Quantitative evaluation of fracture porosity from dual laterlog based on deep learning method

Fracture porosity is one of the key parameters for characterizing fractured reservoirs.However,fracture porosity calculation is difficult with conventional logging data due to severe anisotropy of the reservoirs.To deal with the problem,the equivalent macroscopic anisotropic formation model based on dual laterolog(DLL)data is adopted to cyclically assign such parameters as bedrock resistivity(RB),fluid resistivity in fractures(RFL),fracture dip angle(FDA)and fracture thickness as well as fracture spacing,and to produce massive data for formation modeling.A large number of training data obtained through three dimensional finite element forward modeling and the functional relationship between DLL responses and fracture parameters that are trained and summarized by deep neural network,are combined to establish a new fast forward model for calculating DLL responses in fractured formations.A new fracture porosity inversion model for fractured reservoirs based on gradient optimization inversion algorithm combined with multi-initial inversion strategy is then proposed.While running the model,formation is divided into eight intervals according to bedrock resistivity and fracture dip angle from 0°to 90°is divided every 0.5°to improve the operation speed and efficiency.The results of numerical verification show that when bedrock resistivity is greater than 1000Ωm,the mean absolute error(MAE)of fracture porosity inversion is 0.001658%for horizontal fractures,0.00413%for intermediate fractures and 0.0027%for quasi-vertical fractures.When bedrock resistivity is between 100Ωm and 1000Ωm,MAE of fracture porosity inversion is 0.003%for horizontal fractures,0.0034%for intermediate fractures and 0.00348%for quasi-vertical fractures.Fracture parameters determined by the fracture porosity inversion model with actual data are in good agreement with the results of micro resistivity imaging logging.

How high can fracture porosity become in the ultra-deep subsurface?

Knowledge of how high can fracture porosity become in the ultra-deep burial conditions is important but remains problematic.Fracture aperture and porosity are measured using X-ray computed tomography(CT)at atmospheric pressure and then calculated by image logs.Special attention is paid to how high fracture porosity can become in ultra-deep(>6000 m)settings,and which situations will result in high fracture porosities.In situ stress magnitudes,which can be calculated using well logs,control fracture performances,and dissolution along fracture improve fracture porosity at ultra-deep burial depths.Low horizontal stress difference(Dr<25 MPa),very high fracture density will result in a high fracture porosity.Fracture porosity can keep as high as 2.0%in relatively low in situ stress conditions even at ultra-deep burial depths.In intense in situ stress conditions(Dr>45 MPa),a high degree of dissolution along the fracture dramatically increases fracture porosity.Dissolution will result in the vuggy fracture planes and improve fracture porosity up to 2.0%.The results provide insights into the detection,characterization,and modeling of subsurface fractures.

Response characteristics of array lateral logs and their primary inversion in reservoirs with fracture-induced anisotropy

In order to identify fractured reservoirs and determine their fracture parameters with a high definition array laterolog,we built a fracture-induced anisotropic formation model with a parallel fracture group.The three-dimensional finite element method is used to simulate the responses of the array laterolog,and then the primary inversion method is utilized.Numerical simulation shows that when the fracture spacing is small,the array laterolog response of the fracture group is the same as that of a formation with macroscopic electrical anisotropy.The apparent resistivity of the array laterolog is approximately inversely proportional to fracture porosity.The anisotropy depends on the fracture porosity in the fractured formation,which accordingly results in response variation of the array laterolog.The higher the fracture dip,the larger the apparent resistivity.When the fracture dip is low the difference between the deep and shallow apparent resistivities is small,and when the dip is high the difference turns out to be positive.The fracture parameters were inverted using the Marquardt non-linear least squares method.The results,both fracture porosity and dip show a good match with parameters in the actual formation model.This will promote the application of the array laterolog in evaluating fractured reservoirs.

A numerical approach for pressure transient analysis of a vertical well with complex fractures

A new well test model for a vertical fractured well is developed based on a discrete-fracture model in which the fractures are discretized as one dimensional（1-D） entities.The model overcomes the weakness of complex meshing,a large number of grids, and instability in conventional stripe-fracture models. Then, the discrete-fracture model is implemented using a hybrid element finite-element method.Triangular elements are used for matrix and line elements for the fractures. The finite element formulation is validated by comparing with the semi-analytical solution of a single vertical fractured well. The accuracy of the approach is shown through several examples with different fracture apertures,fracture conductivity, and fracture amount. Results from the discrete-fracture model agree reasonably well with the stripefracture model and the analytic solutions. The advantages of the discrete-fracture model are presented in mesh generation, computational improvement, and abilities to handle complex fractures like wedge-shaped fractures and fractures with branches. Analytical results show that the number of grids in the discrete-fracture model is 10 % less than stripefracture model, and computational efficiency increases by about 50 %. The more fractures there are, the more the computational efficiency increases.

