| Summary: | 碩士 === 大同大學 === 電機工程學系(所) === 96 === This thesis focuses on using the self-organization-fuzzy-neural-network (SOFNN) control system to accomplish the periodic motion control of the linear induction motor (LIM) drive. The structure of the fuzzy-neural-network (FNN) is incorporated into the self-organization concept to form the SOFNN control system for alleviating the computation burden. Moreover, the adaptive laws for network parameters are derived in the sense of the Lyapunov stability theorem, and then the stability of the control system can be guaranteed under the occurrence of system uncertainties and external disturbance. The learning-rate parameters of the SOFNN have a significant effect on the network performance. Therefore, the control system employs the varied learning-rate parameters to increase the training effect. The convergence analyses of the output error are based on the discrete-type Lyapunov function to assure the convergence of the output error. Finally, there are three cases with the parameter variations and the time-varying external disturbances are applied in the simulations. The proposed SOFNN control is compared with the adaptive FNN control system. Besides, the effectiveness of the SOFNN control system is verified by the simulations.
|
|---|
