Search

International audience; Automotive industry is evolving towards more fault-tolerant actuators to fulfill future autonomous vehicles requirements. Critical applications such as steering or braking have to resist to fault occurrences while being low cost due to mass production market. Considering these two criteria, multiphase electric drives offer a good tradeoff between increasing the degrees of freedom and limiting system oversizing. This paper proposes to compare three multiphase electric drives for electro-hydraulic power steering for trucks: a five-phase machine and a six-phase machine both fed by a full bridge inverter (FBI), and a seven-phase machine fed by a standard half-bridge inverter (HBI). Several comparison criteria are considered in the meantime: motor, electronics, control and manufacturing parameters. Some of criteria are obtained from finite-element analysis (FEA), while others are derived from analytic formula or dynamic simulations . Criteria are evaluated for each drive and then results are discussed. In the given low-voltage high-current application, H-bridge inverter topology seems to be a promising solution. Both five-phase and six-phase present similar results. However, six-phase drive could be more interesting, as it could be brought close to standard three-phase solutions and adapted to dual-lane supply.

International audience; Automotive safety applications will be soon a standard based on market evolution trends. Actuators performing safety operations, such as steering or braking, need to be more reliable with a low manufacturing cost. Multiphase drives seem to be a good alternative to actuators redundancy, as it offers greater degrees of freedom with at a limited extra cost. These drives have been studied for a while, especially in other industries such as ship propulsion or aeronautic actuator. However, comparison between multiphase drives for a given application is rare. In this paper, a methodology is proposed to evaluate several drives for a safety automotive application. Some manufacturing assumptions have been made in order to narrow the study to five, six and seven phase permanent magnet synchronous motors (PMSM). Machines have been modeled using 2-D finite elements analysis (FEA) software in order to extract their parameters which will be used in dynamic simulations. Based on the obtained results, each drive is evaluated according to defined criteria and best solutions are highlighted. Considering a low voltage high current application, full bridge inverter topologies feeding a six-phase motor seem to be a promising solution.

Multiphase permanent magnet synchronous machines (PMSMs) have started to be used more and more in the electrification of transportation systems. These systems have several specific requirements such as: high efficiency, high power density, high functional reliability, robustness, and compactness (easier for integration). Basically, to perform the vector control of multiphase PMSM drives, the essential information of the rotor position and speed is obtained by a position sensor mounted at the end of the rotor shaft. In the context of high functional reliability and compactness, the replacement of the position sensor by a “soft” position sensor, using only electrical sensors (currents) is interesting. This replacement helps to avoid the sensitivity of the position sensor to the environment and the necessity of an available rotor shaft end. Therefore, the purpose of this doctoral thesis is to perform accurate sensorless control algorithms for multiphase PMSM drives with the aim of enhancing their reliability and compactness. In this thesis, an algorithm based on the sliding mode observer is proposed to estimate the rotor position and speed by using the estimated back-electromotive forces of all fictitious machines of multiphase PMSM. To solve the problem of PMSM observability at standstill and low speed ranges, an algorithm based on the high-frequency signal injection and specificities of multiphase machines is developed. These algorithms guarantee the precise sensorless control of multiphase PMSM drives in all speed ranges. Experimental results of a non-sinusoidal seven-phase PMSM are given to verify the effectiveness of the proposed sensorless control algorithms in healthy and degraded mode.; L’électrification des systèmes de transport nécessite l’émergence d’entrainements électriques qui répondent à certaines exigences en termes de : rendement et densité de puissance élevés, fiabilité fonctionnelle, robustesse et compacité (facilité d’intégration). Dans ce contexte, les machines synchrones à aimants permanents (MSAP) polyphasées (nombre de phases > 3 ) sont devenues de plus en plus étudiées en raison de leurs densités énergétiques volumique et massique élevées, et leurs tolérances aux défaillances. Pour assurer un contrôle vectoriel performant de ces machines, un capteur de position mécanique est monté sur l’arbre de la machine pour fournir l’information de la position et de la vitesse du rotor. Cependant, dans un contexte de fiabilité fonctionnelle et de robustesse des machines MSAP polyphasées, un défaut sur ce capteur de position ou sur l’une de ses connexions entraine directement une défaillance du système de contrôle. Par ailleurs, dans un contexte de compacité, la nécessité d’un bout d’arbre pour monter ce capteur, qui lui aussi occupe du volume, entraine une augmentation du volume global des entrainements polyphasés. L’objectif de la thèse est de développer des algorithmes spécifiques à la commande sans capteur de machines MSAP polyphasées à force électromotrices non-sinusoïdales dans le but d’augmenter leurs fiabilités fonctionnelles et leurs compacités. Pour cela, des algorithmes basés sur le principe de l’observateur à mode glissant sont développés pour estimer la position et la vitesse du rotor à partir de l’estimation des forces électromotrices des différentes machines fictives équivalentes. Pour résoudre le problème d’observabilité des machines MSAP à zéro et faible vitesse, des algorithmes de commande sans capteur basés sur l’injection d’une tension sinusoïdale à haute fréquence dans la machine fictive qui contribue faiblement à la production du couple sont développés. L’obtention de la position sur toute la plage de vitesse des machines MSAP polyphasées en mode normal et dégradé peut ainsi être obtenue. Les différents algorithmes proposés dans ces travaux de thèse sont validés expérimentalement sur une machine MSAP à 7 phases à forces électromotrices non-sinusoïdales.

