Detection of metal mine hidden goaf and intelligent prediction of boundary

• 1、School of Resources and Safety Engineering, Central South University

Abstract：As the size and location of the hidden goaf uncertainty exists, it causes the disaster controlling and goaf treatment is difficult to implement, precise detection methods and technology of the hidden goaf must to be studied. According to the characteristics of hidden goaf of Luanchuan Molybdenum mine and Dabaoshan Copper mine, a new detection method, high-density electrical method combined with the seismic imaging method, aided with the drilling verification metal hidden goaf, is put forward. Results of engineering application show that this method is of high precision, economic and feasible. Using fractal theory to study fractal characteristics of goaf, the data display, fractal box dimension of goaf reflects the degree of complexity and irregularity. With fractal box dimension and engineering detection data of goaf, neural network intelligent forecast model is established for the upper boundary and lower boundary of goaf. According to the existing characteristic of goaf boundary, the model can intelligent prediction goaf boundary, which is unknown or instrument can not measure, and it saves the cost of hidden goaf exploration and improves the efficiency of detection.

Keywords： hidden goaf intelligent prediction model fractal neural network