Research On Torque Control Of Induction Motor EPS System Based On Adrc

Electric power steering(EPS)system has become the most widely used steering system due to its advantages of good operation stability,easily modular design and installation,and personalized power-assisted modes.Meanwhile,it accords with the automotive future directions,which is low-carbon,lightweight,intelligentThis paper treated the EPS system equipped with induction motor as the research object,studying on the control algorithm of induction motor and EPS system under variable load,to achieve the accurate dynamic torque control and flexible steering control.As a torque servo system,EPS system is required to respond quickly and accurately to the steering torque of the steering wheel.The special structure of induction motor makes it impossible to realize high performance torque control by simple algorithm like DC motor.Therefore,on the basis of vector control for Stator Field Orientation(SFO)and considering the running efficiency of the motor,this paper discusses parameters identification,stator flux observation and the regulation of two axis currents,and designs the EPS system control strategies for power-assistant control,returnable control and compensation control.In vector control of induction motor,motor parameters should be used to estimate the position of flux linkage.The mathematical model of SFO is obtained by vector transformation.Since the SFO does not involve the time-varying parameters of the rotor side,the conventional off-line parameter identification method is adopted.According to the off-line identification principle of different parameters under special working conditions(lock-rotor,no-load),the error term of the parameter identification system is analyzed.The error caused by the current distortion because of the power switch element dead-zone time is analyzed emphatically.The compensation scheme is designed with the coefficient is determined Then the parameter range of the selected motor is obtained by induction motor off-line parameter identification experiment system.In order to obtain accurate motor parameters,particle swarm optimization algorithm based on chaotic sequence is used to optimize the experimental parameters.The fitness function between the current detection value and the calculated value of the system is defined.The parameter optimization model is verified by simulation and the final identification result is obtained.Compare the identification results before and after optimization by the motor running state experiment,so as to verify the motor parameters identified by the optimization model are closer to the actual values.The coupling problem of stator flux linkage and torque current exists in SFO control.Aiming at the deficiency of the conventional stator current decoupling compensation algorithm under the stochastic load of EPS system,a stator flux closed-loop decoupling control strategy based on the Active Disturbance Rejection Control(ADRC)is proposed.For the Linear Extended State Observer(LESO)error arising from larger disturbance of the system changes,this paper uses Parallel LESO,to observe the error and compensate it.The Improved ADRC decoupling control scheme of the stator flux is proposed.Its anti-jamming performance and response performance were better than that of the conventional scheme,this is validated by computer simulation.Based on the analysis of integral saturation and DC bias of common flux observers,and considering the optimization of motor efficiency in EPS system,this paper proposes to add ADRC to the traditional fixed threshold double integral flux observer,so as to realize the non-deviation and non-saturation observation of dynamic flux.According to the working characteristics of EPS system,the necessity of efficiency optimization of induction motor is discussed.By using the loss model method,the stator flux amplitude model which is related to the load torque and speed is established.By integrating the motor parameters in the model,and using the recursive least square method with forgetting factor to estimate,the online identification of time-varying parameters is avoided.According to the stability requirement of d-axis current,the ADRC current regulator is designed.According to the response characteristic of q-axis current,a fuzzy adaptive PID current regulator is designed,and the performances of the two-axis current regulator are verified by simulation.Reasonable static and dynamic control strategy is a necessary condition for EPS system to achieve steering light and comfortable,so as to have a good sense of operation.This paper analyzes the torque characteristics and torque state of torsion bar.Replacing the steering wheel torque with sensor measurement torque,a new power characteristic curve is established.Aiming at the problem of insufficient mechanical returnability,a returnable control algorithm based on steering wheel angle and angular velocity was established.A TD-based steering wheel angular velocity estimation scheme is proposed,and its anti-jamming performance is better than that of the traditional differential algorithm verified by simulation.A load torque estimation scheme based on ESO is proposed for the deterioration of steering performance caused by abrupt load,and a torque compensation algorithm is established.In order to debug the control algorithm of induction motor and steering system proposed in this paper,the ECU of induction motor EPS system was developed,and a test bench of EPS system was built according to the requirements of steering performance test,to verify the basic power-assistant control,returnable control and compensation control of the ECU.