A Wide Adaptation Variable Step-size Adaline Neural Network Parameter Identification IPMSM Model Predictive Control Strategy

Abstract

Model predictive control (MPC), as a frequently adopted control strategy for permanent magnet synchronous motors (PMSMs), exhibits favorable dynamic response capabilities. However, it necessitates an accurate mathematical model of the controlled object, and any parameter mismatch can lead to a decline in control performance. This paper proposes a model predictive current control (MPCC) method based on parameter identification, which can be extended to the parameter identification of plug-in permanent magnet synchronous motors (IPMSMs). A wide-adaptability variable step-size algorithm is designed in response to the varying effects of single variable step-size functions on parameter convergence speed and ripple when the motor experiences different parameter disturbances. This method classifies and fits various variable step-size functions based on the maximum value of the absolute value of different instantaneous errors. This allows different variable step-size functions to adapt to different parameter disturbances, resulting in rapid waveform convergence and consistent ripple size in the identification process. Additionally, a new variable step-size function type was designed with simple parameter settings and easy debugging. Finally, the effectiveness of the proposed method was verified through experiments, and the results showed that the method can achieve fast and accurate identification of multiple parameters under different parameter perturbations, ensuring stable current control.