Certain investigations on performance enhancement of direct torque controlled brushless dc motor using meta heuristic methods

Abstract

Advent of Electric vehicles (evs) and Autonomous Electric Vehicles (aevs) introduces a bigger scope of improvement in the high precision control of Permanent Magnet Brushless Direct Current Motor (PMBLDCM). Highe recision application of PMBLDCM also include the CNC machines, Actuators in robotics, servos etc. Thus pointing the researchers towards the higher precision and efficient working of the PMBLDCM. This research is an attempt to develop a higher precision and accurate control of the speed in the PMBLDCM using the optimized Proportional Integral (PI) controller. The model less algorithms for parameter estimation has grown in a better manner in the recent years. Thus, the mathematical modeling of the PI controller using the different tuning method is replaced with the parameter tuning method. The stochastic nature of the PI controller is utilized for the higher precision and accurate control of the speed and the torque ripple in the PMBLDCM. The optimization algorithm like the Particle Swarm Optimization (PSO) and Cuckoo Search Algorithm (CSA) is utilized to optimize the proportional and integral gains of the PI controller are used in the Direct Torque control of PMBLDCM. MATLAB based simulation is carried out by developing the Simulink model of the PMBLDC motor and developing the optimization algorithm using the Matlab M-file. The performance evaluation of the Direct Torque control (DTC) of the PMBLDCM is carried out and comparing the optimized PI controller with the Fuzzy controlled DTC, PID controlled DTC. The optimized PI controller includes the PSO optimized PI and CSA optimized PI controller. It is observed that the starting current transients is improved with smooth current iv waveform obtained while fuzzy, PSO-PI and CSA-PI controller is applied in the DTC of PMBLDCM. The settling time of the speed curve has improved while the Fuzzy, PSO-PI and CSA-PI controllers are utilized. While the CSA-PI algorithm performs better both in the starting current transients reduction and the settling time reduct