Modeling and evaluation of the safety control capability of coal mine based on system safety

Abstract

The intrinsic level of coal mine safety is directly related to its connatural risks, namely the hazard factors. However, the offset risk factors, namely the safety control capability, ultimately determine the desirable operating level. In this paper, the offset risk factors were studied from the aspects of personnel, material (machine), environment and management to control coal mine safety in contrast to describing system risk. The model was constructed and described using safety entropy and cask theory by fully understanding the hazard factors and offset risk factors. Then a set of model evaluation indexes were derived based on the above analysis. A back-propagation neural network (BPNN) was developed to evaluate the safety control capability. It was trained and tested with data collected from forty-one state-owned coal mines in China; thirty-six were used to train the network and the rest were used to test it. The results showed that simulation performance was acceptable and the goodness of fit was high. The weights of personnel, material (machine), environment and management on the safety control capability are 0.26, 0.29, 0.22 and 0.23 respectively. Finally, the trained network was applied to assess the Wulanmulun mine's safety control capability. The results indicated a high level of safety control capability (0.93).