Abstract
Deep neural networks have achieved great success in microseismic event recognition tasks in rock engineering. However, a well-trained model on one project may not deliver satisfactory results on other new projects due to the different geological conditions, construction methods and monitoring configurations. In this study, the differences of microseismic monitoring in four monitoring project cases, including the geology conditions and monitoring settings, are compared to better understand the challenges of signal recognition by deep learning, and then the drawbacks and unavailability of current algorithms for microseismic monitoring scenarios are discussed. To address the challenges, a unified transfer learning strategy is proposed to perform knowledge transfer for microseismic recognition between multiple projects with varying difficulties, and it consists of a novel unsupervised two-level adversarial adaptation method and a semisupervised active learning mechanism; thus, it operates in a semisupervised fashion. The results experimentally demonstrate that a well-trained model cannot generalize well to other rock projects without performing transfer learning. This framework is evaluated on four real-world collections of monitoring recordings and compared with other benchmark methods. The results show that it outperforms all counterparts and is able to conduct knowledge transfer with varying difficulties. Moreover, the proposed active learning mechanism significantly reduces the manual labeling costs for both hard and easy tasks. The simulations suggest that we have preliminarily achieved deep model reuse in signal recognition of microseismic monitoring.
Highlights
-
The transferability and vulnerabilities of the existing deep neural networks for microseismic event identification between different monitoring projects are discussed.
-
A novel two-level adversarial domain adaptation method is proposed to perform transfer learning in unsupervised setting.
-
A “performance sacrifice” method is proposed to improve algorithm design and explain models’ performance.
-
An active learning mechanism is implemented to greatly reduce manual labelling cost.
-
A unified transfer learning strategy is proposed to perform knowledge transfer for microseismic event recognition of different underground construction projects with varying difficulties.
Similar content being viewed by others
References
Ajakan H, Germain P, Larochelle H, Laviolette F, Marchand M (2014) Domain-adversarial neural networks. arXiv e-prints https://arxiv.org/abs/1412.4446
Bi X, Zhang C, He Y, Zhao X, Sun Y, Ma Y (2021) Explainable time–frequency convolutional neural network for microseismic waveform classification. Inf Sci 546:883–896. https://doi.org/10.1016/j.ins.2020.08.109
Girshick R, Donahue J, Darrell T, Malik J (2013) Rich feature hierarchies for accurate object detection and semantic segmentation. In: Conference on Computer vision and pattern recognition, Columbus, OH, USA, pp 580–587
Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. In: 27th International Conference on neural information processing systems, Montreal, Canada, pp 2672–2680
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. 2015 IEEE International Conference on computer vision. IEEE Computer Society, USA, pp 1026–1034
Isola P, Zhu JY, Zhou T, Efros AA (2017) Image-to-image translation with conditional adversarial networks. In: 30th Conference on computer vision and pattern recognition, Honolulu, HI, USA, pp 5967–5976
Kading C, Rodner E, Freytag A, Denzler J (2017) Fine-tuning deep neural networks in continuous learning scenarios. In: 13th Asian Conference on computer vision. Springer Verlag, Taipei, Taiwan, pp 588–605
Karras T, Laine S, Aila T (2020) A style-based generator architecture for generative adversarial networks. IEEE Trans Pattern Anal Mach Intell 99:1
Liang Z, Xue R, Xu N, Li W (2020) Characterizing rockbursts and analysis on frequency-spectrum evolutionary law of rockburst precursor based on microseismic monitoring. Tunn Undergr Space Technol. https://doi.org/10.1016/j.tust.2020.103564
