Stability Grading And Its Intelligent Prediction Of Complicated Goaf In Metal Mine

With the rapid and continuous development of national economy, China’s demand for natural resources is constantly increasing, especially for mineral resources. People benefit a lot economically from the exploitation of mineral resources, however, unscientific mining and civil predatory mining leads to formation of the complicated goaf (group), which can easily cause a variety of geological disasters. Government, enterprises attach great importance to the various types of disasters induced by the goaf. Therefore, the study of the stability of the complicated goaf and the implementation of safety management measures is necessary. Titled Stability Grading and Its Intelligent Prediction of Complicated Goaf in Metal Mine, this paper carries out the revelant analysis research.In this paper,14indicators impacting the stability of the goaf were selected as index system for grading and evaluation based on the goaf destabilization mechanism and system security analysis, and combined with the analysis of goaf impact factors and engineering practice. unascertained measurement evaluation model was established based on unascertained measure theory; according to the classification standard of respective indicators, the single index unascertained measurement functions were established, from which single index measurement value (ie, the stability degree membership of indicators) of indicators for each evaluation object can be calculated and then single index measurement matrix for each object can be figured out.; the information entropy theory was adopted to calculate the weight value of indicators influencing goaf Stability and multi-indicator measure vector for each evaluation object can be calculated; Finally, Grading was done according to confidence criteria. Combined with measured data of the dabaoshan mine goaf, using this model, stability classification was done to the77goaf in the mining area, which showed the stability of most of studied goaf and provided a reference for the safe handling of the goaf; At the same time, the classification results was compared with the fuzzy mathematical evaluation results, which fully reflects the superiority of the unascertained measurement evaluation model.Finally, the grading evaluation results of74goaf were used to train BP neural network in order to establish the BP neural network forecasting model which can forecast the stability degree of new goaf. The verification results show that the prediction results of the model are satisfying, and the prediction results can actually provide some guiding significance for the project.