Hydraulic fracturing is an important technique in unconventional hydrocarbon recovery,and thus how to monitor its per-formance is crucial to production.Microseismic monitoring,which is a common practice,fails to obtain some useful parameters of the stimulated reservoir.In contrast,magnetotelluric responses are sensitive to low resistivities and could then be used to monitor spatial distribution of fractures.We formulate the 2.5D finite-element equations of controlled-source audio-frequency magnetotellu-rics(CSAMT)and numeric algorithm for CSAMT forward modeling.The algorithm is validated using numerical solutions and an-alytic solutions for two hydraulic fracturing models:one with the emission source close to the wellhead and the other with the e-mission source far away from the wellhead.To examine pre-frac and post-frac variations,we calculate relative apparent resistivity differences between each fractured stage and casing pipe and between two neighboring fractured stages.The results show that the extremum on the relative difference curve will move toward new fractures generated as hydraulic fracturing goes on,and the move-ment correlates with fracture location.This means that hydraulic fractures may be roughly located in accordance with the position of the extremum on the relative difference curve.Pseudo-section map shows sufficiently large apparent resistivity residuals at the positions of fractures,although the center of anomalies is slightly biased toward the side away from the field source instead of be-ing exactly aligned with the center of fractures owing to the influence of the field source;this means that hydraulic fractures may be tentatively located in accordance with apparent resistivity residuals at two neighboring fractured stages.As per the model tests,CSAMT is feasible for hydraulic fracture positioning;our study offers a theoretical approach and technical support to dynamic elec-tromagnetic monitoring of hydraulic fracturing.
frequency domaincontrolled-source magnetotelluricshydraulic fracturingfracturing monitoringforward modelingfracture
国家科技期刊平台
NSTL
维普
万方数据
