Fuzzy-based sensor validation for a nonlinear bench-mark boiler under MPC

Most of the power plants are automated with complete closed loop control systems. For this purpose, numerous sensors are required to monitor the proper functioning of the plant. Due to various reasons, these sensors may not be able record actual value of the measured variable. This misleads the controller taking faulty actions on manipulated variables. It is very important to monitor the measured parameters which help not only for operating the plant safely and also predicting the incipient failures of the sensors. In this paper, a fuzzy based data validation algorithm is implemented to a bench mark boiler which is controlled by a model predictive control is presented. A non-linear bench boiler model has been taken for study and a model predictive control(MPC) is implemented and tested on its combustion control unit. Sensor data validation algorithm is developed for oxygen sensor which measures the excess oxygen present on the air/fuel rate of the combustion control unit.