Neural network adaptive observer based position and velocity sensorless control of PMSM

Neural network has been recognized to be able to offer a number of potential benefits for application in the field of drive. This paper presents a position and velocity sensorless control algorithm for a high performance permanent magnet synchronous motor (PMSM) based on direct neural adaptive observer. The proposed observer comprises two neural networks which are trained to learn electrical and mechanical models respectively. Adaptation is realized by online training using current prediction error. Various advantages of this estimating scheme over other sensorless control scheme, such as robustness, nonlinear adaptation and learning ability is shown by extensive simulations.