Normalized Normal Constraint Algorithm Based Multi-objective Optimal Tuning of Decentralised PI Controller of Nonlinear Multivariable Process – Coal Gasifier
Almost all the industrial processes are multivariable in nature and are very difficult to control, since it involves many variables, strong interactions and nonlinearities. Conventional controllers are most widely used with its optimal parameters for such processes because of its simplicity, reliability and stability. Coal gasifier is a highly nonlinear multivariable process with strong interactions among the loop and it is difficult to control at 0% operating point with sinusoidal pressure disturbance. The present work uses Normalized Normal Constraint (NNC) algorithm to tune the parameters of decentralised PI controller of coal gasifier. Maximum absolute error (AE) and Integral of Absolute Error (IAE) are objective function while the controller parameters of decentralised PI controller are the decision variables for the NNC algorithm. With the optimal controller the coal gasifier provides better response at 0%, 50% and 100% operating points and also the performance tests shows good results.
KeywordsCoal gasifier Multi-Objective Optimization Multivariable process Normalized Normal Constraint Algorithm PID Controller tuning
Unable to display preview. Download preview PDF.
- 8.Kotteeswaran, R., Sivakumar, L.: Lower Order Transfer Function Identification of Nonlinear MIMO System-Alstom Gasifier. International Journal of Engineering Research and Applications 2(4), 1220–1226 (2012)Google Scholar