Groupe de Recherche en Énergie Électrique de Nancy

Human cognitive control is uniquely flexible and has been shown to depend on prefrontal cortex (PFC). But exactly how the biological mechanisms of the PFC support flexible cognitive control remains a profound mystery. Existing theoretical models have posited powerful task-specific PFC representations, but not how these develop. We show how this can occur when a set of PFC-specific neural mechanisms interact with breadth of experience to self organize abstract rule-like PFC representations that support flexible generalization in novel tasks. The same model is shown to apply to benchmark PFC tasks (Stroop and Wisconsin card sorting), accurately simulating the behavior of neurologically intact and frontally damaged people.

This paper presents an energy management method in an electrical hybrid power source (EHPS) for electric vehicular applications. The method is based on the flatness control technique (FCT) and fuzzy logic control (FLC). This EHPS is composed of a fuel cell system as the main source and two energy storage sources (ESSs)-a bank of supercapacitors (SCs) and a bank of batteries (BATs)-as the auxiliary source. With this hybridization, the volume and mass of the EHPS can be reduced, because the high energy density of BAT and high power density of SC are utilized. In the proposed novel control strategy, the FCT is used to manage the energy between the main and the auxiliary sources, and the FLC is employed to share the power flow in the ESS between the SC and the BAT. The power sharing depends on the load power and the state of charge of the SC and the BAT. EHPS is controlled by the regulation of the stored electrostatic energy in the dc buses. The main property of this strategy is that the energy management in the power source is carried out with a single general control algorithm in different operating modes, consequently avoiding any algorithm commutation. An EHPS test bench has been assembled and equipped with a real-time system controller based on a dSPACE. The experimental results validate the efficiency of the proposed control strategy.

Electric motor drives and power electronic converters have become increasingly common in advanced power systems. Passive LC filters are used in these systems to reduce the power ripples. These filters are usually poorly damped for reducing the losses as well as the size/weight and the cost of the system. This leads to instability phenomena if the load power exceeds a power limit depending on the filter parameters. The purpose of this paper is to present tools allowing large signal stability analysis of a dc power system. These tools allow estimation of the domain of attraction of the system operating point. It will be shown that this large signal stability analysis gives useful hints on the design of the system to optimize the stability criteria for constant and variable power loads. The impact of the load dynamics on stability is also studied. An electric drive connected to a dc power supply through a poorly damped LC filter is used as a case study. The simulations and the experimentations confirm the analytical results.

This paper studies the impact of fuel-cell (FC) performance and control strategies on the benefits of hybridization. One of the main weak points of the FC is slow dynamics dominated by a temperature and fuel-delivery system (pumps, valves, and, in some cases, a hydrogen reformer). As a result, fast load demand will cause a high voltage drop in a short time, which is recognized as a fuel-starvation phenomenon. Therefore, to employ an FC in vehicle applications, the electrical system must have at least an auxiliary power source to improve system performance when electrical loads demand high energy in a short time. The possibilities of using a supercapacitor or a battery bank as an auxiliary source with an FC main source are presented in detail. The studies of two hybrid power systems for vehicle applications, i.e., FC/battery and FC/supercapacitor hybrid power sources, are explained. Experimental results with small-scale devices (a polymer electrolyte membrane FC of 500 W, 40 A, and 13 V; a lead-acid battery module of 33 Ah and 48 V; and a supercapacitor module of 292 F, 500 A, and 30 V) in a laboratory authenticate that energy-storage devices can assist the FC to meet the vehicle power demand and help achieve better performance, as well as to substantiate the excellent control schemes during motor-drive cycles.

Photovoltaic (PV) systems are one of the main actors in distributed power generation. In particular, in urban contexts, the PV generators can be subjected to mismatching phenomena due to the different orientation of the modules with respect to the sun rays or due to shadowing. In these cases, the maximum power point tracking (MPPT) function must be designed carefully. In this paper, architecture, including one dc/dc converter for each PV generator, is considered. The converters' output terminals are series connected to a high-voltage dc bus, where also a bidirectional dc/dc converter managing the power from/to a storage device is plugged. The functional constraints deriving from the dc/dc converters' connection, the mismatching phenomena, the MPPT capabilities of the inverter, connected with its input terminals at the dc bus, are taken into account in order to determine the best operating point of the system as a whole. The real-time constrained optimization problem is solved by using the particle swarm optimization method, which needs the knowledge of the actual current versus voltage curve of each PV generator. The practical impact of this need is also discussed in the paper. The feasibility and the performances of the proposed approach are experimentally validated by using a laboratory prototype.

