A novel online model for furnace exit gas temperature of coal-fired boiler

Furnace exit gas temperature (FEGT) is the key parameter in the furnace ash fouling monitoring system. Since the standard least squares support vector machine (LSSVM) is not suitable for online identification and control of FEGT, a novel CM-LSSVM-PLS method is proposed to predict FEGT in this paper. In the process of CM-LSSVM-PLS method, c-means cluster (CM) algorithm is used to partition the training data into several different subsets by considering the characteristics of operational data. Submodels are subsequently developed in the individual subsets based on LSSVM method. Partial least squares algorithm (PLS) is employed as the combination strategy. The online updating algorithm is then applied to the CM-LSSVM-PLS model. The proposed online model is verified through operation data of a 300MW generating unit. The simulation results show that the proposed online updating model is effective for online FEGT forecasting.