Online Adaptive Control Method Based on Dynamic Fuzzy Neural Network

Proposed a novel algorithm of dynamic fuzzy neural network structure based on the expansion of radial basis network, the outlet error and may hold the boundary the effective radius is determined whether the new rule should be joined into the fuzzy rules, and can reach specific function of the system. Its most prominent feature of learning algorithm parameters are adjusted and the structure of identification at the same time, and can be used for real-time modeling and control. Finally the actual case for the simulation analysis, simulation results show that the proposed algorithm is the quick pace of study, more stable, practicable and reliable.