For low-speed electrical machine applications, it is usually weight/size and cost effective to employ a high-speed machine together with a mechanical gearbox. However, the disadvantages associated with magnetic gearboxes can be overcome by mechanically and magnetically integrating a magnetic gear and a permanent magnet brushless machine, to create a ldquopseudordquo direct-drive machine. It is shown that a torque density in excess of 60 kNm/m3can then be achieved, at a power factor in excess of 0.9.

With the development of technologies in recent decades and the imposition of international standards to reduce greenhouse gas emissions, car manufacturers have turned their attention to new technologies related to electric/hybrid vehicles and electric fuel cell vehicles. This paper focuses on electric fuel cell vehicles, which optimally combine the fuel cell system with hybrid energy storage systems, represented by batteries and ultracapacitors, to meet the dynamic power demand required by the electric motor and auxiliary systems. This paper compares the latest proposed topologies for fuel cell electric vehicles and reveals the new technologies and DC/DC converters involved to generate up-to-date information for researchers and developers interested in this specialized field. From a software point of view, the latest energy management strategies are analyzed and compared with the reference strategies, taking into account performance indicators such as energy efficiency, hydrogen consumption and degradation of the subsystems involved, which is the main challenge for car developers. The advantages and disadvantages of three types of strategies (rule-based strategies, optimization-based strategies and learning-based strategies) are discussed. Thus, future software developers can focus on new control algorithms in the area of artificial intelligence developed to meet the challenges posed by new technologies for autonomous vehicles.

This paper introduces a new fault-tolerant operation method for a symmetrical six-phase induction machine (6PIM) when one or several phases are lost. A general decoupled model of the induction machine with up to three open phases is given. This model illustrates the existence of a pulsating torque when phases are opened. Then, a new control method reducing the pulsating torque and the motor losses is proposed in order to improve the drive performances. The proposed method is compared to two other existing techniques. The simulation and experimental results obtained on a dedicated test-rig confirm the validity and the efficiency of the proposed method for a fault-tolerant symmetrical 6PIM drive.

This paper presents an analytical subdomain model to compute the magnetic field distribution in surface-mounted permanent-magnet (PM) motors with semi-closed slots. The proposed model is sufficiently general to be used with any pole and slot combinations including fractional slot machines with distributed or concentrated windings. The model accurately accounts for armature reaction magnetic field and mutual influence between the slots. The analytical method is based on the resolution of two-dimensional Laplace's and Poisson's equations in polar coordinates (by the separation of variables technique) for each subdomain, i.e., magnet, air gap, slot-opening, and slots. Magnetic field distributions, back-EMF, and electromagnetic torque (including cogging torque) computed with the proposed analytical method are compared with those issued from finite-element analyses.

This paper discusses the design, implementation, experimental validation, and performances of a field-programmable gate array (FPGA)-based real-time power converter failure diagnosis for three-leg fault tolerant converter topologies used in wind energy conversion systems (WECSs). The developed approach minimizes the time interval between the fault occurrence and its diagnosis. We demonstrated the possibility to detect a faulty switch in less than 10 mus by using a diagnosis simultaneously based on a ldquotime criterionrdquo and a ldquovoltage criterion.rdquo To attain such a short detection time, an FPGA fully digital implementation is used. The performances of the proposed FPGA-based fault detection method are evaluated for a new fault tolerant back-to-back converter topology suited for WECS with doubly fed induction generator (DFIG). We examine the failure diagnosis method and the response of the WECS when one of the power switches of the fault tolerant back-to-back converter is faulty. The experimental failure diagnosis implementation based on ldquoFPGA in the looprdquo hardware prototyping verifies the performances of the fault tolerant WECS with DFIG.

