Journal of Radio Electronics. eISSN 1684-1719. 2026. ¹7
Full text in Russian (pdf)
DOI: https://doi.org/10.30898/1684-1719.2026.7.8
OF MICROSEISMIC EVENTS DETECTION
IN DAS DATA BASED ON A HYBRID APPROACH
B.V. Yemelyanov 1, O.N. Sherstyukov 1, V.A. Ryzhov 2
1 Kazan (Volga Region) Federal University, 420008, Kazan, Kremlevskaya St., 18
2 Prizmageo LLC, 42003, Kazan, st. Serova, 51/11
The paper was received June 5, 2026.
Abstract. This paper develops an automated method for detecting and spatially localizing microseismic events in borehole distributed acoustic sensing data. The objective of the study is to improve the robustness of weak event detection in the presence of intense nonstationary and spatially correlated interference. The methodological framework includes the creation of a database of synthetic microseismic signals with variable source parameters and signal-to-noise ratios, followed by their integration into real noise records to simulate field conditions. For an initial assessment of the spatiotemporal source parameters, a localization procedure based on the maximum likelihood function was implemented, constructing three-dimensional arrays reflecting the distribution of the energy response in depth-radius-time coordinates. Using these arrays, an encoder-decoder neural network model was developed and trained. This model is designed to refine event coordinates and suppress the influence of complex noise structures. Training and testing were performed on training, validation, and independent test sets using regression and classification quality metrics. The obtained results demonstrate a reduction in the mean absolute and standard deviations of localization coordinates, as well as an increase in classification consistency compared to the use of likelihood function thresholding alone. The proposed hybrid approach is shown to increase the detection probability of weak microseismic events while reducing the number of false alarms in real borehole noise conditions.
Key words: DAS, microseismic monitoring, MLE, U-Net.
Corresponding author: Bulat Vladimirovich Emelyanov, emelyanov.bulat.91@mail.ru
References
1. Hartog A.H. An introduction to distributed optical fibre sensors. – Boca Raton : CRC Press, 2017. – 440 p.
2. Lindsey N.J., Rademacher H., Ajo-Franklin J.B. On the broadband instrument response of DAS systems // Journal of Geophysical Research: Solid Earth. – 2020. – Vol. 125, No. 2. – P. e2019JB018145.
3. Mateeva A., Lopez J., Potters H., Mestayer J., Cox B., Kiyashchenko D., Wills P., Grandi S. Distributed acoustic sensing for reservoir monitoring with vertical seismic profiling // Geophysical Prospecting. – 2014. – Vol. 62, No. 4. – P. 679–692.
4. Karrenbach M., Cole S., Ridge A., Boone K., Kahn D. Fiber-optic distributed acoustic sensing of microseismicity // The Leading Edge. – 2019. – Vol. 38, No. 8. – P. 636–642.
5. Li Z., Zhan Z. Pushing the limit of earthquake detection with DAS // Geophysical Journal International. – 2020. – Vol. 221, No. 2. – P. 1343–1359.
6. Daley T.M., Miller D.E., Dodds K., Cook P., Freifeld B. Field testing of distributed acoustic sensing for subsurface monitoring // Geophysics. – 2013. – Vol. 78, No. 2. – P. O1–O13.
7. Allen R.V. Automatic earthquake recognition and timing from single traces // Bulletin of the Seismological Society of America. – 1978. – Vol. 68, No. 5. – P. 1521–1532.
8. Neidell N.S., Taner M.T. Semblance and other coherency measures for multichannel data // Geophysics. – 1971. – Vol. 36, No. 3. – P. 482–497.
9. Tarantola A. Inverse problem theory and methods for model parameter estimation. – Philadelphia : SIAM, 2005. – 342 p.
10. Yu P., Zhang H., Li X., Chen Y. DASEventNet: AI-based microseismic detection using DAS data // IEEE Transactions on Geoscience and Remote Sensing. – 2024. – Vol. 62. – Art. 5500915.
11. Wamriew D., Saad O.M., Kuleli H.S. Deep neural networks for microseismic detection and location // Sensors. – 2021. – Vol. 21, No. 3. – P. 830.
12. Perol T., Gharbi M., Denolle M. Convolutional neural network for earthquake detection // Science Advances. – 2018. – Vol. 4, No. 2. – eaao1048.
13. Zhu W., Beroza G.C. PhaseNet: a deep-neural-network-based seismic arrival-time picking method // Geophysical Journal International. – 2019. – Vol. 216, No. 1. – P. 261–273.
14. Birialtsev E., Demidov D., Mokshin E. Determination of moment tensor and location of microseismic events under conditions of highly correlated noise based on the maximum likelihood method // Geophysical Prospecting. – 2017. – Vol. 65, No. 1. – P. 1–17.
For citation:
Yemelyanov B.V., Sherstyukov O.N., Ryzhov V.A. Increasing the noise immunity of microseismic events detection in DAS data based on a hybrid approach of MLE and the U-Net model // Journal of Radio Electronics. – 2026. – ¹. 7. https://doi.org/10.30898/1684-1719.2026.7.8 (In Russian)