Data Mining for Complex Thermal System Modeling

As a new modeling thought, the accurate mathematical model of complex thermal system is established by data mining method, which obtains effective information from the large number of real-time operation data and avoids low accuracy of conventional modeling method caused by some assumption. A kind of basic modeling mode, including data preprocessing, mining model, verification model and the strategy from data to mathematic model, is proposed in the paper. Under this mode the oxygen content in boiler flue gas is taken as an example, the partial system mathematical model is established based on partial least-square regression with the real-time data collecting in field. The model error can be controlled within 3.6%. The example results indicate that the proposed modeling thought and the mode are effective for the complex thermal system, and can enhance the modeling accuracy.