Study On Srv Based On Hydraulic Fracturing Simulation And Microseismic Monitoring Information

Tight oil and gas resources constitute a large proportion of unconventional resources.Economic and effective development of tight hydrocarbon reservoirs has become a hot spot.China has abundant resources of tight reservoirs.However,due to their features of low porosity and permeability,hydrocarbon recovery is quite difficult.Hydraulic fracturing is the key technology for the development of tight reservoirs.Economic production from tight formations depends greatly on the effectiveness of the hydraulic fracturing treatment.With the increasing application of microseismic monitoring technology,the positive correlation between stimulated reservoir volume(SRV)and stimulation effectiveness has been fully validated.This has promoted the concept innovation and formed a new concept of fracturing technology aimed at improving the maximum contact area of formation.Therefore,reasonable and effective determination of SRV through microseismic events is of great significance to the evaluation and development of the formation stimulation technique,and also plays an important guiding role in the exploitation of tight reservoirs.For microseismic monitoring,particularly for the location inversion of microseismic events,uncertainties are common under the influences of different data noises.The uncertainties are associated with the collected microseismic data themselves and formation properties such as velocity analysis result.Therefore,large differences are often observed between the fitted SRV from microseismic events and the volume obtained through hydraulic fracture modeling.In view of the above problems,this paper first constructs an improved hydraulic fracturing model based on the fluid flow equations in porous media.The model can simulate the propagation of main hydraulic fractures and fracking fluid leakage from the fracture into the matrix.Compared with the conventional methods which can only capture the main fractures,the improved coupling method can simultaneously simulate the pore pressure changes of the entire formation during fracking process.In the process of fracking,hydraulic fracture will propagate according to the criterion based on stress intensity factor,that is,the stress intensity factor at the fracture tip is greater than the fracture toughness of the rock.This propagation process produces microseismic events centered near the main fractures.In the matrix,when the pore pressure reaches the critical pore pressure,reservoir undergoes microscopic ruptures or slippages and generates microseismic events.In summary,the SRV based on the microseismic events can be determined from the main fractures and the fracturing in the matrix.Usually due to the influence of noises,there is a deviation in the first arrival time received by microseismic geophones.This deviation generates uncertainties for the location inversion of microseismic events.This further leads to a large difference between the computed SRV based on the microseismic information and field monitoring results.In this paper,Gaussian Newton inversion algorithm is applied to the inversion of microseismic events.Different levels of noise(Gaussian noise)are added to the first arrival times during the inversion process.This procedure can reflect the various types of random noise that may be encountered in practice and the overall effects of noise on the location inversion of microseismic events.Due to the influence of noise,the inverted locations of the microseismic events have uncertainties.The outliers in the microseismic point cloud are presented as the obvious isolation with the majority of the microseismic events.From the point of view of physical process,microseismic events are induced by the propagation of the main fracture and the fluid leakage of the fracturing fluid into the matrix.Therefore,the distribution of microseismic points should have a certain degree of density and continuity.Although the reservoir volume stimulated by hydraulic fracturing is so large that the distribution of microseismic event points is widespread,the outliers are obviously isolated from the region where microseismic events are concentrated.Therefore,the outliers are not related to the pressure diffusion in the matrix,and the generation of complex fracture network.This paper aims to reduce the impact of noise on the SRV fitting through removing the outliers based on the concept of outlier.For the forward modeling in the inversion process,this paper applies a finite difference method to solve the Eikonal equation.Compared with the traditional ray tracing methods(such as shooting method and ray bending method),the finite difference method considers the full wavefleld and has its stability,which can effectively overcome the multipath problem and deal with complex formation models.Finally,since the traditional method of computing SRV based on microseismic events is based on the cube calculation,the result can be too optimistic.Therefore,a more reasonable three-dimensional Delaunay triangulation algorithm,Minimum Volume Enclosing Ellipsoid(MVEE)fitting algorithm and the more conservative two discrete bin methods to fit the SRV.The SRV fitting methods are used for the microseismic events inverted from the first arrival times with noise.The fitting results are compared and analyzed with the SRV obtained from the hydraulic fracturing simulation.The results show that the large effects of removing outliers.After removing the outliers in the microseismic events with noise,the SRV obtained by the two discrete bin algorithms are not different from those of hydraulic fracturing model,which indicates that although the two methods are more conservative,but with good stability(robustness)and low sensitivity to noise.On the other hand,the MVEE and 3D Delaunay triangulation algorithms are not robust and can be easily affected by the noise in the observed data.Both approaches,however,clearly delineate not only the areas generating microseismic events from the mathematical realm,but also the more detailed and quantitative SRV geometry.Therefore,in practice a combination of multiple fitting algorithms can be helpful for evaluating the fracturing effectiveness and predicting the oil and gas production.Combining with hydraulic fracturing and microseismic location inversion,this paper presents a method framework to compare the stability and effectiveness of each SRV fitting algorithm.