Objective To achieve safe and efficient hydraulic fracturing, it is necessary to conduct timely and accurate microseismic monitoring in the reservoir stimulation process. The purpose is to analyze and assess the fracturing effects.

Methods Using a permanently deployed fiber-optic cable outside the casing of a coalbed methane (CBM) well, this study employed distributed optical fiber acoustic sensing (DAS)for full-borehole, wide-azimuth, and high-density real-time monitoring of microseismic events induced by the hydraulic fracturing of coal seams. Furthermore, this study proposed a novel method for localizing microseismic sources based on information about wave travel time and a source-constrained mapping algorithm. First, effective microseismic events were identified by calculating the interchannel similarity coefficient from optical fiber data. Second, the incidence locations of microseismic sources in the fiber were obtained based on the P- and S-wave peaks. Third, the wave travel time differences derived using the short term averaging/long term averaging (STA/LTA) algorithm were used to determine the spatial distances between microseismic events and the fiber. Finally, the constrained mapping and localization of microseismic sources were performed by combining the incidence locations in the fiber and the distances from microseismic sources to the monitoring fiber.

Results and Conclusion  The results indicate that DAS exhibited a strong directional sensitivity to seismic waves in the acquisition process of optical fiber data. Specifically, DAS was sensitive to P-waves with an incidence angle of 0° and S-waves with an incidence angle of 45°but almost insensitive to P-waves with an incidence angle of 90° and S-waves with an incidence angle of 90°. In the case of consistent noise intensity, the sampling effects in the seismic wavefield based on various gauge lengths exhibited decreasing significance in the order of 1 m, 10 m, and 100 m. In the case of a noise intensity of 300%, the spatial sampling data based on a gauge length of 1 m still displayed relatively continuous waveforms, highlighting the seismic source wavelets of 50 Hz. The predicted fracture networks primarily exhibited NNW-SSE-directed distributions, aligning with the direction of the regional maximum principal stress, and prediction errors of within 10 m mostly. The application in wellfield M of the Chuannan coalfield in the Sichuan Basin indicates that the constrained localization algorithm yielded prediction results consistent with the experimental verification results and can effectively indicate the fracture network zones of coal seams. Therefore, this study holds great significance for enhancing the exploitation efficiency of tight hydrocarbon reservoirs with low porosity and permeability.