Microseismic Monitoring Signal Waveform Recognition and Classification: History
View Latest Version
Please note this is an old version of this entry, which may differ significantly from the current revision.
Contributor:
Hongmei Shu
,
Ahmad Yahya Dawod

Microseismic event identification is of great significance for enhancing our understanding of underground phenomena and ensuring geological safety. Microseismic monitoring entails the continuous surveillance of minuscule seismic events during mining activities. These imperceptible events provide valuable information about evolving geological conditions. They serve as early warning signals, offering crucial insights into potential hazards and enabling timely preventive measures. This not only safeguards the well-being of miners but also enhances the overall efficiency and sustainability of mining practices.

  • microseismic events
  • machine learning
  • signal processing
  • waveform recognition

1. Introduction

In the contemporary landscape of mining operations, a profound transformation is underway, marked by the transition to intelligent mining [1]. This paradigm shift is catalyzed by the integration of cutting-edge technologies such as the Internet of Things (IoT) [2], big data analytics [3], and artificial intelligence (AI) [4]. Central to this transformation is the unwavering commitment to enhancing mining safety [5]. As mining operations plunge deeper into the Earth’s crust, the stability of rock formations is increasingly susceptible to disruptions caused by human activities [6]. These disruptions give rise to geological hazards like rock bursts and mining-induced seismic events, posing grave threats to the safety of miners and the productivity of mining endeavors [7][8]. Within this context, microseismic monitoring technology has emerged as a fundamental pillar for ensuring geological safety in mining operations (Figure 1).
Applsci 13 12739 g001
Figure 1. Application of microseismic monitoring in smart mining.

2. Microseismic Monitoring Signals

Microseismic events refer to weak seismic activities caused by minor displacements and stress changes within underground rock formations [9]. In mining environments, factors such as mining activities and rock movements can lead to minor fractures and deformations in rocks, resulting in microseismic events. These events can occur naturally or be triggered by human activities like mining operations and blasting. Monitoring and analyzing microseismic events are crucial for mine safety. Using the example of MMS in a metal mine, the system collects underground microseismic signals. These signals include seismic waveforms from microseismic events, which record the seismic signals generated when underground rock formations undergo slight changes. These signals may encompass different types of events such as rock fractures and blasting. The monitoring system captures these signals using sensors and transmits them to the surface or data centers for further analysis. These signals exhibit unique characteristics in terms of waveform, amplitude, frequency, etc., which can be used to identify various types of microseismic events and provide essential information and warnings for mine safety [10].
For instance, microseismic monitoring systems play a crucial role in ensuring coal mine safety by capturing MS from coal-rock fractures and blasting activities. However, the challenge lies in distinguishing between these signals due to their waveform similarities. Scholars have extensively studied this issue to enable accurate identification of authentic MS within the monitoring system [11][12]. Seismic data analysis methods, commonly used for assessing seismic activity in volatile mines, encounter challenges due to localization errors and incomplete data catalogs caused by unfavorable seismic detector layouts [13]. Furthermore, dynamic disasters like stress-type and fracture-type rock bursts significantly impact mine safety. To address this, a study integrates spatiotemporal parameters through a big data platform and employs the AdaBoost algorithm to predict rock burst risks, contributing to accurate and timely warnings [14]. Shu et al. examined the features and classification of MS in coal mine workings, as well as its importance in the early detection of gas and coal outbursts [15]. In addition, Yin et al. employed a data-driven approach based on deep learning to successfully predict coal seam floor water inrush using microseismic monitoring data [16]. These applications highlight the significance of MMS and data analysis in enhancing safety production in coal mining.

2.1. Microseismic Monitoring Data

Microseismic monitoring data exhibit unique characteristics that are vital for effective event recognition and classification [17]. Firstly, these data demonstrate variations in signal intensity over both time and space, which may be attributed to complex changes in underground rock formations and mining activities. Second, mechanical processes, equipment breakdowns, subsurface water movement, or environmental conditions can contribute to high levels of noise in microseismic data. This complicates data processing because a good distinction between signals and noise is required. Furthermore, due to geological effects, the waveforms of microseismic data are frequently complex and diverse, reflecting the physical qualities of rocks and differences in subsurface structures. Finally, microseismic data span a wide range of frequencies and energy levels, implying that MS can manifest in a variety of frequency ranges and energy levels. This variability causes difficulties in event recognition and classification, necessitating the investigation and processing of various features.
These microseismic surveillance data characteristics represent the complexity of the underlying rocks and mining activities, which require advanced data analysis techniques for interpretation and understanding. Variations in signal intensity across time and place suggest that the distribution of microseismic events is nonuniform, which may be related to the nonuniformity of underground rock layers or mining activity. To improve the detection and analysis of microseismic events in the presence of high noise levels, preprocessing measures such as blurring and filtering are required. The complex waveforms may reflect a variety of subsurface structures, which is important for recognizing various types of occurrences such as rock bursts, demolition, and microfractures. Because of the variety of frequencies and energy levels, multiscale and multifeature analysis approaches must be used to capture the many properties of microseismic signals.
Understanding these properties of microseismic monitoring data is therefore critical for accurate identification and classification of microseismic occurrences. When confronted with such complicated data, researchers must devise proper data processing and analysis methods to distinguish between important events and irrelevant noise, as well as extract key information about microseismic events. Furthermore, different types of microseismic events may necessitate alternative feature extraction and classification approaches to ensure accuracy and reliability. These data features enable in-depth investigations of subterranean engineering and rock behavior and provide critical information for mining activities and underground engineering projects.

