Research on microseismic signal identification through data fusion

• Published in: Computers & geosciences 2024-10, Vol.192, p.105708, Article 105708
• Main Authors: Zhang, Xingli, Zhang, Zihan, Jia, Ruisheng, Lu, Xinming
• Format: Article
• Language: English
• Subjects: Blasting signal; Data fusion; Dual branch network model; Microseismic signal; Signal identification
• ISSN: 0098-3004
• DOI: 10.1016/j.cageo.2024.105708

The present study proposes a double-branch classification network, DPNet (Double Path Net), for the classification and identification of microseismic and blasting signals based on multimodal feature extraction. The vibration signals’ one-dimensional spectrogram and two-dimensional wavelet time–frequency graph are inputted into the double branch network. Subsequently, convolutional neural networks and ResNet are employed to extract the one-dimensional frequency features and two-dimensional time–frequency features of the vibration signals, respectively. Experimental results demonstrate that our proposed method achieves outstanding classification performance with an accuracy of 97.34% for microseismic signals and blasting signals. This research not only provides innovative solutions to practical problems but also introduces a novel idea of multimodal feature extraction at a theoretical level. By successfully applying it to efficiently classify complex signals in mining engineering, we offer a feasible solution that holds promising prospects for practical applications in this field. •The original signals are processed by FFT and wavelet transform, which lays a foundation for feature extraction.•A double branch classification network DPNet (Double Path Net) is proposed for the classification of microseismic and blasting signals based on multimodal features.•The one-dimensional spectrum and two-dimensional time–frequency graphs of the signals are used as inputs, and the one-dimensional frequency features and two-dimensional time–frequency features of the signals are extracted by DPNet respectively.•DPNet combines the advantages of CNN and ResNet to classify and identify the microseismic signals and blasting signals of coal rock fracture.•The experiment proves that DPNet achieves 97.34% accuracy in the classification task of microseismic and blasting signals, with excellent performance.