It is known that the interaction between poorly dampedLCinput filters and constant power loads (CPLs) leads to degradation of dynamic performance or system instability. This paper addresses a large-signal stability study and stabilization of an electrical system containing a dc power supply, anLCfilter, and a CPL. This latter is realized here by a voltage source inverter supplying a motor drive. To stabilize the system, the control structure is slightly modified to implement a nonlinear stabilization block that virtually increases the dc-link capacitance and, hence, the damping of the system. The main idea consists in adding a capacitive power component to the CPL power reference. This allows reducing the real dc-link capacitance value and volume, which is, for weight and size reasons, an important issue in aerospace applications. The impact on the large-signal stability will be analyzed by estimating the domain of attraction of the operating point. An illustrative example consisted of anLCinput filter connected to an inverter-permanent-magnet synchronous motor designed for aircraft applications treated by simulations and experimentation, which confirm the validity of the proposed approach.

This paper presents a new space vector pulsewidth modulation (SVPWM) technique for the control of a six-phase voltage source inverter (VSI)-fed dual stator induction machines (DSIM). A DSIM is an induction machine which has two sets of three-phase stator windings spatially shifted by 30 electrical degrees and fed by two three-phase VSIs. Despite their advantage of power segmentation, these machines are characterized by large zero sequence harmonic currents, and in particular those of order$6k\pm 1$, which are due to the mutual cancellation between the two stator windings. The proposed SVPWM scheme, while easy to implement digitally, reduces significantly these extra stator harmonic currents. Experimental results, collected from a 15 kW prototype machine controlled by a digital signal processor, are presented and discussed.

In this paper, a model-reference-based online identification method is proposed to estimate permanent-magnet synchronous machine (PMSM) parameters during transients and in steady state. It is shown that all parameters are not identifiable in steady state and a selection has to be made according to the user's objectives. Then, large signal convergence of the estimated parameters is analyzed using the second method of Lyapunov and the singular perturbations theory. It is illustrated that this method may be applied with a decoupling control technique that improves convergence dynamics and overall system stability. This method is compared with an extended Kalman filter (EKF)-based online identification approach, and it is shown that, in spite of its implementation complexity with respect to the proposed method, EKF does not give better results than the proposed method. It is also shown that the use of a simple PMSM model makes estimated parameters sensitive to those supposed to be known whatever the estimator is (both the proposed method and EKF). The simulation results as well as the experimental ones, implemented on a nonsalient pole PMSM, illustrate the validity of the analytic approach and confirm the same conclusions.

In this paper, an efficient and reliable neural active power filter (APF) to estimate and compensate for harmonic distortions from an AC line is proposed. The proposed filter is completely based on Adaline neural networks which are organized in different independent blocks. We introduce a neural method based on Adalines for the online extraction of the voltage components to recover a balanced and equilibrated voltage system, and three different methods for harmonic filtering. These three methods efficiently separate the fundamental harmonic from the distortion harmonics of the measured currents. According to either the Instantaneous Power Theory or to the Fourier series analysis of the currents, each of these methods are based on a specific decomposition. The original decomposition of the currents or of the powers then allows defining the architecture and the inputs of Adaline neural networks. Different learning schemes are then used to control the inverter to inject elaborated reference currents in the power system. Results obtained by simulation and their real-time validation in experiments are presented to compare the compensation methods. By their learning capabilities, artificial neural networks are able to take into account time-varying parameters, and thus appreciably improve the performance of traditional compensating methods. The effectiveness of the algorithms is demonstrated in their application to harmonics compensation in power systems.

In this paper, an oscillation compensation technique is proposed to improve the stability margin of an electrical system constituted by a dc power supply, an LC filter, and a constant power load. This is realized here by an actuator (inverter-permanent-magnet synchronous motor). To design the compensator, input impedance of the constant power load and output impedance of the filter are required and derived in this paper. To develop the load input impedance expression, small signal approximation is employed and all dynamics are taken into account except by the inverter ones only, which can often be neglected in practical applications. Then, the control structure of the whole system is slightly modified to implement the oscillation compensation block that increases the stability margin, and thus, permits to reduce the dc-link capacitance value. In this paper, the proposed method is applied to an actuator designed for aerospace applications. The influence of the actuator control parameters and the input filter parameters on the stability of the dc-link voltage is discussed. Simulations and experimentations confirm the validity of the proposed approach.

