A fuzzy logic approach to accommodate thermal stress and improve the start-up phase in combined cycle power plants

Abstract

Use of combined cycle power generation plant has increased dramatically over the last decade. A supervisory control approach based on a dynamic model is developed, which makes use of proportional-integral-derivative (PID), fuzzy logic and fuzzy PID schemes. The aim is to minimize the steam turbine plant start-up time, without violating maximum thermal stress limits. An existing start-up schedule provides the benchmark by which the performance of candidate controllers is assessed. Improvements regarding possible reduced start-up times and satisfaction of maximum thermal stress restrictions have been realized using the proposed control scheme.