Adaptive Neuro-Fuzzy Modeling and Control of IP Drum Level of a Power Plant for Improving Transient Response

Abstract

Heat recovery steam generator (HRSG) boiler is one of the main components of combined cycle power plants that its proper and safe operation is subject to drum level being in a specified range. In this paper, an application of ANFIS structure is presented for modeling the dynamic behavior of IP drum level changes of Qom Combined Cycle Power Plant, with emphasis on accurate modeling of its transient behavior in order to improve the transient response and consequently prevent steam unit from tripping. Next, the response of the developed model is compared with the experimental data to validate its accuracy. Then, a self-tuning PID controller based on BP neural network is developed to control the drum level changes. Simulation results show improved performance of this controller in terms of less overshoot and settling time, compared to the classic PID controller used in Qom power plant.