Font Size: a A A

Research On Direct Torque Control Strategy Of Switched Reluctance Motor Based On Neural Network

Posted on:2023-03-28Degree:MasterType:Thesis
Country:ChinaCandidate:M SunFull Text:PDF
GTID:2532306782962809Subject:Control Engineering
Abstract/Summary:
Switched Reluctance Motor(SRM)has been widely used in electric vehicles,aerospace,mining,textile industry and other fields because of its simple structure,low cost and high efficiency.However,due to the double salient pole structure of the stator and the rotor,it has serious nonlinearity,resulting in an inevitable large torque ripple,and the radial force between the stator and the rotor is constantly changing with the relative position of the stator and the rotor.If the change occurs,the stator and the casing of the machine body will vibrate violently in severe cases,which will lead to the occurrence of high motor noise.This defect has seriously hindered the promotion-of switched reluctance motors in some specific applications.Compared with other motors,the traditional current-voltage chopper control is not ideal for SRM torque ripple suppression.Therefore,this thesis adopts the Direct Torque Control(DTC)strategy and study its effect on SRM torque ripple suppression.This study firstly introduces the research status and main research hotspots of switched reluctance motor.Expounds the SRM mechanism and its operation principle.And analyzes the SRM linear model,quasi-linear model and nonlinear model.Then analyzes the current chopping control,voltage PWM.Angle control strategy these three basic control strategies.Through the analysis of the causes of torque ripple,it can be seen that the above traditional control strategy is not ideal for the suppression of torque ripple.The effect of suppressing torque ripple is verified.Then this study combined with the improved sector direct torque control,the fuzzy PI speed controller and the BP neural network PID controller are designed respectively for application and comparative analysis in the direct torque control.For the three-phase 6/4-pole SRM,a DTC simulation model was built on the MATLAB/Simulink platform,and the working principle of each module was analyzed in detail.Combined with the results of the system simulation experiment,it is verified that the BP neural network PID controller suppresses the torque ripple.effectiveness and feasibility.Finally,based on the theoretical analysis and simulation results of the SRM control system,using STM32F303RCT6 as the main control chip,the controller including the current and voltage detection circuit,power supply circuit,drive circuit,position detection circuit and other hardware circuits is designed,and the corresponding software is given.Subroutine design flow,the SRM experimental platform is built for verification.The experimental results show that the direct torque control strategy proposed in this thesis can effectively suppress the torque ripple of the SR motor and make the SRM speed control system run stably.
Keywords/Search Tags:Switched reluctance motor, direct torque control, improved DTC, BP neural network
Related items