A new approach to direct torque control of interior permanent magnet synchronous motor

This paper presents an approach to direct self-control of permanent magnet synchronous motor (PMSM). This method is based on neural networks. The neural network is used to select the switching state of the inverter. In order to reduce the training patterns and increase the execution speed of the training process, the inputs of switching table is converted to digital signals, i.e., one bit represent the flux error, one bit the torque error and three bits the region of stator flux. The proposed three-layer neural network controller contains only twelve neurons in an attempt to minimize computational requirements of the digital signal processor. Computer simulations of the permanent magnet synchronous motor and neural network system are presented. The simulations results show that it is possible to replace switching table of the direct self-control (DSC) for permanent magnet synchronous motor by a neural network controller and the system using this neural networks controller has quickly parallel speed and high torque response.