Research on MISO Fuzzy Neural Network and its Application

In this paper, a MISO fuzzy neural network algorithm is presented. This algorithm consists of the excellences of fuzzy algorithm and neural network algorithm. In the parameter learning phase it changes the parameters based on the Lyapunov stability theory to ensure the stability. Meanwhile, it didn't need to seek the whole minimum value when it modifies the parameters. So the algorithm can reach the stability result more quickly than the conventional fuzzy neural algorithm. The analyses of theory prove the stability of the algorithm. Then we use this algorithm to control the dissolved oxygen in wastewater treatment process, and compares with the conventional fuzzy neural algorithm. The results of simulations show the superiority of this algorithm and nicer robustness in the process.