Abstract
The identification of suspicious microseismic events is the first crucial step in processing microseismic data. In this paper, we present an automatic classification method based on a deep learning approach for classifying microseismic records in underground mines. A total of 35 commonly used features in the time and frequency domains were extracted from waveforms. To examine the discriminative ability of these features, a genetic algorithm (GA)-optimized correlation-based feature selection (CFS) method was applied. As a result, 11 features were selected to represent microseismic records. By dividing each microseismic record into 50 frames, an 11 × 50 feature matrix was utilized as the input. A convolutional neural network (CNN) with 35 layers was trained on 20,000 samples recorded at the Huangtupo Copper and Zinc Mine. There are 5 types of events: microseismic events, blasting, ore extraction, mechanical noise, and electromagnetic interference. The event type was correctly determined by the trained CNN classifier 98.2% of the time, outperforming traditional machine learning methods.
Highlights
Microseismic monitoring is routinely performed to observe variations in ground pressure within mines
TRAINING AND TESTING THE convolutional neural network (CNN) Based on the microseismic records from the Huangtupo Copper and Zinc Mine, we randomly segment the dataset into a training set, validation set and testing set, which comprise 80%, 10% and 10% of the data, respectively
These findings demonstrate that the deep learning approach has excellent efficiency and reliability for the classification of microseismic data
Summary
Microseismic monitoring is routinely performed to observe variations in ground pressure within mines. A microseismic monitoring system can indicate the stability of a rock mass via an analysis of microseismic events [1]. To analyze microseismic events in real time, scientists worldwide have proposed a variety of automatic processing methods, such as P- and S-wave arrival picking [2], source localization [3], and source parameter calculation [4]. The microseismic records collected for rock masses are commonly triggered by rock fracturing, blasting, the use of a drill jumbo, and ore extraction. The identification of suspicious microseismic events constitutes the first crucial step of microseismic data processing and is usually performed by experienced analysts through the manual visual scanning of waveforms.
Sign-in/Register to view highlights & summary 
Published Version (
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