A Robust Self-Organizing Neural Fuzzy Controller for Dynamic Braking Resistor

摘要

A robust self-organizing neural fuzzy controller (RSONFC) is designed for dynamic braking resistor of power system. First, a neural network (NN) based model is introduced to implement the self-organizing fuzzy controller. The control rules and membership functions of fuzzy system are represented as the processing nodes in the NN model. Then, the traditional heuristic fuzzy control rules and input/output fuzzy membership functions can be optimal tuned from training examples due to excellent learning capability of NN. For the robustness considered, a multilayer neural network is used to learn the relations of operation conditions and optimal parameters of the controller. To demonstrate the effectiveness of the proposed controller, comparative studies with a variable structure controller (VSC) are conducted on a single-machine-infinite-bus power system with rather encouraging results.