Optimization of BLDC-based electric vehicles: Vehicle dynamics modelling through dual-motor approach and designing a novel augmented TLBO algorithm for PID control

Abstract

The work primarily focuses on increasing the efficiency of EV drive in electric two-wheeler by working on several aspects, such as modulating the vehicle’s design, optimizing the control strategy, and increasing the speed range using a dual-motor approach. The dynamics of electric two-wheeler have been discussed with a mathematical vehicle model and further tuning of several aspects. Besides, this paper also introduces a novel Augmented Teaching and Learning based Optimization (ATLBO) technique designed exclusively to control BLDC motors for the electric two-wheeler vehicle. Besides, the designed technique has been implemented for the widely used commercial e-bike of Hero Company. Most two-wheeler electric motors are mounted on the rear side of the wheel, which has a limited speed range. Therefore, an analysis has been performed to increase the vehicle’s speed range using a dual motor, from 45 km hr−1 to 62 km hr−1, proving to be a viable alternative to a single motor generally used in an electric bike. ATLBO technique has been designed against a conventional TLBO to optimize the proportional-integral-derivative (PID) controller for the speed control of a linear brushless DC (BLDC) motor. The proposed method has several advantages, including ease of implementation, a consistent convergence characteristic, and high computational efficiency. Furthermore, the literature has validated the merits of the presented novel control technique. The only disadvantage of using a dual motor is the initial cost, but the overall cost is moderated in the long-term usage for its augmented performance parameters. The performance parameters of the above technique are analyzed against other optimization techniques like conventional Teaching and Learning based optimization (TLBO), Particle Swarm Optimization (PSO). MATLAB/Simulink models the brushless DC motor and implements ATLBO, TLBO, and PSO algorithms. It has been found that the response obtained from ATLBO is comparatively much faster than other optimization techniques, which supports the motor for quick acceleration as well as more efficient in improving the step response characteristics such as rise time, settling time, and steady-state error in the speed control of a linear BLDC motor.