Study on transportation law and inverse prediction of high-temperature points during spontaneous combustion of coal in closed coal storage yard
In order to investigate the transportation law and inverse prediction method of high-temperature points during the spontaneous combustion of coal in closed coal storage yard,the transportation law of high-temperature points of coal in closed coal storage yard was studied by constructing a simulation experimental bench of coal spontaneous combustion,and the corre-lation between the surface layer and internal temperature data of coal spontaneous combustion was analyzed.The random forest(RF)and BP neural network(BPNN)before and after the optimization of genetic algorithm were used to establish a predic-tion model of high-temperature points during the spontaneous combustion of coal in closed coal storage yard.The results show that during the spontaneous combustion process of loose coal,the transportation direction of high-temperature points presents a nonlinear moving law,which is mainly affected by the effect of fissure in the process of coal combustion.The temperature data between the surface layer sites and the lower layers presents the strong correlation.Both RF and BPNN have strong robustness and fault tolerance in the prediction results,and the excellent nonlinear mapping ability of BPNN makes its processing results better than RF.After the optimization of RF and BPNN,the optimization of genetic algorithm improves the prediction accuracy of both RF and BPNN,and the impact on RF is the greatest,with the accuracy of model reaching more than 0.98 after optimi-zation.The research results can provide a reference for the prevention and control of natural ignition of coal in the closed coal storage yards.
NETL
NSTL
万方数据
