Research Of Boiler-Turbine Unit Fuzzy Rule Based Multiple Model Coordinated Control System

The coordinated control system (CCS) for boiler-turbine unit is a rather complicated multi-variable nonlinear control system.Many advanced control algorithms have been found to have preliminary, successful applications in the system.Some control difficulties have been overcome.However, the problems encountered with the frequent load change of large magnitude, which are badly needed to tackle in electrical industries, are still unsolved.In order to give some hints on the solution of these problems, we have studied in this dissertation a fuzzy rule based multiple model control scheme and its related theories.Our study addresses the following topics:A conception of relative nonlinear measure on the equilibrium manifold parameter has been derived based on the definitions of equilibrium manifold and nonlinear measure about nonlinear system.And the nonlinear measure curve of a material boiler-turbine CCS model has been drawn.From the curve, we linearised the CCS nonlinear model and constructed the T-S fuzzy model of the unit by linking together the linear models of the unit with membership functions.A kind of fuzzy supervisory control system in which using the multi-variable inner model controller or robust controller as its local controller and T-S fuzzy model as the inner model of the supervisory controller has been proposed.And a kind of theory on its global asymptotical stability has been presented.From the T-S model of a CCS, the fuzzy supervisory controller has been provided.We proved the global asymptotical stability of the supervisory control system, studied the problems like performance of setpoint following, performance to inner disturb, the affection of supervisory controller's parameters to the control system and the intrinsical relationship on fuzzy supervisory control and fuzzy variable structure control.A setpoint following multi-variable PDC control algorithm has been introduced, in which the problems of stability demand, input-output constrains, dynamic performance etc.are all solved by a set of LMI convex optimization algorithm.The PDC control algorithm has been presented from the T-S fuzzy model of a CCS, the research to the CCS model between the PDC control and the fuzzy supervisory control in which using robust controller as its local controllerhas been conducted.The PDC algorithm can overcome the weaknesses of normal fuzzy control about stability and dynamic performance analyzing.After analyzing the shortages like dynamic process performance etc.of DEB400, a practical application of the fuzzy supervisory control on a 500MW turbine-generator unit is presented in this paper.The correlative problems like feedforward, restructure the state variable of boiler-turbine control loop etc.have been studied.Finally, we analyzed the main problems existed in CCS at present, and gave some viewpoints on the future development on CCS.