An Improving Method to Decoupling and Linearization of Induction Motor Based on Neural Network Inverse

Wang Xin,Dai Xianzhong
DOI: https://doi.org/10.3321/j.issn:1000-6753.2008.04.006
2008-01-01
Abstract:Due to parameters’ variation during operation of induction motor, the decoupling and linearization implemented by field oriented control (FOC) and analytical inverse control(ANIC) is destroyed. For that, based on current-fed induction motor model whose outputs are rotor flux magnitude and speed, a neural network inverse decoupling and linearization method is proposed, theoretically analysis shows that the method can implement the adaptive decoupling and linearization of induction motor, and make the couple between the rotor flux and speed weaker and the design of outer loop controller easier, so the whole system control performance is further improved. At last, taking induction motor system which adapted proposed method as object, the study of simulation is done. The comparison simulation results show that the proposed decoupling and linearization method is valid.
What problem does this paper attempt to address?