Parameter Optimization of Robust Non-fragile Fractional Order PID Controller for Pump Turbine Governing System

Due to the reversible properties of pump turbine, its operation along the "S" shape curve can lead to unstable operation, like remarkable unit frequency oscillation, under unload condition. To solve the problem, a robust non-fragile fractional order PID controller (FOPID) of pump turbine governing system is proposed in this paper. Furthermore, a combination of bacterial foraging algorithm (BFA) and multi-scenario functions are employed to select parameters for the novel FOPID controller, in which the multi-scenario functions are composed by the time domain integral performance indexes and the overshoot of system output. Simulation tests under unload running condition are used to demonstrate the feasibility and robustness of the proposed controller. The analysis results indicate that the proposed FOPID controller using BFA represents a sound improvement over the original method, and is highly practical.