Study on Index Weighting Coefficients of Bump Tendency and Bump Hazard of Coal Body

To accurately reflect the bump tendency of coal specimen, a lot of experimental data had been used to conduct statistical analysis. Based on the genetic algorithm, the BP neural network method was optimized, the index weighting coefficients of bump tendency were determined, including bump energy index, elastic energy index and dynamic damage time. Furthermore, the weighting coefficients were corrected by the expert scoring method. According to the statistic law of experimental data, the fuzzy membership functions were respectively established, and the comprehensive fuzzy evaluation system of bump tendency had formed. Compared with the actual condition, the results showed that this method contained the advantages of BP neural network and expert scoring method, which were optimized by the genetic algorithm, it had good credibility.