National audience; This paper proposes a sensorless control based on Sliding Mode Observer (SMO) for a Five-phase Interior Permanent Magnet Synchronous Machine (FIPMSM), with a consideration of the third harmonic component. Compared to the conventional three-phase machines, the third harmonic of electromotive force contains more information. Thus, in this paper, we consider the first and third harmonic components of the five-phase machine to estimate the rotor position that is necessary for the control. Simulation results of the implemented SMO are shown to verify the feasibility of the proposed sensorless control strategy

International audience

International audience; This paper proposes several easy-to-implement control strategies when seven-phase axial flux brushless DC machines with trapezoidal back electromotive forces operate in normaland faulty modes by taking into account constraints on voltage and current. The constraints are related to the converter and machine design in terms of maximum values of current andvoltage. The considered faults are cases in which one or two phases of the machine are open-circuited. Numerical computations based on analytical formulations are applied toobtain torque-speed characteristics, including the fluxweakening operation. The methods determine current references to ensure the torque optimizations while currents and voltages are within their limits. The usefulness of the methods is verified by numerical results.

International audience; Multiphase machines have recently gained importance in the research community for their use in applications where high power density, wide speed range and fault-tolerant capabilities are needed. The optimal control of such drives requires to consider voltage and current constraints imposed by the power converter and the machine itself. If classical three-phase drives have been optimally controlled under such limits for a long time, the same cannot be said in the case of multiphase drives. This paper deals with this issue, where an optimal control technique based on Cascaded Model Predictive Controls (MPC) is presented for a five-phase permanent magnet synchronous machine (PMSM). A Continuous-Control-Set MPC (CCS-MPC) numerically computes optimal current references in real-time in order to exploit the maximum performance for given DC bus voltage and current limits. Then, a Finite-Control-Set MPC (FCS-MPC) is used to carry out the current control in the machine, directly applying the switching state that minimizes a cost function related to the current tracking. Obtained mixed microprocessor-based and FPGA-based real-time simulations prove the interest of the proposal, which ensures the optimal control of the multiphase drive operating under current and voltage constraints.

International audience; Dans le cadre d'une commande en mode dégradé d'un entrainement comprenant deux machines polyphasées connectées en série et pilotées de façon indépendante, l'article s'intéresse à analyser la performance de l'entrainement dans le cas d'un défaut de court-circuit d'un transistor. Deux reconfigurations modifiant l'algorithme de commande ont été présentées avec les objectifs d'éliminer l'ondulation de couple et de comparer la valeur de courant crête obtenue avec chaque stratégie en vue d'obtenir un dimensionnement de l'entrainement au plus juste.

International audience; Cet article a pour objectif de caractériser différentes configurations de bobinages, de génératrices synchrones à aimants permanents déposés en surface, polyphasées pour des applications faibles vitesses et fortes puissances, associées à un multiplicateur de vitesse à faible rapport et une turbine hydrolienne à pas fixe. L'objectif est de comparer les structures polyphasées à bobinages concentrées à une structure de référence à bobinage distribué avec un pas diamétral et une encoche par pôle et par phase. L'idée est de qualifier l'influence du bobinage et de la structure polyphasée choisie, en termes de capacité de fonctionnement en défluxage en mode normal et dégradé pour un même cahier des charges relatif à une génératrice hydrolienne. L'étude concerne des machines à 5 phases régulièrement reparties et des machines à double étoiles triphasées décalées de 30°.

This paper deals with the real time Fault Detection and Identification (FDI) process of inverter Open Switch Fault (OSF) and Open Phase Fault (OPF) in five-phase PMSM designed for aerospace applications in which the electric drive system has particular operating characteristics either in healthy states or in the faulty ones. Two original contributions are considered in this paper. They consist in normalizing the input variables of the FDI process applied to multiphase system and compensating the noise (switching and sensors noises) and dccomponent resulting from the fault occurrence. The proposed strategy uses multiple normalized criterias derived from the measured phase currents and the references currents obtained from the outputs of the speed controllers. The FDI shows independence with respect to transient states and to switching and measurement noises. Moreover, it can be easily included in an existing software without any additional sensors. The validity of the proposed method is verified by Matlab/Simulink simulation tests.

This paper presents a new idea by using the Artificial Neural Networks (ANNs) for estimating the parameters of the machine which achieving the maximum efficiency of the Synchronous Reluctance Motor (SynRM). This model take into consideration the magnetic saturation, cross-coupling and iron loss. With Finite Element Analysis (FEA), the characteristics of the SynRM including inductances and iron loss resistance are determined. Because of the non-linear characteristics, an ANN trained off-line, is then proposed to obtain the d-q inductances and iron loss resistance from Id,Iq currents and the speed. After learning process, an analytical expression of the optimal currents is given thanks to Lagrange optimization. Therefore, the optimal currents will be obtained online in real time. This method can be achieved with maximum efficiency and high-precision torque control. Simulation and experimental results are presented to confirm the validity of the proposed method.

This paper presents an original method, based on artificial neural networks, to reduce the torque ripple in a permanent-magnet non-sinusoidal 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. The study 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. Either the torque or the speed control scheme, both integrate two neural blocks, one dedicated for optimal currents 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. CPER Région Alsace 2007-2013