Analysis And Strategy Of Position Estimation Error Of Permanent Magnet Synchronous Motor Sensorless Control

Permanent magnet synchronous motors(PMSMs)are widely used in aerospace,deep sea,deep earth and other high-tech fields,air conditioning,washing machines and other household occasions because of its simple structure,high efficiency and high power factor.Sensorless control technology is the research hotspot of the PMSM control because it can reduce the volume and the cost of the PMSM drive system,and improve the reliability of the drive system.The inverter nonlinearity,current measurement errors and PMSM model parameter errors can increase the position estimation error and reduce the performance of the PMSM sensorless control.The inter-turn short-circuit(ITSC)fault increases the current and the temperature of the short-circuit coil,speeds up the damage of winding insulation,increases the position estimation error,and affects the performance of voltage compensation and parameter identification.It is important for improving the performance and reliability of the sensorless control to compensate inverter nonlinearity and current measurement errors,identify PMSM model parameters,diagnose the ITSC fault and reduce the position estimation error.Aiming at the chattering problem of the sliding mode observer(SMO)and the phase delay of the position estimation algorithm,an adaptive SMO in the synchronous rotating coordinate system(??-axis)is proposed.The gain of the SMO switching function is reduced and the phase compensation block is omitted by introducing adaptive parameters to the ??-axis SMO.The experimental results verify that the algorithm can effectively reduce the high frequency error and phase delay of the estimated position.The parameter model of voltage source inverter(VSI)nonlinearity such as deadtime effect is established,and the expression of dq-axis voltage errors caused by VSI nonlinearity are derived.Considering the influence of VSI nonlinearity,the small signal model of the ??-axis adaptive SMO based position estimation algorithm is established.The transfer function of the d-axis back electromagnetic force(EMF)estimation and the d-axis voltage error is derived.And the transfer function of the position estimation error and the d-axis voltage error is derived.Aiming at the declined accuracy of VSI nonlinearity identification method caused by the position estimation error,motor parameter error and observer delay,an identification and compensation strategy based on the d-axis back EMF estimation is proposed.The experimental results show that the identification and compensation strategy can compensate the VSI nonlinearity,reduce the position estimation error and improve the performance of the PMSM sensorless control.Considering the DC offsets and the gain coefficient errors of phase currents,the mathematical model of current measurement errors is established,and the expression of the equivalent d-axis voltage error of the effect of current measurement errors and VSI nonlinearity on ??-axis SMO-based position estimation algorithm is derived.Aiming at the interference problem caused by the current measurement errors with the VSI nonlinearity compensation method,a look-up table-based compensation strategy is proposed to compensate the d-axis voltage error.In order to realize the compensation of current errors and VSI nonlinearity,the d-axis voltage compensation signal is updated by the iterative learning algorithm(ILC)to minimize the d-axis back EMF harmonics.The experimental results verify that the algotithm can effectively reduce the position,velocity estimation error and phase current harmonics.The steady-state voltage equation of the PMSM considering VSI nonlinearity,current measurement errors,motor parameter error and position estimation error is established,and the influence of the position estimation error on parameter identification is analyzed.Aiming at the rank defect problem caused by the position estimation error,a nonlinear regression model of the PMSM is established,and the identification and correction strategy based on the Gauss-Newton iterative method is proposed to estimate VSI nonlinearity and motor parameters.The experimental results verify that the proposed strategy can correct the model parameters of the position estimation algorithm and effectively reduce the position estimation error.The mathematical model of the PMSM with the ITSC fault is established,the expression of the equivalent d-axis voltage error of the effect of ITSC fault,VSI nonlinearity,current measurement errors and motor parameter error on position estimation is derived,and the amplitude and phase characteristics of the harmonics of the d-axis voltage error are analyzed.Aiming at the problem of mutual interference between diagnosis of the ITSC fault and elimination of the current error,a fault diagnosis strategy based on the d-axis voltage compensation signal is proposed.The experimental results verify that the fault diagnosis strategy can eliminate the current relative gain error,locate the ITSC fault and effectively improve the reliability of the PMSM sensorless control.