Calibration Of Microseismic Velocity Model Based On Parallel Method

At present,the proportion of low permeability oil and gas production in China is increasing year by year.Low permeability oil and gas will be the main target of China's oil and gas development in the future.Horizontal drilling technology and fracturing reservoir reconstruction technology are the key engineering technologies for low permeability oil and gas development to obtain high yield.However,the effect of fracturing cannot be evaluated at the time of fracturing,and can only be evaluated through the later development,which to some extent increases the cost of fracturing.In order to solve this problem and realize real-time evaluation of fracturing effect,microseismic monitoring technology has been put forward and developed.Microseismic monitoring technology is a technology that USES hydraulic fracturing,oil and gas production or conventional petroleum engineering operations(such as water injection,gas injection and thermal flooding)to cause changes in the underground stress field,resulting in seismic waves generated by rock fractures or faults,fracturing fracture imaging or monitoring reservoir fluid movement,which started in the 1970 s.Microseismic monitoring has unique advantages in judging fracturing effect in real time and adjusting fracturing plan.It is an effective means to improve the development effect of tight,low permeability,shale gas and other complex reservoirs.Microseismic monitoring is mainly associated with oil and gas field exploitation of oil and gas field development produces the crack monitoring,when the development process of the hydraulic fracturing cracks in a series of micro seismic events,monitoring the micro seismic events,fixing the seismic source and observed the change over time,is one of the main purpose of the micro seismic monitoring.In the process of source location,micro-seismic monitoring is usually affected by many factors,among which the velocity model error has the most significant impact.The initial velocity model based on acoustic logging data has a large error and is not suitable for seismic source location.In the process of conventional velocity model correction processing,the time is often too long,which is very unfavorable for the realization of real-time micro-seismic monitoring.How to improve the speed of velocity model correction,so as to improve the efficiency of microseismic monitoring data processing,research on micro-seismic monitoring is also very necessary.In this paper,based on the data of the first to last time difference obtained from the underground micro-seismic monitoring,the correction method of the micro-seismic velocity model is studied.It can effectively improve the efficiency of velocity model correction and meet the need of real-time microseismic monitoring.In this paper,focusing on the research purpose and significance,the correlation analysis of microseismic monitoring technology at home and abroad is carried out,and the correlation method of correction of microseismic velocity model is summarized.In this paper,forward numerical simulation and inversion velocity model correction are studied.For the forward numerical simulation part,the forward numerical simulation method based on ray theory is studied in depth.Starting from the principle of several main ray tracing methods,and considering the requirements of forward calculation for computing speed and accuracy,a fast ray tracing algorithm between two points for horizontal layered media is selected.For the correction part of the inversion velocity model,first of all,for the nonlinear problem related inversion method,to understand several common nonlinear inversion algorithm,and mainly for the simulated annealing method for in-depth research,from the simulated annealing method principle,process and algorithm structure of the relevant discussion;Then,on the basis of simulated annealing method,it proposes to improve it in parallel,and finally chooses the method of multiple markov chains to improve the simulated annealing method by comparing and analyzing several different parallel schemes,so as to improve the computational efficiency.In the trial calculation of synthetic data,the parallel simulated annealing method of multiple markov chains was tested under the conditions of different number of threads and different markov chain lengths.Finally,the velocity model of a certain area is corrected by the perforating data of downhole microseismic monitoring,and the re-location of the perforation is carried out according to the correction results.From the positioning results and calculation time,perforation positioning based on the corrected velocity model can be well positioned near the real seismic source,and the calculation time can also meet the demand of real-time microseismic monitoring,which proves the feasibility of the velocity model correction method in this paper.