Font Size: a A A

Research On Weak Signal Feature Model Of Coalbed Methane Hydraulic Fracturing Based On Microseismic Technology

Posted on:2021-05-28Degree:MasterType:Thesis
Country:ChinaCandidate:G ChenFull Text:PDF
GTID:2480306302969149Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
In recent years,microseismic monitoring technology has been widely used in the exploitation of oil and shale gas.Microseismic monitoring technology is to receive the signals of geophones placed in adjacent wells or on the ground,and locate microseismic events based on these signals,so as to monitor and detect the hydraulic fracturing process.Through the research of various current microseismic source location methods and algorithms,the error is small between 500 meters and 1000 meters underground.In this depth range,the distance between the source and the geophone is not very far,and during the period of seismic wave transmission and reception If the signal attenuation rate is kept constant,then the seismic wave transmission distance is proportional to the loss.Use deep learning multi-layer convolutional neural networks to build a signal processing model,use unprocessed hydraulic fracturing seismic signals to train the model,and use artificially processed hydraulic fracturing seismic signals to test the model to continuously improve the accuracy of the model and let it It can accurately realize the twoclassification of noise and seismic signals,identify the phase of seismic waves,and accurately locate rock rupture points through seismic source location algorithms according to seismic wave propagation time,instead of manual labeling,improving source location accuracy,saving costs,and improving efficiency.Establish a microseismic ground monitoring system and conduct field experiments.Manually mark the seismic phase,wave arrival time and source location of the hydraulic fracturing seismic data monitored on the ground.At the same time,these data will be transmitted to the model for seismic phase and wave arrival time marking and Model simulation of seismic source location,and finally compare the model simulation experiment results with field experiment data.
Keywords/Search Tags:Coalbed methane mining, Hydraulic fracturing, Microseismic, Deep learning, Convolutional neural network
PDF Full Text Request
Related items