Research On Control System Of High Speed Permanent Magnet Synchronous Motor For Spinning Based On Neural Network

Considering the low speed and low performance of textile machinery and equipment,it is important to introduce advanced control algorithms to read high-speed and high-capacity textile equipment.High speed permanent magnetic synchronous motors are gradually preferred by the textile industry due to their small size,light weight and energy saving.However,high-speed permanent magnetic synchronous motors have shortcomings such as non-linear and multiple control variables,which during operation are easily affected by internal parameters(magnetic field,temperature),external environment(load,speed change)and other factors.Usually,PMSM adopts traditional PI control.This method has low strength and cannot meet high performance control requirements in the use of PMSM systems.Based on this,the PP neural network is integrated into the PI control mechanism.The fusion of the two mechanisms effectively solves the motor control problem.Specific experiments demonstrate that the integration of the PP neural network improves the robustness of the control system and that the control effect is excellent.The main research content of this article is as follows:First,in this paper,we analyze the theoretical state of domestic and international motor control along with motor characteristic selection,compare and discuss,and this time use vector control(FOC)as a high-speed motor control strategy.The article describes in detail and how spatial vector pulse width modulation(SVPWM)works.The calculation method is finally a simulation model of a high-speed motor is designed in the simulation software.Second,This paper analyses the principle of PID control algorithms and BP algorithms aimed at the problems of the traditional PI control system such as mild interference and instability and,given BP’s self-adjusting algorithms,integrates the neural network algorithm into the PI control system.Finally,an online control system suitable for the characteristics of the engine has been developed and the robustness of the control system improved.Then,in view of the difficulties in the general optimization of the BP neural network,the selection of nonlinear functions and the selection of hyperparameters,optimization schemes,respectively,were proposed and a high-speed motor control system with better performance was proposed.designed.which improved the real-time performance of the control system and enhanced the robustness of the control system.Finally,the PI closed-loop control system and the enhanced closed-loop control model of the enhanced BP neural network were built into the simulation software for simulation comparison experiments.double PID closed loop control of the network The effect is more stable.Finally,based on the above theory,it designed and built the STM32F407 control board as the central motor control experiment platform,focusing on the hardware circuit for the software design of the whole system,completed the main function,interrupt and subroutines of the corresponding function module.In addition,he wrote the software for obtaining the engine speed and collected the speed information in the software for obtaining the engine speed.By the experimental comparison of the traditional single and double closed loop PI and the improved BP neural network combined with the online speed control system of the PID controller,it was found that the combination of an improved BP neural network compared to traditional closed circuit Single and double PI,the PID controller has greater robustness,better real-time performance and shorter response time.