a hybrid PCM/water system. The effects of the PCM thickness and velocity inlet of the water on the cell temperature were investigated. Results showed that, in comparison to the passively air cooled cell, both systems decreased the maximum surface temperatures, thus improving the uniformity of the temperature distribution and keeping the battery in a safe temperature range.

The thermal runaway (TR) is the main safety concern of lithium-ion batteries (LIBs). Methods for predicting and preventing TR are critical to achieve greater battery safety. Many researchers have studied the reactions that take place inside the cell and that because of their exothermicity trigger the TR. In this work the coupled electrochemical-thermal model for a lithium-ion cell was extended with contributions from exothermic reactions based on an Arrhenius law to model mechanisms of abuse, which could lead to a thermal runaway. Firstly, differential scanning calorimetry (DSC) tests were conducted on the individual components of the cell to characterize the reactions of the TR process in terms of onset temperature, thermal and kinetic parameters. The kinetic parameters of each reaction were identified by the Kissinger method. Then the thermal and kinetics parameters of the reactions occurring during the thermal runaway together with the phenomena involving the electrolyte (i.e., evaporation, boiling and venting) were included in the Battery and Fuel Cell Module of COMSOL Multiphysics simulator, to simulate the behaviour of a cylindrical 18650 cell under thermal abuse conditions. In particular, the results of the model appear to agree with the experimental data, concerning to a NCA 18650 cell subjected to radiative heat flux in a cone calorimeter.

The aim of this article is to introduce COFFEE, a concept of open and collaborative platform in the field of electrical engineering. The platform intends to make energy research accessible, and improve collaborations between researchers, public authorities, design offices and citizen collectives. The COFFEE concept is presented supported by a literature review on open energy modeling and collaborative platforms. Following a “user experience” inquiry conducted with a representative panel, the results are used to specify a first implementation of the COFFEE concept, and can serve as guidelines for the implementation of open energy modelling platforms. These platforms could become the spearheads of electrical engineering laboratories, promoting reproducibility and collaborations between energy stakeholders.

The increasing electrification of urban public transports requires improving the design of the electrical infrastructures to take into account all the technical and financial challenges involved in the creation of a new line. This paper presents a new optimization tool dedicated to the sizing of tramway electrical infrastructures: power substations, overhead transmission lines, feeders and equipotential bonding. The purpose is to determine the number, positions and technical characteristics of all these components to achieve the best trade-offs between investment costs, energy costs and the quality of the traffic power supply. The sizing problem is formulated as a multi-objective optimization problem and solved using the NSGA-II genetic algorithm. The proposed method is applied to a simple test case and gives good results.

Accurate modeling of bifacial module energy production is conditioned to the correct modeling of the front and rear irradiance. This paper compares the existing approaches used to estimate the incident irradiance on the back side and the front side of a photovoltaic (PV) bifacial module, by studying the performance of each model in terms of accuracy and computation time. In this study, we have selected three software with different approaches. We started with Bifacial_radiance which uses the ray-tracing technique. The second software is Sandia model which is a three-dimensional implementation of view factor method under MATLAB™. We complete our study with pvfactors that employs a two-dimensional configuration factor model. This study aims to propose the most time-efficient way to compute the irradiances received by bifacial panels, which will serve to predict the energy production of power plants. Having a fast model allows to develop efficient real-time management strategies for power supply systems that use bifacial modules. According to this study, pvfactors has the lowest execution time and gives almost the same output results as Bifacial_radiance and Sandia model that use complex algorithms.

This article presents a generic and self-adaptive construction algorithm for a surrogate model. This method makes use of two major tools: Latin HyperCube, which serves to efficiently spread a large number of samples; and Kriging, which is very efficient for surrogate modeling in the domain of black box models. The efficiency of this method is investigated in the case of a finite element model of a surface permanent magnet synchronous machine. During this study, Kriging surrogate models are compared with various samples in terms of both accuracy of construction and calculation speed. Next, the self-adaptative algorithm is applied in order to derive an accuracy criterion in a minimal amount of time and compare one with a Kriging model built using the same number of samples, yet without our tool to determine any accuracy lost due to the black box feature of the model and the hypotheses used.

Electric load forecasting is of utmost importance for governments and power market participants for planning and monitoring load generation and consumption. Reliable Short-Term Load Forecasting (STLF) can guarantee market operators and participants to manage their operations correctly, securely, and effectively. This paper presents the optimization of neural networks for power forecasting by means of whale optimization algorithm: two types of artificial neural networks namely, Feed-Forward Neural Network (FNN) and Echo State Network (ESN) have been used for STLF. ESN’s simplicity and strength have room for improvement. Therefore, an optimization algorithm called the Whale Optimization Algorithm (WOA) has been used to improve ESN’s performance. WOA-ESN was used for STLF of the first case study, namely Puget power utility in North America. The considered forecasting error indicators showed significant accuracy and reliability. WOA-ESN model and recursive approach resulted in better accuracy measures in terms of standard performance metrics.

The efficiency of direct-start applications such as pumps or fans can be improved by replacing a squirrel cage induction motor (SCIM) with a line start permanent magnet synchronous motor (LSPMSM). LSPMSM is a super-premium efficiency class IE4 motor, which combines the features of both conventional SCIM and permanent magnet synchronous motor (PMSM). In this paper, a reluctance network approach (RNA) is devised to estimate the maximum steady-state torque of LSPMSM. A reluctance network (RN) in both nonlinear and linear conditions is utilized to investigate the effect of flux-bridge saturation on the computed back electromotive force (EMF). The value of back EMF calculated from RNA is used to calculate the steady-state torque of LSPMSM. Finally, a two-dimensional (2D) finite element method (FEM) simulation is performed to validate the results obtained by the proposed model.

With the development of axial flux technology and industrial evolution, traditional machines cannot fit application requirements. Radial flux machines represent the majority of machines in high-speed application but they are not always an optimal solution according to the criteria of the considered applications. The design of the high-speed axial flux machines is challenging where many multi-physics critical issues remain to be solved. This paper reviews the high-speed axial flux machines in terms of different features such as machine types and designing structure, mechanical constraints, specific losses, materials, and application domains. The purpose is to give an overview of different technics and solutions in the literature to meet the needs of the high-speed axial flux machines to investigate their development and integration in different applications.