Robust speed control of brushless dc motor based on adaptive neuro fuzzy inference system for electric motorcycle application

Electric vehicles have been widely discussed in some articles since the cost of fuel for conventional vehicles in this era is not stable and tends to increase. And also, conventional vehicles are also not fully eco-friendly and have poor efficiency. Electric vehicles, mostly, use Brushless Direct Current (BLDC) motor as the prime mover, since it has a simple structure, good performance and high efficiency. This paper presents an Adaptive Neuro Fuzzy Inference System (ANFIS) controller to control the speed of BLDC motor applied for electric motorcycle. ANFIS controller was designed and evaluated, then compared to Proportional-Integral-Derivative (PID) and Fuzzy-PID controllers. ANFIS is trained based on the data of Fuzzy-PID performances with slight modification. According to the study, ANFIS controller has better performances compared to PID and Fuzzy-PID controllers with average steady state error of 0.13% when the speed reference changes and 0.16% when the load changes. Moreover, ANFIS controller obtains 0.27 s for rise time according to 3000 rpm of speed reference, while the other controllers have longer time to reach the speed reference.