Modeling and control of Superheated steam Temperature control process

Abstract

newlinePower plants are virtually the prime movers of Indian economy The newlinecontribution of thermal power plants to the total power generated has considerably newlineincreased over the past decade The superheated steam temperature control is critical newlineto the efficient operation of steam boiler turbines The superheated steam newlinetemperature control process of 210MW Plant of Mettur Thermal Power Station newline MTPS Mettur India is taken up as case study newlineModeling is the fundamental step to process control From the real newlinetime data obtained from the process using Auto regressive with Exogenous newlineinput ARX method four transfer function models are obtained These newlinemodels are used for the design of controllers for the process Artificial neural newlinenetworks are popularly used for the identification of nonlinear dynamical newlinesystems The nonlinear models developed are the neural network model and newlinethe Adaptive network Based Fuzzy Inference System ANFIS model newlinePID controllers are versatile simple and robust and therefore find newlineWide spread application in industry A new scheme for the tuning of PID newlinecontroller parameters using symbolic computation is proposed Discrete PID newlinecontrollers are designed for the discrete models by tuning of the controller newlineparameters based on the reduction of the cumulative error newlineGenetic Algorithm GA is a powerful optimization method which newlinecan be used in control systems in conjunction with other techniques to find newlineoptimum controller parameters Simulated Annealing SA is an optimization newlinemethod similar to the physical process of heating up a solid until it melts newlinefollowed by cooling it down to crystallize into a perfect lattice A hybrid newlinegenetic algorithm simulated annealing approach is presented to determine the newlineparameters of a PID controller newline newline