PI fuzzy gain-scheduling speed control of a gas turbine power plant

Speed control during startup is one of the most critical tasks of gas turbine power plant operation. This paper introduces the PI fuzzy gain-scheduling (PI-FGS) controller to solve the foremost speed control problems, including tracking of the acceleration pattern and rejection of disturbances caused by operation events throughout startup. Fundamentally, the PI-FGS synthesizes a gain-scheduling controller from multiple locally-tuned generalized PI algorithms by means of a fuzzy system. The PI-FGS structure permits independent tuning of the tracking and rejection responses at any operating point. In addition, tuning can be carried out on-site based on the operator experience and inspection of the current plant response. Simulation experiments show that the PI-FGS may improve speed control performance well over that provided by commercially available speed controllers, which are based on a single conventional PI algorithm