Paper
Study of RBF Neural Network Speed Controller Based on the Fuzzy PI in DTC System
- Authors:
- Kai Xu; Guowei Xu; Ping Qu
- Abstract
- The role of proportional-integral (PI) controller as a speed controller for induction motor in high performance drive system is vital. However, the PI controller is slow in adapting to speed changes, load disturbances and parameters variations without continuous tuning of its gains. Hence, an on-line self- tuning scheme using Fuzzy logic controller was proposed. On the basis of conventional PI regulator, a Fuzzy PI speed controller was designed according to speed error and its rate of change, which could adjust the proportional coefficient kp and integral coefficient ki dynamically to adapt the speed variations. Simultaneously, a new strategy of RBF neural network which replaces the Fuzzy inference is also proposed and applied to DTC system. It is unnecessary to initiate complex search and inference, and may have the output results through the high speed parallel distribution computation. The comparison with conventional PI speed controller shows that the proposed method reduces the flux, speed, current and torque ripples. The validity of the proposed method is verified by the simulation results.
- Keywords
- Direct Torque Control(DTC); Induction Motor; PI Speed Controller; Fuzzy Inference; RBF Neural Network
- StartPage
- 21
- EndPage
- 25
- Doi