A data-driven soft sensor modeling for furnace temperature of Opposed Multi-Burner gasifier

The Opposed Multi-Burner (OMB) Coal-Water Slurry (CWS) gasification is a new large-scale coal gasification technology with higher product yield, lower oxygen and coal consumption than that of Texaco CWS gasification technology. However, current furnace temperature measurements of OMB and other gaisifiers are unstable and even short-life due to the extreme internal environment: high temperature, strong corrosion, etc. Therefore a new data-driven soft sensor modeling technique for furnace temperature of OMB gasifier is proposed and the selection of secondary variables and model structure of BP neural network is studied in this paper. Results indicate that, the furnace temperature predictive model integrating Principal Component Analysis (PCA) and BP neural network has a promising performance with good predictive precision.