Speed Sensorless Control of Permanent Magnet Synchronous Motor Based on Wavelet Neural Networks

Lefu Yang
2010-01-01
Abstract:Permanent-magnet synchronous-motor is a muti-variable,non-linear,strong coupling system that is highly sensitive to the outer interfere and inter perturbation.To improve the system robustness,based on wavelet neural network time-frequency localization properties,zoom features,self-learning,adaptive,robust,and strong nonlinear approximation ability of the character and to establish non-linear mapping,it is estimated rotor position and calculate the output of the rotor speed,the introduction of sliding mode variable structure control strategy for PMSM control,and under the same conditions also PI control of a simulation study results showed that the proposed control strategy can effectively improve the system spped performance.
What problem does this paper attempt to address?