| Hydraulic fracturing is an industrial technology of underground reservoir stimula-tion that creates a network of interconnected fractures by injecting high-pressure fluid into the reservoir to increase permeability and enhance fluid flow.Hydraulic fractur-ing technology is currently widely used to develop low-permeability unconventional oil and gas resources,such as shale gas and coalbed methane.During the hydraulic fracturing process,the generation of fractures is usually accompanied by the activity of microseismic.Moreover,due to the migration of fluids and the stress disturbances gen-erated by the fracturing process itself,fault activation is caused on the existing faults around the reservoir and even in the distant regions,and earthquakes larger than the normal microseismic energy can be induced.Generally,an earthquake with a relatively small magnitude caused directly by hydraulic fracturing is called a "microseismic";an earthquake that is larger than the normal microseismic energy due to fault activation caused by hydraulic fracturing is called "induced earthquake".Microseismic and in-duced earthquake monitoring utilizes arrays of geophones deployed on surface or in deep wellbores.Due to the presence of fractures and fluids,hydraulic fracturing stim-ulates the target reservoir and its surrounding media,and the location of microseismic and induced earthquakes can be used to describe the distribution of fractures and faults.Therefore,earthquake location and seismic velocity tomography are of great signifi-cance for the development of hydraulic fracturing microseismic and induced earthquake monitoring of shale gas.This study mainly focuses on developing advanced earthquake location and seis-mic tomography methods suitable for surface and downhole earthquake monitoring,and studying the effects of hydraulic fracturing to reservoir velocity structures using surface and downhole microseismic monitoring dataset,respectively.The main con-tents include:(1)Development of real-time earthquake location method and real-time seismic velocity tomography method based on Kalman filter frameworkFor linear earthquake location and seismic velocity tomography systems,they can be transformed into Kalman-filtering-like systems,ie for current earthquake location and seismic velocity models,seismic locations,velocities and their uncertainties can be iteratively updated,after receiving new arrival time data.For the real-time seismic localization algorithm based on Kalman filter framework,the effectiveness of the algo-rithm is verified based on synthetic dataset and real dataset.From the application results of the 2003 San Simeon Mw 6.6 earthquake and the 2004 Parkfield Mw 6.0 earthquake,the Kalman filter real-time location method can update the earthquake location in real time,which is of great significance for real-time monitoring and early warning of earth-quakes.For the real-time seismic velocity tomograohy algorithm,the test based on the two-dimensional synthetic dataset shows that compared with the traditional traveltime tomography method,the new algorithm can quickly determine the change of the ve-locity structure in near real time,and can simultaneously obtain the uncertainty of the velocity model.In the case of assuming a known velocity change region,the method can also image only a specific region by constructing a model covariance matrix containing only information corresponding to the velocity anomaly region,which can effectively avoid artifacts in the inversion process.(2)Development of azimuth-constrained double-difference earthquake location and seismic tomography method for downhole microseismic monitoringFor deep wellbores,especially single-well microseismic monitoring,it is impossi-ble to determine the location of the earthquake using only the information of the earth-quake arrival time.Based on the traditional double-difference earthquake location and seismic tomography method,we added the azimuth information of the microseismic events to constrain the earthquake location,that is,keeping the azimuth of the relo-cations of microseismic events correct.The new method is applicable to single well microseismic monitoring dataset,and the real dataset test results show the effectiveness of the method.(3)Earthquake relocation and seismic tomography for the surface microseismic monitoring dataset from the Fox Creek shale gas development in Alberta,CanadaThis study area is an important shale gas development area in Canada,and the hy-draulic fracturing activities in this area induced some earthquake events with relatively large magnitude(Mw>0).For the combined monitoring dataset of shallow boreholes and surface monitoring microseisms in four horizontal fracturing wells,we refined the induced earthquakes based on an improved double-difference imaging algorithm and jointly inverted Vp,Vs and Vp/Vs near the target reservoir.After joint inversion,we obtained a more accurate seismic location and were able to characterize the fault better.Combining the research results of predecessors in this frac-turing area with the results of seismic tomography and earthquake relocations obtained by our joint inversion,we believe that in the process of hydraulic fracturing,upwardly expanding vertical fractures are generated,which are connected to each other,resulting in horizontal seismicity.The upward extension of the fracture network and the existing faults on the west side of the fracturing well form a gas migration channel,which causes the gas in the reservoir to migrate upward and to the west to the pre-existing fault,thus forming a low Vp,high Vs and low Vp/Vs anomalies.For the seismic events at the northern and eastern sides of the fracturing well,there is no channel for gas migration,only cracks and faults exist,resulting in low Vp and low Vs anomalies.(4)Single well microseismic relocation and seismic tomography for hydraulic frac-turing developmentThe dataset of downhole microseismic monitoring in a shale gas hydraulic frac-turing was processed using a newly developed azimuth-constrained double-difference earthquake location and seismic tomography algorithm.The results show that the re-location results obtained by our new method can better characterize the fracture dis-tribution.In this study,we also identified two pre-existing faults that were activated sequentially during the fracturing process due to increased fluid pore pressure in the fault zone.The results of velocity tomography not only clearly show the low velocity anomalies associated with the pre-exsiting fault zone,but also show that the microseis-micity caused by hydraulic fracturing correspond to low velocity anomalies around the fracturing zone due to the presence of fractures and fluids.Therefore,we can use the areas of low Vp and low Vs anomalies to estimate the stimulated reservoir volume.The new way of SRV estimation does not depend on the completeness of microseismic events and thus can provide a more accurate and less biased estimation.The real-time seismic tomography algorithm based on the Kalman filter frame-work is also applied to the four-stage hydraulic fracturing monitoring dataset to study the effect of each segment of fracturing activity on the subsurface medium.The re-sults show that the changes in Vp and Vs are in good agreement,and the low-velocity anomalies caused by the first two stages of fracturing are significantly different from the velocity anomalies caused by the latter two stages of fracturing.The low velocity anomalies generated by the first two stages of fracturing activities may be related to the pre-exsiting faults that were reactivated during this two stages of fracturing,while the results of the latter two stages of fracturing are more likely to directly correspond to the fractures produced by fracturing. |
PDF Full Text Request