Predictive control of superheated steam temperature based on decentralized fuzzy inference

A predictive control scheme based on decentralized fuzzy inference is developed for the control of superheater steam temperature. Firstly, the future output of the control object is predicted by the object's step response model. Then a decentralized fuzzy inference structure is designed. The deviations between the predicted and expected values in the future instant are regarded as inputs of fuzzy inference units to perform the fuzzy inference process and acquire the inference results. The inference results then are weighted and synthesized and the control output of the system is generated. Simulation results for the superheated steam temperature are presented to show the desired performance of the developed approach.