Research And Design Of Mine Rock Stability Early Warning System Based On Acoustic Emission

With the rapid development of my country’s economy and industry,the demand for mineral resources is increasing day by day and underground operations in mines are becoming more and more frequent.With that comes the issue of safety inside the mine.These include the imbalance of rock stability and extrusion fracture caused by mining.As a result,accidents such as roof falls and collapses occur,which pose a major threat to the life and safety of mine operators.Therefore,the mine rock formation stability early warning system came into being,which plays a vital role in the safety production of the mine and the life safety of the relevant personnel.It can warn relevant operators of possible dangers and locate the location where dangers may occur.The existing rock formation stability early warning systems have many shortcomings.On the one hand,the accuracy of acoustic emission signal identification is not high,and the identification of acoustic emission signal needs to be determined by experience;On the other hand,the positioning accuracy is not enough.Aiming at these problems,this paper conducts in-depth research,and designs and develops a corresponding early warning system.The main work of this paper is as follows:(1)The recognition accuracy is affected by the noise contained in the acoustic emission signal.Aiming at this problem,the paper de-noises the sound signal received by the sensor by wavelet threshold,improves the soft and hard threshold function,and proposes a new threshold function.The experimental results show that the denoising effect of the new threshold function is better than that of the soft and hard threshold functions.(2)Only by identifying the real acoustic emission signal and extracting the relevant early warning parameters can it provide effective early warning for the staff.Therefore,this paper proposes an acoustic emission signal recognition method based on spectrogram and CNN.Extracting the spectrogram of the acoustic emission signal which is used as the input of the CNN to realize the identification of the acoustic emission signal.The experimental results show that for acoustic emission signals,the recognition accuracy of CNN is significantly higher than that of the BP network,and the recognition accuracy of CNN can reach 96%.(3)Acoustic emission source location can inform safety workers where mines may collapse.Therefore,this paper proposes an improved time-difference localization method for acoustic emission sources combined with probability density fitting.Firstly,Using the cross-correlation method to obtain the arrival time difference between different sensors,then solving the travel time equation and combining the probability density function in the probability and statistics method to fit the positioning results.Thus,the influence of uneven wave speed on the positioning results can be reduced.The maximum deviation of algorithm positioning is 0.5 meters.(4)The thesis designs a visual early warning system for mine rock formation stability.Through the interface,the waveform of the acoustic emission signal of 12 channels,the characteristic parameters of a selected channel,the corresponding acoustic emission source positioning coordinates,and the three-level alarm mark can be displayed.