Migration-based Detection Of Microseismic Events:Comparison And Optimzation Of Stacking Objective Functions

A microseismic event is defined as an earthquake with a magnitude of less than 3.0and could not be felt by public on the surface.Microseismic monitoring is one of the key technologies for effective exploitation of oil and gas reservoirs.Currently,it has been the most reliable technology for imaging and evaluating the fracture caused by hydraulic fracturing,and be able to monitoring the important parameters accurately and real-time,such as spatial fracture distribution,orientation,and reservoir stress state of the reservoir,which can be the guidance for hydraulic fracturing operating.There are two types of microseismic monitoring: downhole monitoring and surface monitoring.Downhole monitoring is currently the most accurate and reliable monitoring technology.The disadvantage is that large-scale applications are often restricted due to the lack of monitoring wells at appropriate distances.Compared with downhole monitoring,surface microseismic monitoring results in relatively poor quality data.The advantage is that the surface microseismic monitoring has a large monitoring range and low cost,which is suitable for monitoring various complex environments and conditions.At the same time,it is a unique advantage in inversion of the focal mechanism with wide-angle azimuth coverage.In recent years,people have done many researches on surface micro-seismic monitoring technology,it has been proved that surface monitoring can obtain reliable observation data,and micro-seismic events with large magnitudes could be detected,and these events basically meet the requirements of crack characterization,and are monitoring economically.However,we found the problems in processing massive data real-time,identifying weak signal and estimating uncertainty quantitatively when surface microseismic monitoring technology is applied.For solving the current deficiencies and technical difficulties in surface monitoring,we focuses on the research of microseismic data based on surface monitoring in this paper,selecting objective function and doing the optimization of some key parameters.First,three migration stacking methods commonly used in micro-seismic monitoring are chose,including STA/LTA method,Semblance stacking method and cross-correlation coefficient stacking method.By testing synthetic micro-seismic records,the three methods are verified for detecting events effectively,we believe that the STA/LTA method has poor recognition ability for low signal-to-noise ratio seismic events.And then,we picked up a sets of microseismic event signals,did the works on optimizing the key parameters of the three methods;at the same time,The concept of local stacking is proposed for solving the problem of large data and inaccurate velocity model.The 1130 detectors were divided into 113 groups according to the principle of ten detectors in a local range as a group.When a signal can be detected in 50 groups,it means the occurrence of a micro-seismic event;finally,by processing the data of a fracturing section in southwestern Sichuan,three migration stacking methods respectively detect different numbers of micro-seismic events,We found that the STA/LTA method has the highest computational efficiency,and the detectable earthquake magnitude is-3.01,Compared to another two method,Semblance stacking method had a poor preformation on solving the microseismic data.The cross-correlation method could detect a better numbers of microseismic.