A FUZZY COMBINED FORECASTING MODEL OF COAL SPONTANEOUS COMBUSTION

Pengtao Jia¹, Jun Deng², Shuhui Liang¹

¹School of Computer Science and Technology, Xi’an University of Science and Technology, Xi’an Shaanxi 710054, China

²School of Energy Engineering, Xi’an University of Science and Technology, Xi’an Shaanxi 710054, China

This paper focuses on the effective analysis of the coal spontaneous combustion monitoring data, so as to realize the accurate and reliable coal spontaneous combustion limit parameter prediction. Firstly, a weighted multimember fuzzy operation model was constructed. When the additive generator of the model changes, this model can generate new operation clusters. Based on it, a new combined forecasting model of coal spontaneous combustion limit parameter is proposed. The new model can use linear and nonlinear models as its single forecasting models. Its combination is variable and has good generalization ability. Then, the BP neural network model and the support vector machine were used as the single forecasting models of the new model. Finally, for realizing the optimal combination of single models, genetic algorithm and least square method were used to evaluate parameters of new model. The experimental analysis shows that the new model leads to less error and better performance than single models. It can be concluded that the new combined forecasting model is suitable for coal spontaneous combustion.