Imaging Discrete Fracture Network using the Location and the Moment Tensor of Microseismic Events

Imaging Discrete Fracture Network using the Location and the Moment Tensor of Microseismic Events
Other Titles
미소진동 위치정보 및 모멘트 텐서를 이용한 개별 균열망 영상화
Jeongmin Yu
Alternative Author(s)
Issue Date
Hydraulic fracturing is widely used in the development of shale gas and enhanced geothermal system (EGS). Multi-stage fracture treatments are used to re-activate complex fracture networks that are used for the investigation of the flow path or the prediction of production. Also, characterization of fractures is required for non‐invasive site investigation at radioactive waste storage/disposal sites. Microseismic monitoring which provides the information of the locations of the microseismic events and their moment tensor is most commonly used to gain a better understanding of the behavior of fractures. The location of microseismic events can be used in estimations of the geometry and distribution of fractures and faults, and the moment tensor provides information on the orientation of fracture planes and event magnitudes. Therefore, various methods were applied in the derivation of the discrete fracture networks (DFN), which shows the geometries of complex fractures using planar polygons, from microseismic data. Past researches on imaging DFN took lots of time due to the transformation of the coordinate or could not depict the orientation of fracture planes accurately because only microseismic event locations were used. In this dissertation, an effective fracture networks imaging method using the microseismic event locations and the moment tensor has been developed. For this, the conventional RANSAC method was improved upon to detect fracture plane quickly and efficiently in 3D point cloud data. In addition, through the application of DBSCAN clustering based on data density to microseismic location data in pre-processing step, the accuracy of the locations of fracture planes were increased. Furthermore, the orientations of fracture planes can also be accurately obtained by applying DBSCAN clustering to the selection of inliers during RANSAC process using moment tensor information To assess the performance of the proposed method, the developed algorithm was applied to synthetic microseismic data included errors in the event locations and the noise. The results demonstrate that the developed method can reliably extract the fracture planes almost identical with the true model. In addition, the effect of the input parameters used in the developed algorithm were analyzed and compared with the conventional RANSAC algorithm. Furthermore, to investigate the field applicability of the proposed method, it was applied to the microseismic data acquired in the shale gas / oil play with two clustering methods (stage and DBSCAN). The results of both the methods depicted the dominant strike of fracture planes accurately.
Appears in Collections:
Files in This Item:
There are no files associated with this item.
RIS (EndNote)
XLS (Excel)


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.