2.2. Waveform Features

Microseismic events manifest in various forms of waveform images, each possessing unique characteristics. To comprehensively analyze and classify microseismic events, researchers can employ methods for extracting relevant features from these waveform images. Figure 2 illustrates different types of microseismic event waveform images, encompassing microseismic events (commonly referred to as rock microfracture events), blasting events, rock drilling events, power interference events, and other noise events. The diversity in these images reflects the waveform features of different events; thus, feature extraction from these images can aid in better understanding and distinguishing various types of microseismic events.
Applsci 13 12739 g003
Figure 2. Various types of microseismic event waveform images. (a) Different waveforms of rock microfracture (b) Different waveforms of blasting (c) Drilling (d) Orepass blasting (e) Electrical noise (f) Background noise.
Microseismic event classification can typically be based on triggering mechanisms, waveform characteristics, and occurrence locations [18]. Common classifications of microseismic events include naturally occurring microseismic events, which are triggered by natural processes such as underground structural movement and rock deformation; human-induced microseismic events, which are caused by human activities such as mining operations and blasting; and rock burst events, which are caused by the rupture or collapse of rocks. Based on different triggering mechanisms, monitored microseismic events can be divided into the following categories:
  • Rock microfracture microseismic events, which are usually associated with stress changes and cracks in underground rock formations. The waveform characteristics of these events show small amplitude, high frequency, short duration, and may display certain periodicity.
  • Blasting events are seismic signals generated by blasting activities in mines or underground projects. The waveform characteristics are usually high amplitude, low frequency, longer duration, and specific spectral features.
  • Rock drilling events, which originate from mechanical activities such as rock gouging and drilling in mines and other sites, produce noisy signals. Although these signals may appear in microseismic monitoring, they are unrelated to seismic activity and need to be distinguished from microseismic events. Their waveform characteristics are generally characterized by high amplitude noise, broad bandwidth, transient signals, and lack of significant periodicity.
  • Other noise events, which refer to noise sources other than rock drilling events, such as unloading, equipment operation, power interference, and groundwater flow, may interfere with the monitoring and identification of MS. The waveform characteristics of these noise events are usually random, irregular, and without obvious frequency and amplitude patterns.
Extracting features from waveform images is a critical step in microseismic event analysis because these features can be used to characterize the event in both the time and frequency domains. These features may include amplitude, frequency, energy distribution, waveform shape, etc. By quantitatively analyzing these features, researchers can establish models for microseismic event recognition and classification [19]. These models can assist mining and underground engineering monitoring systems in more accurately identifying and responding to potential microseismic events, thereby enhancing the safety and sustainability of underground operations. Figure 2 provides a visual reference, emphasizing the diversity of waveform images and underscoring the necessity of feature extraction from these images. By subdividing microseismic events based on different classification criteria, people can gain a more accurate understanding of different types of underground activities and their potential impacts, thus better safeguarding the safety of mines and projects.

