Uncertain Robustness Optimization of Steam-Turbine Generator Set Control System

In order to meet the control specifications of thermal process with uncertainties, two uncertain robustness optimization methods are presented, utilizing stochastic and fuzzy theory. Fuzzy credibility robustness, based on newly developed fuzzy credibility theory, is a new concept other than probability robustness, which could provide more information. For the stochastic and fuzzy multi-objective programming problem respectively, NSGA-II algorithm combined with Monte-Carlo experiments are used to obtain the Pareto robustness solutions. Pessimistic value criterion is adopted to compare two uncertain variables. The above two methods are applied to a steam-turbine generator set control system optimization problem. Compared with those controller design methods based on nominal conditions, the results demonstrate the two new design methods have better robustness, and prove their feasibilities and advantages.