Design, control and performance comparison of PI and ANFIS controllers for BLDC motor driven electric vehicles

Abstract

The research and usage of electric vehicles (EVs), including two and four-wheeler vehicles, are rapidly increasing worldwide as alternatives to oil/gas-based vehicles. Brushless direct current (BLDC) motors are popular for industrial and traction applications due to their inherent advantages. In EVs, achieving low error in steady-state and transient responses is crucial for smooth acceleration at the wheel. This paper presents the design and control of a BLDC motor for speed control during acceleration and deceleration, considering error as a key factor in the MATLAB/Simulink environment. Proportional-integral (PI) and fuzzy controllers are commonly used for motor control to improve steady-state and transient performance, thereby reducing error. In this study, the PI and adaptive neuro-fuzzy inference system (ANFIS) controllers are designed and compared for a 5-kW, 48-V, and 100-Amp BLDC motor in EV applications. The results demonstrate that the ANFIS controller enhances the dynamic performance of the BLDC motor and improves other operating characteristics such as rise time, settling time, peak overshoot percentage and the vehicle response in terms of speed and distance.