This paper puts into perspective the state-of-the art knowledge on bearing currents and diagnosis tools, which rely on shaft voltages and bearing currents phenomena. Today, shaft signals measurements are progressively included in predictive health monitoring of large turbo and hydro generators. In addition, bearing fault diagnosis through shaft signals have recently gained some attention to improve reliability and estimation of the remaining useful life in electrical drives. The main objective of this paper is to provide diagnosticians with a broad overview of the current trends in fault diagnosis through shaft signals. Therefore, the study presents a critical review of the measurements methods, advantages, and disadvantages of fault diagnosis methods, and suggests future possible enhancements for bearing health monitoring through shaft signals.

This paper presents an original method, based on artificial neural networks, to reduce the torque ripple in a permanent-magnet nonsinusoidal synchronous motor. Solutions for calculating optimal currents are deduced from geometrical considerations and without a calculation step, which is generally based on the Lagrange optimization. These optimal currents are obtained from two hyperplanes. This paper takes into account the presence of harmonics in the back-EMF and the cogging torque. New control schemes are thus proposed to derive the optimal stator currents giving exactly the desired electromagnetic torque (or speed) and minimizing the ohmic losses. The torque and the speed control scheme both integrate two neural blocks, one dedicated for optimal-current calculation and the other to ensure the generation of these currents via a voltage source inverter. Simulation and experimental results from a laboratory prototype are shown to confirm the validity of the proposed neural approach.

Fault detection (FD) in power electronic converters is necessary in embedded and safety critical applications to prevent further damage. Fast FD is a mandatory step in order to make a suitable response to a fault in one of the semiconductor devices. The aim of this study is to present a fast yet robust method for fault diagnosis in nonisolated dc-dc converters. FD is based on time and current criteria which observe the slope of the inductor current over the time. It is realized by using a hybrid structure via coordinated operation of two FD subsystems that work in parallel. No additional sensors, which increase system cost and reduce reliability, are required for this detection method. For validation, computer simulations are first carried out. The proposed detection scheme is validated on a boost converter. Effects of input disturbances and the closed-loop control are also considered. In the experimental setup, a field programmable gate array digital target is used for the implementation of the proposed method, to perform a very fast switch FD. Results show that, with the presented method, FD is robust and can be done in a few microseconds.

Hydrogen is recognized as a promising and attractive energy carrier to decarbonize the sectors responsible for global warming, such as electricity production, industry, and transportation. However, although hydrogen releases only water as a result of its reaction with oxygen through a fuel cell, the hydrogen production pathway is currently a challenging issue since hydrogen is produced mainly from thermochemical processes (natural gas reforming, coal gasification). On the other hand, hydrogen production through water electrolysis has attracted a lot of attention as a means to reduce greenhouse gas emissions by using low-carbon sources such as renewable energy (solar, wind, hydro) and nuclear energy. In this context, by providing an environmentally-friendly fuel instead of the currently-used fuels (unleaded petrol, gasoline, kerosene), hydrogen can be used in various applications such as transportation (aircraft, boat, vehicle, and train), energy storage, industry, medicine, and power-to-gas. This article aims to provide an overview of the main hydrogen applications (including present and future) while examining funding and barriers to building a prosperous future for the nation by addressing all the critical challenges met in all energy sectors.

We present a method that minimizes the cogging torque in rotor surface mounted permanent-magnet motors. The key idea is to set the distribution of the air-gap flux density by segmenting the magnet pole into several elementary magnet blocks. By choosing either the appropriate elementary magnet block span or the relative position of the magnet blocks, the cogging torque may be significantly reduced. Our analytical approach uses Fourier series to predict the cogging torque harmonics, and finite-element computations. Our numerical results confirm the analytical conclusions.