Brushless Dc Motor Control Strategy Based On Neural Network Research

In recent years, advances in rare-earth permanent magnetic material, power devices, micro-processor, converter idesign technique and control theory have made BLDCM play a vital role in motion-control applications. However, the control performance of the BLDCM drive is still influence by uncertainties, which usually features parameter variations, external load disturbances and nonlinear dynamics. To achieve high-performance BLDCM drive, which has great ability of adaptation and better performance against disturbances, advanced control schemes have to be developed to deal with these uncertainties.This thesis, on the basis of the research on structure and learning algorithm of neural network, a high-performahce neuro-controller with simultaneous online identification and control is proposed for controlling BLDCM. The dynamics of the motor are controlled using two different neural based identification and control schemes, as system is in operation. In the first .scheme, an attempt is made to control the rotor angular speed, utilizing a single three-hidjden-layer network, the second scheme attempts to control the stator currents. This schemes incorporates three multilayered feedforward neural networks that are online trained, using the Levenburg -Marquardt training algorithm. The control of the direct and quadrature components of the stator current can successfully traced trajectories after relatively short online training periods. The control strategy adapts to the uncertainties of the motor dynamics and their inherent nonlinearities. Promising simulation results have been observed when the neural controller is trained in an environment contaminated with noise.