Research On Shift Displacement Estimation And Control Of Electromagnetic Direct Drive Transmission System

New energy vehicles have put forward higher requirements for vehicle control execution technology,especially the underlying drive device.High-efficiency,energy-saving and high-quality direct drive technology has become the focus of research in the field of drive.Among them,electromagnetic linear actuators have significant advantages in terms of quality,volume,response and other performance compared with the combination of rotary motors and motion conversion devices in the field of linear motion control.Electromagnetic linear actuators face some common basic problems in the process of marketization:the use of expensive displacement sensors directly affects the economy,market competitiveness and reliability of the actuators in complex environments.The problem of actuator displacement estimation and control is a common basic problem that scholars at home and abroad are most eager to solve as soon as possible,and it is also a technical bottleneck that needs to be broken through as soon as possible in the marketization of electromagnetic linear direct drive devices.In this paper,the Direct-Driving Automated Mechanical Transmission（DAMT）based on electromagnetic linear actuator is studied to address the problems of decreasing compactness,increasing cost and decreasing reliability due to the presence of displacement sensors.A Least Square Support Vector Machine（LSSVM）based displacement estimation method is developed to realize the sensorless control of electromagnetic actuator and DAMT by combining theoretical analysis,mathematical modeling,simulation and experimental verification.The specific work includes the following aspects.（1）The analysis and modeling of the electromagnetic direct-drive gearshift system based on the electromagnetic linear actuator were carried out.Firstly,the structure and principle of electromagnetic direct-drive transmission system were analyzed;the structure and principle of the actuator were analyzed and the mathematical model was established;the mathematical model of the shift force was established based on the analysis of the shift process,and the shift control strategy based on the shift process was proposed.（2）A displacement estimation method applied to electromagnetic linear actuators was studied.A displacement estimation method based on LSSVM was proposed to overcome the disadvantages of existing displacement estimation methods such as low accuracy,poor stability,weak generalization ability and the need for a large number of training samples,and to achieve a high estimation accuracy with a small training sample.The multi-objective particle swarm optimization algorithm was used to optimize the regularization parameters and kernel width parameters based on the multi-objective particle swarm optimization algorithm with accuracy and root mean square error as the optimization objectives,which further improved the performance of the displacement estimation model.In addition,the correlation analysis between the evaluation indexes and the performance parameters of the displacement estimation model was carried out to clarify the influence law of the performance parameters on the performance of the displacement estimation model,and thus an improved full-service displacement estimation model applicable to the full-service conditions was established.The rules for collecting training data of the displacement estimation model were developed,and the data were collected and simulated.The results showed that the estimation accuracy was above95%,the root mean square error was below 1×10^-4,and the estimation time was below 0.1ms under the specific working conditions.（3）A displacement sensor-free control method for electromagnetic linear actuators was studied.The control system controller of the electromagnetic linear actuator was designed,and the control system with PID control algorithm,fuzzy control algorithm and fuzzy adaptive PID control algorithm were designed and analyzed respectively;the displacement sensor-free control of the linear actuator is realized by combining with the displacement estimation model based on LSSVM.The results show that the accuracy of the control estimation was above 94%;the dynamic test rig of the electromagnetic linear actuator was built,the test process was designed,the data was collected by the test rig and the displacement was estimated,and the results showed that the accuracy of the estimation was above 92%,thus verifying the feasibility of the displacement estimation model in practical applications.（4）The displacement sensor-free control method of DAMT was studied.Based on the modeling analysis of the shifting process,the control model of DAMT was established and the evaluation system of the shifting quality was built;the data was collected from the control model of DAMT,the training set and test set were established and the model was trained;the displacement estimation model was applied to the DAMT and the sensor-free control of DAMT was realized;the simulation verification of the in-,out-and shifting of DAMT under typical working conditions was carried out.The results show that the accuracy of the estimation was over 94%.A general test rig for the electromagnetic direct-drive transmission system was built,the test procedure was designed,the data was collected and tested by the test rig,and the shift displacement was estimated by the displacement estimation model based on LSSVM,and the results showed that the accuracy was over 93%.