A Robust DPCC with Super-Twisting Sliding Mode Observer and Parameter Identification for IPMSM

To solve the problems of performance degradation caused by parameter mismatch in deadbeat predictive current control(DPCC) under complex operation conditions, an interior permanent magnet synchronous motor(IPMSM) robust DPCC method based on disturbance sliding-mode observer is proposed, which can estimate and compensate voltage error due to parameter mismatch. In order to solve chattering problem of SMO, an improved SMO based on super-twisting algorithm(STA) is designed, which can further improve steady-state performance, and reduce harmonics. Furthermore, Adaline neural network is used to design parameter estimator which can extract parameter error information contained in voltage error. Finally, the effectiveness of proposed STA-SMO based DPCC and Adaline NN is verified on an IPMSM by simulation.