Research On Brushless DC Motor Control System Based On BP Neural Network

With its unique advantages in torque,accuracy and efficiency,brushless DC motors have been widely used in various fields such as aerospace,biomedicine,household appliances,and automotive electronics.Because of the nonlinearity of the brushless DC motor,it is difficult to explain it with an accurate system model.In the final analysis,the change of its motor parameters is easily affected by external environment such as magnetic circuit saturation,which also includes factors such as load interference.PID control is widely adopted in the industrial field by virtue of its algorithm’s simplicity and super robustness.It is also recognized by everyone in terms of the reliability of the control effect.However,in PID control,the relationship between the proportional term,the integral term,and the differential term is not a simple linear function that can be described.The relationship between the three can be said to be complementary to each other.Only by clearly understanding the precise model of the system can an accurate control model be established.The brushless DC motor is doomed to the non-linearity of its motor system due to its multi-variable and strong coupling characteristics.If you want to find the best PID control parameters in its complex nonlinear system model,the possibility can be said to be minimal.Therefore,the research on high-efficiency control of brushless DC motors has become a new hot spot.Artificial neural networks have the ability to fit arbitrary functions without relying on the specific system model of the control object,including non-linear functions of course.The BP neural network is the most frequently used artificial neural network,and its effectiveness is also the highest.However,it is obviously unrealistic to realize the high-efficiency control of the brushless DC motor by only relying on a single control method.Therefore,a control scheme that combines BP neural network and PID control is proposed,and on this basis,a lot of improvements have been made to the neural network algorithm.The ultimate goal is to realize the improved BP neural network algorithm for PID control.Online self-correction of speed control parameters.Through in-depth analysis of the mathematical model and control requirements of the brushless DC motor,the corresponding control system simulation model is designed and built in MATLAB,and the control system adopts double closed-loop control.The experimental results show that compared with the traditional PID control,it obviously has certain control advantages.The improved BP neural network PID control speed control used in the system has the fastest response speed and the smallest fluctuation.Different from the unimproved neural network PID control,the adaptive setting time of the PID control parameters has been effectively improved.Through the analysis of its three-phase current,back electromotive force and electromagnetic torque,it is found that the control scheme not only improves the operation efficiency of the motor,but also greatly reduces the energy loss during the operation of the motor.It not only improves the intelligence of related products,but also achieves significant results in reliability and maintainability.Committed to achieving the goals of energy saving,emission reduction and operating cost control for small,medium and micro enterprises,and to provide a driving force for the industrial upgrading of the motor industry.