Robust Position Sensorless Control Stratgy Of High-Frequency Signal Injection For Pmsm Low-Speed Operation

Permanent magnet synchronous motors(PMSMs)have significant advantages such as high power density,high operating efficiency,and have been widely used in intelligent home appliances,transportation electrification,and aerospace.In many applications,it is difficult to install the position sensor due to the compact space inside the motor or the complex and harsh operating environment,so it is of great theoretical significance and application value to study the position sensorless control technology of PMSM.At present,one of the technical challenges of sensorless control systems for PMSM is to improve the performance in the low-speed opreation area,and the high-frequency(HF)signal injection method is an effective technical route to achieve low-speed operation control.The following key problems exist: the additional noise caused by the high-frequency injection method affects the operational adaptability of the system,the single-resistance sampling method in the low-speed non-reconfiguration region leads to system control performance degradation,and the immunity of the system under complex working conditions needs to be further improved,etc.Therefore,the robust control strategy of HF signal injection method needs to be further studied,which is of great significance for the promotion and application of high-performance PMSM sensorless drive system.The robust position sensorless control stratgy of high-frequency signal injection in the low-speed operation is explored for the PMSM drive for air-conditioners.In order to solve the problem of harsh noise in the traditional fixed-frequency HF signal injection method,a random-initial-phase HF square-wave voltage signal injection based sensorless control method is proposed.Based on the idea of random pulse width modulation,by building the probability selection mechanism,several HF square-wave voltage signals with different initial phases are randomly injected into the windings of PMSM to expand the high-frequency distribution range of the energy spectrum and reduce the additional HF noise caused by the induced current.A rotor position extraction scheme based on the increment of induced HF current information is studied to meet the special requirements of signal processing in the proposed method.Based on the principle of power spectral density,the difference of HF noise producted by the traditional fixed-frequency HF signal injection method and the proposed random HF voltage signal injection method is analyzed,which elaborates the mechanism of expanding HF energy spectrum by the proposed method theoretically.The experimental results show that this method can ensure the sensorless low-speed stable operation and reduce the additional noise caused by HF signal injection.Vector phase-shift modulation strategy is an effective method to solve the problem of non reconstruction area in low speed operation of low-cost PMSM drive systems with single resistance sampling.However,due to the time-sharing sampling characteristics of digital control system,there is obvious current reconstruction error in low speed operation area.For a sensorless permanent magnet synchronous drive system with single resistance sampling,this paper proposes a phase current reconstruction error suppression method based on the estimation of the local average current change rate(ELAC2R)during the switching period.Because the current errors in different sectors vary greatly,the current ripple and reconstruction errors caused by the voltage vector phase-shift single current sensor sampling system in the low speed region are modeled and analyzed by comparing the actual sampling points with the ideal sampling points.Furthermore,according to the current ripple characteristics between the ideal sampling point and the actual sampling point,more accurate motor phase current information can be obtained quickly through the sampling value.This method does not need to change the circuit topology and does not increase the complexity of the modulation strategy.Compared with the traditional method,the proposed method does not need to inverse the inductance matrix containing the rotor position information when obtaining the current change rate,and the phase current sampling error compensation process is simplified,reducing the computational burden of the control chip.The experimental results show that the proposed method can effectively improve the phase current reconstruction performance of PMSM,and reduce the harmonic content of motor current.Active disturbance rejection control(ADRC)theory has a special advantage in improving the robustness of the sensorless PMSM control system.However,the traditional linear and nonlinear ADRC strategies have their own limitations in suppressing different forms of disturbances under complex conditions.To further expand the ability of ADRC strategies to suppress the disturbances and improve the robustness of the sensorless PMSM control system,a cascaded ADRC strategy based on nonlinear efal functions with variable gains is proposed to improve the control performance of the motor speed loop.The proposed nonlinear error description function(efal function)can combine the advantages of the linear function and the exponential function,which can effectively suppress the position observer error and improve the observation effect of the disturbances in the ADRC drive system.On this basis,cascaded disturbance observation is adopted to ensure fast estimation and compensation of the disturbances,and the improvement effect of the proposed method on system robustness is demonstrated from the perspective of description function,amplitude frequency characteristics,disturbance compensation performance and system stiffness.The experimental results show that the proposed method can effectively suppress the disturbances of the sensorless PMSM control system,reduce the overshoot and response time of the speed loop,and improve the dynamic response capability of the system.Based on the above research,the proposed algorithm is implemented in a singleresistance sampling PMSM driver based on Renesas control chip,and the rotor position observer with random HF square wave voltage signal injection,the motor stator current reconstruction strategy based on current rate of change estimation,and the ADRC strategy based on cascaded extended state observer are integrated.The effectiveness of the proposed method is verified through the experimental platform.The developed driver is also studied in the engineering application of air conditioner system,and the test in the working chamber shows that the proposed PMSM position sensorless control method has good performance.