Performance Analysis Of Radial Basis Function Neural Network Based Model Predictive Control For Biomass Boiler Process

Abstract

As the world population has increased, electricity demand and consumption have risen rapidly in recent years. As a result, fossil fuel costs have risen and the associated air pollution has become a global issue. A biomass boiler power plant is one of the most effective solutions to this problem. The boiler plant is a multi-input, multi-output, time-varying and nonlinear system by nature. The majority of biomass boiler controllers use classical PID controllers. The inability of these classical controllers to control time-varying and nonlinear systems is the source of the plant's poor performance. To make the controller better, first determine the system model of the nonlinear, MIMO and time-varying system by employing a data-driven Radial Basis Function Neural Network model and then input the model result into the model predictive controller.