Abstract
The classification of multichannel microseismic waveform is essential for real-time monitoring and hazard prediction. The accuracy and efficiency could not be guaranteed by manual identification. Thus, based on 37310 waveform data of Junde Coal Mine, eight features of statistics, spectrum, and waveform were extracted to generate a complete data set. An automatic classification algorithm based on artificial neural networks (ANNs) has been proposed. The model presented an excellent performance in identifying three preclassified signals in the test set. Operated with two hidden layers and the Logistic activation function, the multiclass area under the receiver operating characteristic curve (AUC) reached 98.6%.
Highlights
As a real-time, continuous, and dynamic method, microseismic monitoring has been widely used in coal mines at home and abroad in recent years [1,2,3]. e microseismic monitoring system mainly monitors rock stability by analyzing the microseismic events produced by mining activities [4, 5]
By eliminating the pollution of electromagnetic interference signals, mechanical drilling signals, and blasting vibration signals, we can determine the occurrence law of pure microseismic events in the mine and identify high-quality microseismic signals for fine positioning. e relationship between the loss and the training iterations is shown in Figure 12 under different combinations of hidden layer numbers and activation functions
The loss is 0.299. e reason why we did not study artificial neural networks (ANNs) with three hidden layers and above is that the deeper the number of layers, the stronger the ability of classification, and the better the effect is in theory
Summary
As a real-time, continuous, and dynamic method, microseismic monitoring has been widely used in coal mines at home and abroad in recent years [1,2,3]. e microseismic monitoring system mainly monitors rock stability by analyzing the microseismic events produced by mining activities [4, 5]. As a real-time, continuous, and dynamic method, microseismic monitoring has been widely used in coal mines at home and abroad in recent years [1,2,3]. E most widely used method was the short-term average/long-term average (STA/LTA). Algorithm proposed by Allen [11], which calculated a short-term and a long-term average of the magnitude of the signal to recognize earthquakes from single traces. To improve the confidence of arrival-time picking in low signal-to-noise ratio (SNR) condition, Akram and Eaton [13] combined the ratio of the peak eigenvalues (PER) and STA/LTA to enhance signal coherency. The problem of STA/ LTA algorithm is that it is sensitive to the background noise and window size is crucial for the accuracy of the picking results. It is widely accepted that the accuracy of classical algorithms is not satisfactory
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
