Prediction Modeling for Heat Transfer of Boiler Based on Partial Least-square Regression

Due to the complexity of heat transfer mechanism, heat transfer model in furnace with simple form and applying to real-time calculation accuracy usually has a low accuracy. By using partial least squares regression (PLSR) method the boiler heat transfer model was established. This method combines the forecasting by means of model with non-model style data connotation analysis, thus, the model by mean of PLSR overcomes multivariable correlativity. The prediction model for boiler heat transfer based on PLSR has high accuracy, clear physical meaning, and more reasonable explanation to heat transfer mechanism of boiler, as compared with the ordinary least-squares regression model. The modeling method presented in this paper can be used for simulation and model-based control system design on ships steam power system and other large power system.