Research On Motor Control System For Electric Scooter Based On BP Neural Network

Aiming at the problems of safety and comfort in using electric scooter,it is of great practical significance to study deeply the electric scooter which is safe,reliable,cost-effective and comfortable to drive by using modern control technology.Brushless DC motor has become the preferred motor for electric bicycle because of its high efficiency,long life,low noise and good mechanical performance.However,brushless DC motor is a multi-variable,strong coupling,non-linear complex system,and conventional PID control algorithm is difficult to achieve the desired speed control effect.In this paper,BP neural network with strong learning ability,adaptive ability and non-linear approximation ability is introduced into the control of electric scooter.BP neural network PID controller is designed,and the simulation model of BLDCM speed control system is built.The results show that the BP neural network PID control has better stability and robustness than the conventional PID control.The main contents of this paper are as follows:(1)In order to meet the requirements of driving safety and comfort of electric scooter,three aspects of software program algorithm of electric scooter are studied.Firstly,the differential steering control method of electric scooter is adopted,which solves the problems of side slip or side slip when electric scooter steers,and ensures the driver's driving safety.Secondly,considering the slow reaction of middle-aged and old people when driving electric scooter,a intelligent speed control method of electric scooter is proposed,that is,intelligent speed control on uphill,downhill,backward and steering driving.Finally,a double-closed-loop PID control method for electric scooter motor is proposed,which optimizes the speed regulation performance of the motor and achieves a more comfortable driving experience.(2)Aiming at the problems of low control precision and poor anti-interference ability of conventional PID control in motor speed regulation system,BP neural network algorithm is introduced.In order to solve the problems of local minimum and slow convergence of BP neural network control algorithm,a BP neural network control algorithm based on online adjustment of learning rate is proposed,and a BP neural network PID controller of speed loop is designed.The system simulation results show that the improved BP neural network PID control algorithm makes the electric scooter control system have better dynamic performance and steady-state accuracy.(3)Based on the above control algorithm and the functional requirements of electric scooter,the hardware and software of electric scooter motor control system are designed,including the hardware composition,the selection of main components,the design of control system module circuit,software framework,main program and software design of each module.The experimental platform of the control system of the electric scooter is built,and the hardware test of the control system and the road test of the electric scooter are carried out.The test results prove the correctness of the hardware and software of the electric scooter.