Neural Network-Driven Sensorless Speed Control of EV Drive Using PMSM

To reduce pollution and energy consumption, particularly in the automotive industry, saving is main concern, hence, Electric vehicles (EVs) are getting significantly more attention than with internal combustion engines (IC engines). motors used Vehicles must have high efficiency for maximum utilization of from batteries or fuel cells. Also, these be compact, lightweight, less expensive very easily recycled. Further, to obtain better dynamic performance, various motor control strategies speed motor. And increased reliability, sensorless techniques that offer sufficiently performance used. The largely divided into three groups: state observer methods, indirect measurement methods saliency-based methods. Generally, uses back emf flux linkage estimate Since directly proportional rotor speed, at low-speed based method will give poor performance. current-based Model Reference Adaptive System (MRAS) model also popular estimating low speed; however, assessments deteriorate during applications such as EV. This paper presents an artificial neural network (ANN)-deployed permanent magnet synchronous (PMSM) drive EVs. In this paper, estimation using MRAS discussed compared proposed ANN-based controller, which shows significant improvement EV drives. MATLAB simulation experimental results presented validate algorithm.