Research On Disturbance Suppression Method Of Permanent Magnet Synchronous Motor Servo System

With the advent of neodymium iron boron(REFEB)permanent magnet materials,permanent magnet synchronous motors have been widely developed in electric drive systems.However,since the permanent magnet synchronous motor system is a multivariable,strongly coupled nonlinear system,it is very difficult to directly control the motor.The permanent magnet synchronous motor can be decoupled through coordinate transformation,the motor model is simplified,and the control is similar to a DC motor.The method directly controls the decoupled current,which will greatly simplify the motor control system.The permanent magnet synchronous motor will generate a large number of periodic torque pulses in the torque output due to cogging torque,magnetic flux harmonics,and current measurement errors.At the same time,the time-varying nature of the load will adversely affect the system's variable speed regulation,which is inconsistent with the concept of high-precision tracking control.This article will suppress them by the following two methods.Permanent magnet synchronous motors often use PID control strategies,but the parameter setting usually uses human experience and has poor reliability.Repetitive control is based on the sampling parameters of the previous cycle to compensate the control signal of the next cycle,and to perform q-axis current compensation on the permanent magnet synchronous motor system under vector decoupling.As an additional control module,the solution is easy to implement and does not require accurate knowledge of the motor parameters.This paper proposes an approximate adaptive dynamic programming technique that iteratively solves the ARE equation by using the state and input of online information.All repeated iterations can be performed by repeatedly using the same state and input information at a certain fixed time interval.Under the condition that the system is stable,prove the convergence and make experimental verification.Aiming at the limitation of system information transmission and receiver,a sensorless system is adopted,and in order to improve the robustness of the system,a sliding mode variable structure control strategy is used.Due to the chattering problem caused by sliding mode variable structure control,this paper uses a new sigmoid function to replace the original sgn function,and continuous state function feedback control is used in the boundary layer to make the gain reduce jitter as much as possible while ensuring stability.Small jitter.At the same time,in order to reduce the reason that the signal to noise ratio of the signal can not be accurately estimated at zero speed,a compensation gain method is proposed to ensure that the gain value converges to zero at zero speed.In order to suppress the disturbance of the back electromotive force considering that the arc tangent function will amplify the error,the phase-locked loop technology is used to extract the signal,which makes the system signal analysis more accurate.