Interval Type-2 Fuzzy Neural Network Based Cascade Predictive Control of Superheated Steam Temperature

Abstract The superheated steam temperature (SST) suffers from complex dynamic characteristics such as nonlinearity and large time delay. To deal with the negative effects, this paper proposes a cascade predictive control strategy (CPC-PI) based on interval type-2 fuzzy neural network (IT2FNN) to improve the performance of SST. First, this paper proposes the IT2FNN model for modeling the SST model. And then on the basis the local linearized IT2FNN model, the CPC-PI strategy is designed. In the following, fuzzy self-regulation of the weight factor in the CPCPI is carried out to further improve the dynamic response speed and stability of SST. Finally, the comparative simulations verify that the IT2FNN-based CPC-PI control strategy with self-regulated weight factor is superior to the PI-PI control strategy and the IT2FNN-based CPC-PI control strategy with fixed weight factor.