Probability-based robust optimal PI control for shell gasifier in IGCC power plants

Abstract

In order to achieve a robust control performance of the gasifier which has dynamic characteristics of multi-variable coupling, large inertia and multi-disturbance, an optimization method for decentralized PID/PI controller parameters based on probabilistic robustness is developed. First, the control structure and target of the Shell gasifier is analyzed and a crude model is introduced. Model uncertainties and other detailed industrial requirements could be considered at the same time in the method. The probability of satisfaction with the dynamic performance is computed statistically, and then it is presented as the objective function to optimize the controller parameters based on genetic algorithm. The Monte Carlo experiment was applied to test the robustness of the control system. In comparison with the tuning methods based on internal model control (IMC) and the optimization algorithm under the nominal model, simulation results show the method could exploit the potentialities of PID/PI controllers in a maximal probability.