Long M, Wang J (2015) Learning transferable features with deep adaptation networks. arXiv e-prints https://arxiv.org/abs/1502.02791
Ma TH, Tang CA, Tang LX, Zhang WD, Wang L (2015) Rockburst characteristics and microseismic monitoring of deep-buried tunnels for Jinping II Hydropower Station. Tunn Undergr Space Technol 49:345–368. https://doi.org/10.1016/j.tust.2015.04.016
Perol T, Gharbi M, Denolle M (2018) Convolutional neural network for earthquake detection and location. Sci Adv 4(2):e1700578
Qian B, Xu N, Xiao P, Zhao T, Dong L (2019) Damage analysis and deformation early warning of surrounding rock mass during top arch excavation of underground powerhouse of Shuangjiangkou hydropower station. Chin J Rock Mech Eng 38(12):2512–2524. https://doi.org/10.13722/j.cnki.jrme.2019.0476
Ross ZE, Meier M-A, Hauksson E (2018) PWave arrival picking and first-motion polarity determination with deep learning. J Geophys Res Solid Earth 123(6):5120–5129. https://doi.org/10.1029/2017jb015251
Salakhutdinov R, Hinton G (2009) Semantic hashing. Int J Approx Reason 50(7):969–978
Sun B, Saenko K (2016) Deep CORAL: correlation alignment for deep domain adaptation. In: 14th European Conference on computer vision. Springer Verlag, Amsterdam, Netherlands, pp 443–450
Sun B, Feng J, Saenko K (2016) Return of frustratingly easy domain adaptation. In: 30th AAAI Conference on artificial intelligence. AAAI Press, Phoenix, Arizona, pp 2058–2065
Tang S, Wang J, Tang C (2020) Identification of microseismic events in rock engineering by a convolutional neural network combined with an attention mechanism. Rock Mech Rock Eng 54(1):47–69. https://doi.org/10.1007/s00603-020-02259-0
Tzeng E, Hoffman J, Zhang N, Saenko K, Darrell T (2014) Deep domain confusion: maximizing for domain invariance. arXiv e-prints arXiv: 14123474.
Tzeng E, Hoffman J, Darrell T, Saenko K (2015) 15th simultaneous deep transfer across domains and tasks. In: International Conference on Computer Vision, Santiago, Chile, pp 4068–4076.=
Tzeng E, Hoffman J, Saenko K, Darrell T (2017) Adversarial discriminative domain adaptation. In: IEEE Conference on computer vision and pattern recognition (CVPR), Honolulu, HI, USA, pp 2962–2971
Wang JX, Tang SB, Heap MJ, Tang CA, Tang LX (2021) An auto-detection network to provide an automated real-time early warning of rock engineering hazards using microseismic monitoring. Int J Rock Mech Min Sci 140:104685. https://doi.org/10.1016/j.ijrmms.2021.104685
Wilkins AH, Strange A, Duan Y, Luo X (2020) Identifying microseismic events in a mining scenario using a convolutional neural network. Comput Geosci 137:104418. https://doi.org/10.1016/j.cageo.2020.104418
Xu NW, Dai F, Liang ZZ, Zhou Z, Sha C, Tang CA (2014) The dynamic evaluation of rock slope stability considering the effects of microseismic damage. Rock Mech Rock Eng 47(2):621–642. https://doi.org/10.1007/s00603-013-0432-5
Xu NW, Li TB, Dai F, Li B, Zhu YG, Yang DS (2015) Microseismic monitoring and stability evaluation for the large scale underground caverns at the Houziyan hydropower station in Southwest China. Eng Geol 188:48–67
Yosinski J, Clune J, Bengio Y, Lipson H (2014) How transferable are features in deep neural networks? In: 27th International Conference on neural information processing systems, Montreal, Canada, pp 3320–3328
Zhu JY, Park T, Isola P, Efros AA (2017) Unpaired image-to-image translation using cycle-consistent adversarial networks. In: IEEE International Conference on computer vision, Venice, Italy, pp 2242–2251
Acknowledgements
This work was supported by the National Natural Science Foundation of China (51874065 and U1903112).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, J., Tang, S. Novel Transfer Learning Framework for Microseismic Event Recognition Between Multiple Monitoring Projects. Rock Mech Rock Eng 55, 3563–3582 (2022). https://doi.org/10.1007/s00603-022-02790-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00603-022-02790-2