Study On Detecting Technique For Underground Coal Mine Dynamic Disasters Source

Most of the China’s coal are buried deep and in complex geological conditions，Over96%of the total reserves of resources are suitable for underground mining， the complexityof the coalfield geological conditions has brought a lot of dynamic disaster for coal minesafety production. The disaster with a sudden, serious and strong failure features has broughtmajor injuries and property losses to mine safety production. As the number of high gasmine and coal and gas outburst mines outburst in Guizhou ranking the first，Therefore，the coal and gas outburst is one of the urgent needs to address the dynamic disasters in theprovince’s coal mine safety production，and is the top priority of the coal mine disaster，This paper selects Guizhou Wangjiazai Coal Mine as research background，grounded on tothe combined effects of Theory and coal and rock acoustic emission. Acoustic emissiondetection technology on-site monitoring of the mining face，And BP neural network hasbeen taken into account to study the face prominent risk prediction.（1）This paper first describes the main features of the disaster of mine power，theDisaster source of coal and gas outburst power detection technology status and developmenttrend，and select the power of dynamic non-contact continuous detection of acousticemission as coal and gas outburst disaster source detection techniques.（2）Using Wangjiazai coal mine rock mechanics，the seepage parameters theRFPA2Dsoftware to establish numerical models and model results show that the combinedeffect hypothesis applies to mine coal and gas outburst mechanism，will highlight theprocess is divided into brewing，the initial growth mitigation and again stimulate the fivestages， and highlight process changes.（3）Emission detection technology based on the principles and theoretical models ofthe acoustic emission of coal or rock sound of coal and gas outburst in the theoretical aspectsof the feasibility analysis，and identified three and highlight the risk of acoustic emissioncommon monitoring parameters：the acoustic emission event，total events and the rate ofenergy.（4）Coal and gas outburst acoustic emission BP neural network prediction method,the application of BP neural network theory to conduct a feasibility analysis completed onthe structure of BP neural network design and programming.（5）Design the installation of acoustic emission sensors, measurements taken faceacoustic emission structure in the commonly used monitoring parameters of training samples by using MATLAB software to complete the training of BP neural network and BP neuralnetwork used in coal and gas outburst risk predictionin. By comparing the traditional methodof cuttings gas desorption index, show that the method is simple, the prediction is fast andreliable degree of advantage.