An Adaptive Neural-Network Model-Following Speed Control of PMSM Drives for Electric Vehicle Applications

A robust speed control technique for permanent-magnet synchronous motor (PMSM) drives is proposed in this paper for electric vehicle applications. The robust controller consists of a neural-network controller (NNC) in the speed feed-back loop in addition to an on-line trained neural-network model-following controller (NNMFC) in the feedforward loop. The adaptive neural-network model-following controller (ANNMFC) combines the merits of the feed-back NNC and the feed-forward NNMFC for PMSM drive. The weights of the NNMFC are trained on-line according to the model-following error between the outputs of the reference model and the PMSM drive system to realize high dynamic performance in disturbance rejection and tracking characteristics. The NNMFC generates an adaptive control signal which is added to the feed-back neural-network speed controller output to attain robust model-following characteristics under different operating conditions regardless of parameter uncertainties and load disturbances. A computer simulation is developed to demonstrate the effectiveness of the proposed robust ANNMFC. The results confirm that the proposed robust speed controller grants robust performance and precise tracking response to the reference model regardless of load disturbances and PMSM parameter uncertainties using the adaptive neural-network model-following control. Key-Words: PMSM, Vector Control, Neural Network (NN), Model Following Controller (MFC).