State Parameters Estimation for Distributed Drive Electric Vehicle Based on PMSMs’ Sensorless Control

To improve the accuracy of the state parameters estimation for distributed drive electric vehicles (DDEVs), a method for estimating the state parameters of DDEV based on permanent magnet synchronous motors’ (PMSMs) sensorless control is proposed in this article. First, aiming at the problem that the speed and rotor position of the PMSMs are difficult to obtain, a sensorless control method for PMSM based on the square root seventh-degree cubature quadrature Kalman filter with robust factor (RF-SR7thCQKF) algorithm is designed. The algorithm combines the seventh-order orthogonal sphere-radius criterion with Bayesian filtering to obtain the seventh-degree cubature quadrature Kalman filter (7thCQKF), which can obtain higher estimation accuracy than the third- and fifth-order criteria by increasing the number of sampling points. To further improve the stability and filtering accuracy, the orthogonal triangle (QR) decomposition is introduced of the 7thCQKF to ensure the semi-positive characterization of the covariance matrix, and the effect of the filtering accuracy degradation due to the large measurement value error is weakened by the introduction of the robust factor. Second, the hyperbolic tangent adaptive sliding mode observer (HT-ASMO) for tire forces observation is designed, and the HT-ASMO can attenuate the jitter frequency of the system, reduce the time to reach the sliding mode switching surface from any point, and effectively improve the estimation accuracy of the tire forces. Finally, the PMSMs with sensorless control are used as the drive motors of the DDEV, and the RF-SR7thCQKF algorithm is used to estimate the vehicle state parameters based on the four-degree-of-freedom vehicle dynamics model. After simulation and experimental verification, the vehicle state parameters estimation method based on the PMSMs’ sensorless control has a high estimation accuracy.