This entry is adapted from the peer-reviewed paper 10.3390/app132312739

References

  1. Li, J.; Zhan, K. Intelligent Mining Technology for an Underground Metal Mine Based on Unmanned Equipment. Engineering 2018, 4, 381–391.
  2. Abdalzaher, M.S.; Soliman, M.S.; El-Hady, S.M.; Benslimane, A.; Elwekeil, M. A Deep Learning Model for Earthquake Parameters Observation in IoT System-Based Earthquake Early Warning. IEEE Internet Things J. 2022, 9, 8412–8424.
  3. Arrowsmith, S.J.; Trugman, D.T.; MacCarthy, J.; Bergen, K.J.; Lumley, D.; Magnani, M.B. Big Data Seismology. Rev. Geophys. 2022, 60, e2021RG000769.
  4. Mousavi, S.M.; Beroza, G.C. Machine Learning in Earthquake Seismology. Annu. Rev. Earth Planet. Sci. 2023, 51, 105–129.
  5. Du, J.; Chen, J.; Pu, Y.; Jiang, D.; Chen, L.; Zhang, Y. Risk Assessment of Dynamic Disasters in Deep Coal Mines Based on Multi-Source, Multi-Parameter Indexes, and Engineering Application. Process Saf. Environ. Prot. 2021, 155, 575–586.
  6. Pu, Y.; Apel, D.B.; Liu, V.; Mitri, H. Machine Learning Methods for Rockburst Prediction-State-of-the-Art Review. Int. J. Min. Sci. Technol. 2019, 29, 565–570.
  7. Zhang, C.; Jin, G.; Liu, C.; Li, S.; Xue, J.; Cheng, R.; Wang, X.; Zeng, X. Prediction of Rockbursts in a Typical Island Working Face of a Coal Mine through Microseismic Monitoring Technology. Tunn. Undergr. Space Technol. 2021, 113, 103972.
  8. Cui, F.; Sun, J.; Lai, X.; Jia, C.; Zhang, S. Study on the Energy Release Law of Overburden Rock Breaking and Anti-Rockburst Technology in the Knife Handle Working Face of a Gently Inclined Coal Seam. Appl. Sci. 2023, 13, 11809.
  9. Pu, Y.; Apel, D.B.; Hall, R. Using Machine Learning Approach for Microseismic Events Recognition in Underground Excavations: Comparison of Ten Frequently-Used Models. Eng. Geol. 2020, 268, 105519.
  10. Jiang, R.; Dai, F.; Liu, Y.; Wei, M. An Automatic Classification Method for Microseismic Events and Blasts during Rock Excavation of Underground Caverns. Tunn. Undergr. Space Technol. 2020, 101, 103425.
  11. Duan, Y.; Shen, Y.; Canbulat, I.; Luo, X.; Si, G. Classification of Clustered Microseismic Events in a Coal Mine Using Machine Learning. J. Rock Mech. Geotech. Eng. 2021, 13, 1256–1273.
  12. Kan, J.; Dou, L.; Li, J.; Song, S.; Zhou, K.; Cao, J.; Bai, J. Discrimination of Microseismic Events in Coal Mine Using Multifractal Method and Moment Tensor Inversion. Fractal Fract. 2022, 6, 361.
  13. Wang, C.; Si, G.; Zhang, C.; Cao, A.; Canbulat, I. Variation of Seismicity Using Reinforced Seismic Data for Coal Burst Risk Assessment in Underground Mines. Int. J. Rock Mech. Min. Sci. 2023, 165, 105363.
  14. Wang, R.; Chen, S.; Li, X.; Tian, G.; Zhao, T. AdaBoost-Driven Multi-Parameter Real-Time Warning of Rock Burst Risk in Coal Mines. Eng. Appl. Artif. Intell. 2023, 125, 106591.
  15. Shu, L.; Liu, Z.; Wang, K.; Zhu, N.; Yang, J. Characteristics and classification of microseismic signals in heading face of coal mine: Implication for coal and gas outburst warning. Rock Mech. Rock Eng. 2022, 55, 6905–6919.
  16. Yin, H.; Zhang, G.; Wu, Q.; Yin, S.; Soltanian, M.R.; Thanh, H.V.; Dai, Z. A Deep Learning-Based Data-Driven Approach for Predicting Mining Water Inrush from Coal Seam Floor Using Microseismic Monitoring Data. IEEE Trans. Geosci. Remote Sens. 2023, 61, 4504815.
  17. Dong, L.J.; Wesseloo, J.; Potvin, Y.; Li, X.B. Discrimination of Mine Seismic Events and Blasts Using the Fisher Classifier, Naive Bayesian Classifier and Logistic Regression. Rock Mech. Rock Eng. 2016, 49, 183–211.
  18. Li, L.; Tan, J.; Schwarz, B.; Staněk, F.; Poiata, N.; Shi, P.; Diekmann, L.; Eisner, L.; Gajewski, D. Recent Advances and Challenges of Waveform-Based Seismic Location Methods at Multiple Scales. Rev. Geophys. 2020, 58, e2019RG000667.
  19. Jinqiang, W.; Basnet, P.; Mahtab, S. Review of Machine Learning and Deep Learning Application in Mine Microseismic Event Classification. Min. Miner. Depos. 2021, 15, 19–26.
More
© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.
This entry is offline, you can click here to edit this entry!
Video Production Service
Feedback