Analysis of permanent magnet synchronous motor using artificial neural network for electric vehicles

Abstract

This paper presents a modern approach of speed control for Permanent magnet Synchronous Motor (PMSM) using online artificial neural network (ANN). The overall system will be simulated under various operating conditions. The use of ANN as a controller makes the drive robust, with faster dynamic response higher accuracy and insensitive to load variation .On line training of the ANN is carried out and the results show fast convergence. After training the ANN the system is tested for a step change in load, the simulation results showing good dynamic response with fast recovery time.The DC motors have been widely used as precision control motors till mid 1970s. Recent developments in microprocessors, magnetic materials and semiconductors technology have offered an excellent opportunity to use AC motors in high performance drives systems. PMSM became at the top of ac motors in the medium range of power and it became very popular choice in drive technology over the last few years due to some of its inherent advantages [l].,These advantages include high torque to current ratio, large power to weight ratio, higher efficiency and robustnessThe speed controller used in drive system plays an important role to meet the other required criteria of the High Performance Drive. It should enable the drive to follow any reference speed tacking into account the effects of load impact, saturation and parameter variations. Conventional controllers such as P, PI and PID controllers need accurate mathematical models describing the dynamics of the system under control [2]. But these types of controllers are difficult to design if an accurate system model is not available. Moreover, unknown load dynamics and other factors such as noise, temperature, saturation, etc. affect their performance. Some adaptive control techniques, such as the variable structure and self -tuning don't need a model for system dynamics, the dynamic model is rather developed based on the on-line input/output response of the system. Unfortunately, these methods track only linear systems. But some investigator overcame this difficulty by updating these models every several sampling intervals [3-61.Although these adaptive controllers are proven to be effective, their hardware implementation is elaborate .ANN using parallel and distributed processing units can achieve the functions of system modeling and control for nonlinear systems [I]. Artificial neural networks have several key features that make their highly suitable for high performance drive applications. For example, ANN can cr 0-78034482-2104/$20.00 02004 IEEE. 143

Cuckoo Search Based DTC of PMSM

“…Recently, researches are made in DTC of PMSM based on FL-type-1, FL-type-2 and artificial neural network and in controlling PMSM using the DTC strategy with PS optimization technique or by using combination of control strategies and optimization techniques like tuning PI-controller using swarm algorithm or intelligent FLC using PS optimization [10][11][12][13][14][15][16]. Nowadays, PS algorithms have become famous due to its simplicity.…”