Review of permeability evolution model for fractured porous media

The ability to capture permeability of fractured porous media plays a significant role in several engineering applications, including reservoir, mining, petroleum and geotechnical engineering. In order to solve fluid flow and coupled flow-deformation problems encountered in these engineering applications,both empirical and theoretical models had been proposed in the past few decades. Some of them are simple but still work in certain circumstances; others are complex but also need some modifications to be applicable. Thus, the understanding of state-of-the-art permeability evolution model would help researchers and engineers solve engineering problems through an appropriate approach. This paper summarizes permeability evolution models proposed by earlier and recent researchers with emphasis on their characteristics and limitations.

Integrated petrophysical log characterization for tight carbonate reservoir effectiveness: A case study from the Longgang area, Sichuan Basin, China

Ultra-low porosity and permeability, inhomogeneous fracture distribution, and complex storage space together make the effectiveness evaluation of tight carbonate reservoirs difficult. Aiming at the carbonate reservoirs of the Da＇anzhai Formation in the Longgang area of the Sichuan Basin, based on petrophysical experiments and logging response characteristics, we investigated the storage properties of matrix pores and the characteristics of fracture development to establish a method for the characterization of effectiveness of tight reservoirs. Mercury injection and nuclear magnetic resonance （NMR） experiments show that the conventional relationship between porosity and permeability cannot fully reflect the fluid flow behavior in tight matrix pores. Under reservoir conditions, the tight reservoirs still possess certain storage space and permeability, which are controlled by the characteristic structures of the matrix porosity. The degree of fracture development is crucial to the productivity and quality of tight reservoirs. By combining the fracture development similarity of the same type of reservoirs and the fracture development heterogeneity in the same block, a three-level classification method of fracture development was established on the basis of fracture porosity distribution and its cumulative features. According to the actual production data, based on the effectiveness analysis of the matrix pores and fast inversion of fracture parameters from dual laterolog data, we divided the effective reservoirs into three classes： Class I with developed fractures and pores, and high-intermediate productivity; Class II with moderately developed fractures and pores or of fractured type, and intermediate-low productivity; Class III with poorly developed fractures and matrix pores, and extremely low productivity. Accordingly log classification standards were set up. Production data shows that the classification of effective reservoirs is highly consistent with the reservoir productivity level, providing a new approach for the effectiveness evaluation of tight reservoirs.

Key geological factors controlling the estimated ultimate recovery of shale oil and gas: A case study of the Eagle Ford shale, Gulf Coast Basin, USA

Based on 991 groups of analysis data of shale samples from the Lower Member of the Cretaceous Eagle Ford Formation of 1317 production wells and 72 systematic coring wells in the U.S. Gulf Basin, the estimated ultimate recovery(EUR) of shale oil and gas of the wells are predicted by using two classical EUR estimation models, and the average values predicted excluding the effect of engineering factors are taken as the final EUR. Key geological factors controlling EUR of shale oil and gas are fully investigated. The reservoir capacity, resources, flow capacity and fracability are the four key geological parameters controlling EUR. The storage capacity of shale oil and gas is directly controlled by total porosity and hydrocarbon-bearing porosity, and indirectly controlled by total organic carbon(TOC) and vitrinite reflectance(Ro). The resources of shale oil and gas are controlled by hydrocarbon-bearing porosity and effective shale thickness etc. The flow capacity of shale oil and gas is controlled by effective permeability, crude oil density, gas-oil ratio, condensate oil-gas ratio, formation pressure gradient, and Ro. The fracability of shale is directly controlled by brittleness index, and indirectly controlled by clay content in volume. EUR of shale oil and gas is controlled by six geological parameters: it is positively correlated with effective shale thickness, TOC and fracture porosity, negatively correlated with clay content in volume, and increases firstly and then decreases with the rise of Ro and formation pressure gradient. Under the present upper limit of horizontal well fracturing effective thickness of 65 m and the lower limit of EUR of 3×10^(4) m^(3), when TOC<2.3%, or Ro<0.85%, or clay content in volume larger than 25%, and fractures and micro-fractures aren’t developed, favorable areas of shale oil and gas hardly occur.