Prediction of Key Parameters of Coal Gasification Process Based on Time Delay Mining Fuzzy Time Cognitive Maps

Although China is a country short of oil and gas, its coal resources are relatively abundant. At present, the development of coal chemical industry has become a concern, and coal chemical industry cannot develop without the preparation of synthesis gas, coal gasification is the necessary means to prepare synthesis gas. Due to the time delay between process variables, the prediction model cannot reflect the real causality of each variable in the system, which leads to the problems of poor interpretation and low accuracy of prediction results. It is difficult to deal with this kind of multi-variable time series data by traditional model method. For this reason, this paper presents a time-delay mining fuzzy time-cognitive maps (TM-FTCM), which uses cross-correlation function (CCF) to mine the time-delay information from the data. It effectively solves the problem of inaccurate prediction model due to delay between industrial process variables. The validity of the proposed method is verified by the experimental analysis of the coal gasification process data.