A NEURAL-NET MODEL-BASED MULTIVARIABLE LONG-RANGE PREDICTIVE CONTROL STRATEGY APPLIED IN THERMAL POWER-PLANT CONTROL

A constrained multivariable control strategy along with its applicatio n in more efficient thermal power plant control is presented in this p aper. A neural network model-based non-linear long-range predictive co ntrol algorithm is derived, which provides offset-free closed-loop beh avior with a proper and consistent treatment of modeling errors and ot her disturbances. A multivariable controller is designed and implement ed casing this algorithm. The system constraints are taken in to accou nt by including them in the control algorithm using real-time optimiza tion. By running a simulation of a 200 MW oil-fired drum-boiler therma l power plant over a load-profile along with suitable PRBS signals sup erimposed on controls, the operating data is generated Neural network (NN) modeling techniques have been used for identifying global dynamic models (NNARX models) of the plant variables off-line from the data. To demonstrate the superiority of the strategy in a MIMO case, the con troller has been used in the simulation to control main steam pressure and temperature, and reheat steam temperature during load-cycling and other severe plant operating conditions.