Comparative Performance Analysis of PI Speed Controller Tuning Techniques for a Vector-Controlled BLDC Motor Drive

Abstract

The performance of PI based speed controller for Brushless DC motor Drive has been analysed in this paper. BLDC motors demand precise control strategies to mitigate steady-state errors and enhance transient response in high-performance applications like electric vehicles. Traditional manual tuning methods may not provide optimal PI controller gains. Therefore, this study uses three optimization techniques, namely Particle Swarm Optimization (PSO), Firefly algorithm (FA), and Cuckoo Search Algorithm (CSA), to tune the PI speed controller. The performance of the controllers is evaluated under three different operating conditions: start-up, urban driving cycle, and hill-climbing conditions, representing typical real-world operating conditions of an EV. The speed response is analysed in terms of speed overshoot, speed error, speed ripple, rise time, and settling time. The response of the controllers under load variations is also examined to evaluate their ability to maintain the desired speed when an external load acts as a disturbance. In the urban driving cycle, the controller performance is further analysed under frequent speed increase, decrease, and short-duration stops, representing typical traffic conditions. The results show that the optimized PI controllers provide improved speed response compared with the conventional PI controller. Among the considered techniques, the CSA-PI controller provides lower speed overshoot, speed error, and speed ripple, along with a faster rise time, while during loading or uphill condition the FA-PI perform well as it give only 0.2% speed dip during sudden load change. However, this study provides a basis for selecting a suitable PI tuning technique according to the operating requirements of BLDC motor drives used in EV applications.