Study On Stability Prediction Of Goaf In Panlong Lead-Zinc Mine Based On Acoustic Emission Technology

The stability of goaf will directly affect the safety production of mine and the personal safety of workers,because the research on the stability of goaf has always been a very important topic in mine production.Based on the key R & D plan of Guangxi and the engineering background of Guangxi Panlong lead zinc mine of Zhongjin lingnan mining industry in guangxi,this paper carried out the prediction research on the stability of goaf.The specific research contents and results of this paper are as follows:(1)Based on iceemdan and mc-cnn model,the recognition and classification of acoustic emission signals from underground mines were realized.Firstly,the AE signal was processed by iceemdan,and the IMF component of the original signal can be obtained.According to the kurtosis value of IMF,the components with more feature information were selected and converted into the input of multi-channel image;Secondly,different weights were given according to the kurtosis values of the input components of each channel,so as to highlight the feature information contained in the image;Finally,the actual collected data were trained and verified on the model,and the accuracy of classification and recognition was 97.64%.Compared with traditional methods such as artificial neural network,the accuracy of this method was much higher than other methods;(2)Then,on the premise of completing the identification of acoustic emission events,taking the parameters of acoustic emission events as the research object,the grey catastrophe prediction model was established.Firstly,the grey model was used to process the acoustic emission data and synthesize it into the system potential function,and then the change trend of acoustic emission event parameters was simulated and predicted according to the catastrophe theory,so as to realize the prediction of goaf stability;The fitting accuracy of the prediction function model with three parameters was calculated,and the fitting grade was good.Using the established prediction function model to predict the acoustic emission data of next month,the prediction results were the same as the development trend and in line with the actual situation;(3)The stability prediction model of goaf with three state variables was established by using acoustic emission parameters,the specific expression of the model was obtained by using the model inversion method,the expression of the corresponding stability judgment criterion was obtained according to the nonlinear theory,and the acoustic emission parameters were input into the equation of the judgment criterion,so as to judge the stability;The data not involved in the model inversion were substituted into the model to obtain the prediction results.The two prediction results were consistent with the reality,which can be used as an effective means for the prediction of goaf instability.The research shows that the goaf stability prediction model based on grey catastrophe theory and model inversion method based on acoustic emission monitoring information can successfully predict the goaf stability,and provide an effective guarantee for the safe production and sustainable development of mines.