Intelligent Microseismic Monitoring of Mineral Equipment Manufacturing Industry Based on Deep Learning in the Artificial Intelligence Environment

Abstract

Abstract To prevent major disasters caused by mine engineering, the structure and early warning effect of microseismic monitoring systems in the mineral equipment manufacturing industry based on deep learning (DL) are explored under the background of artificial intelligence (AI). The purpose is to provide a reference for predicting the law of strata movement under high-intensity mining conditions. In this study, firstly, the principle of the microseismic monitoring sensor system is analyzed, and the structure of the microseismic monitoring system of intelligent mining face in the mineral equipment manufacturing industry is further understood. Secondly, as one of the AI technologies, DL introduces the Convolutional Neural Network (CNN) and transfer learning (TL) into the processing and intelligent warning of mine microseism signals. Moreover, an intelligent microseismic monitoring system based on CNN and TL for the mineral equipment manufacturing industry is constructed to realize the identification of microseismic events. Finally, taking the Xiaojihan coal mine as an example, the microseismic activity regularity of its mining face is analyzed, and the performance of various microseismic signal recognition models is compared. The results reveal that the TL-CNN algorithm in the model constructed in this study has the best performance. Compared with other methods, Mean Absolute Error, Root Mean Square Error, and Mean Absolute Percentage Error indicators have decreased by at least 28.2%, 21.0%, and 36.2%. This shows that the TL-CNN model-based multi-input sequence model is more suitable for forecasting rockburst risk. The mining microseismic signal processing model based on the CNN discussed here provides a reference basis for ensuring the accuracy of rockburst microseismic warning to some extent.