A hybrid recognition model of microseismic signals for underground mining based on CNN and LSTM networks

Microseismic (MS) monitoring technology has been widely used to monitor ground pressure disasters. However, the underground mining environment is complex and contains many types of noise sources. Furthermore, traditional recognition method entails a process with low accuracy for MS signals, so it difficult serve safe production mines. Therefore, this study established hybrid model combining singular spectrum analysis (SSA) method, convolutional neural networks (CNN), long short-term memory (LSTM). First, principal components signals were extracted SSA then spatial temporal features separately CNN LSTM. Based on actual field data collected from Xiadian Gold Mine, was compared CNN, LSTM, back-propagation (BP), as well commonly methods including support vector machine (SVM), decision tree (DT), K-nearest neighbor (KNN), linear discriminant (LDA). The results show that proposed can accurately extract further improve signals' performance. mechanical in particularly increased using model, which avoids confusion signals.