Abstract
The microseismic monitoring signals which need to be determined in mines include those caused by both rock bursts and by blasting. The blasting signals must be separated from the microseismic signals in order to extract the information needed for the correct location of the source and for determining the blast mechanism. The use of a convolutional neural network (CNN) is a viable approach to extract these blast characteristic parameters automatically and to achieve the accuracy needed in the signal recognition. The Stockwell Transform (or S-Transform) has excellent two-dimensional time-frequency characteristics and thus to obtain the microseismic signal and blasting vibration signal separately, the microseismic signal has been converted in this work into a two-dimensional image format by use of the S-Transform, following which it is recognized by using the CNN. The sample data given in this paper are used for model training, where the training sample is an image containing three RGB color channels. The training time can be decreased by means of reducing the picture size and thus reducing the number of training steps used. The optimal combination of parameters can then be obtained after continuously updating the training parameters. When the image size is $180\times140$ pixels, it has been shown that the test accuracy can reach 96.15% and that it is feasible to classify separately the blasting signal and the microseismic signal based on using the S-Transform and the CNN model architecture, where the training parameters were designed by synthesizing LeNet-5 and AlexNet.
Highlights
The technology of Microseismic monitoring in mines is very important, as an effective means for forecasting and creating the early warning of a rock burst disaster is needed
This paper focuses on the existing problems of the mixture of mine microseismic signal and blasting signal, analyses the characteristics of the time-domain waveform of the monitoring signals, studies the convolutional neural network (CNN) algorithm and the concept and algorithm implementation of the Stockwell Transform (S-transform), and puts forward a recognition method for mine microseismic and blasting signals
In view of the superior time-frequency characteristics of the S-Transform, the original microseismic data are transformed first by the algorithm to obtain a two-dimensional time-frequency image, followed by classifying them by use of the CNN algorithm to achieve a high accuracy in the microseismic signal and the blasting signal classification
Summary
The technology of Microseismic monitoring in mines is very important, as an effective means for forecasting and creating the early warning of a rock burst disaster is needed. In view of the superior time-frequency characteristics of the S-Transform, the original microseismic data are transformed first by the algorithm to obtain a two-dimensional time-frequency image, followed by classifying them by use of the CNN algorithm to achieve a high accuracy in the microseismic signal and the blasting signal classification. The model architecture for the classification of the microseismic signal and the blasting signal is designed based on the combined advantages of LeNet-5 and AlexNet. The model architecture of the two-dimensional time-frequency image by use of the S-Transform is: input layer - convolution layer 1 (ReLU, batch standardization) - pooling layer 1 (LRN) convolution layer 2 (ReLU, batch standardization) - pooling layer 2 (LRN) - fully connected layer 1 (ReLU, Dropout) fully connected layer 2 (ReLU, Dropout) - SOFTMAX Classification result. A total of 116 blasting samples were obtained in one month, 90 of which were taken as blasting training samples and 26 as blasting test samples. 116 microseismic samples were obtained, 90 of which were taken as microseismic training samples and 26 as microseismic test samples
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
