Improving the prediction accuracy of coal mine accidents can effectively support the prevention of coal mine accidents. In order to identify the occurrence raw of the coal mine accidents, the number of coal mine accidents from 2000 to 2018 was taken as a sample. The data set was constructed by forecasting the data of the next year with the data of every three years. Therefore, a total of 16 groups of data were obtained. Then the 16 groups of data were divided into the training group and the test group. The BP neural network improved by genetic algorithm (GA-BP) and wavelet neural network were used to establish the prediction model respectively. The prediction results of the two methods were analyzed. The results showed that the predicted result of GA-BP neural network was closer to the actual value. Therefore, the prediction model constructed by GA-BP was used to predict the number of coal mine accidents in 2019 and 2020, which were 199 and 176.
白彦龙1，陈 昱1，白长江1，梁建明2，西 龙1，薛玉壁1，彭佳佳1，王 欣1. 基于神经网络的煤矿事故数量预测研究[J]. 煤炭与化工, 2020, 43(9): 91-94,97.
Bai Yanlong1, Chen Yu1, Bai Changjiang1, Liang Jianming2,. Application of neural network in predicting the number of mine accidents. CCI, 2020, 43(9): 91-94,97.
