Particle Swarm Optimization for tuning a Fuzzy Supervisory Controller Parameters (Takagi Seguno and Mamdani Engines)

Abstract

Steam generation systems are a crucial part of most power plants. Therefore, boiler control is an important problem for power plants that are frequently changing load or subject to sudden load disturbances, which are common in current market driven electricity industry. Hence, the control of such processes needs the design of more effective controllers. Although Fuzzy logic controllers seem more adequate for controlling chemical processes, since they provides solutions to incompletely defined and nonlinear processes industrials does not prefer such type of controllers since the tuning of these controllers requires the adjustment of a large number of parameters, which is tedious. The fact that let the industrials prefer the use of simple PID controllers rather than complicated fuzzy or neural controllers. Hence in this paper a simple PD controller is used to control the steam flow parameters of whereas its parameters are deduced using fuzzy supervisory controller the parameters of the supervisory controller are optimized using particle swarm optimization algorithm, the most known inference Engines SEGUNO and MAMDANI are considered.