阅读说明：本技术 基于网格划分的储层参数测井评价方法 (Reservoir parameter logging evaluation method based on grid division ) 是由 张晋言 刘伟 刘兵开 吕增伟 于宁宁 张文姣 赵昕 于 2019-02-13 设计创作，主要内容包括：本发明公开了一种利用网格划分建模技术解决复杂油气藏储层参数建模的方法,将建模时输入的测井曲线等自变量信息,在其分布范围内按照一定的规则进行等分,在对自变量参数进行网格划分的基础上,统计岩心分析得到的参数值在每一个网格内的分布特征,之后通过统计方法确定最能代表某网格内储层参数分布特征的数值作为该网格内的参数值,最终通过建立网格划分模型。本发明通过对实际资料处理分析,与传统的多元统计回归建模方法相比,发现能够更好的解决大量数据条件下具有复杂的相关关系的数据建模问题,有效的降低了建立的参数模型与岩心分析数据之间的误差,提高了建模的精度,有较高推广价值和社会效益。(The invention discloses a method for solving parameter modeling of a complex oil and gas reservoir by utilizing a grid division modeling technology, which comprises the steps of equally dividing independent variable information such as a logging curve input during modeling according to a certain rule in a distribution range, counting the distribution characteristics of parameter values obtained by core analysis in each grid on the basis of grid division of the independent variable parameters, determining a numerical value which can represent the most distribution characteristics of reservoir parameters in a certain grid as the parameter values in the grid by using a counting method, and finally establishing a grid division model. Compared with the traditional multivariate statistical regression modeling method, the method disclosed by the invention has the advantages that through processing and analyzing the actual data, the data modeling problem with complex correlation under the condition of a large amount of data can be better solved, the error between the established parameter model and the core analysis data is effectively reduced, the modeling precision is improved, and the method has higher popularization value and social benefit.)

1. A reservoir parameter logging evaluation method based on grid division is characterized by comprising the following steps:

step (1), determining logging curve data serving as independent variable parameters, and establishing an independent variable grid;

step (2), projecting the core analysis data into grids, counting and calculating distribution characteristic values such as the average value, the maximum value, the minimum value and the peak value of the core analysis data in each grid;

step (3), determining a numerical value which can represent the most reservoir parameter distribution characteristics in a certain grid as a parameter value in the grid by using a statistical method;

step (4), calculating a grid characteristic value close to no data point by using an interpolation method, and establishing a grid model;

and (5) predicting reservoir parameters, extracting logging curve related information, projecting the logging curve related information into a grid, and predicting reservoir parameter values by adopting a triangle-based cubic equation interpolation method.

2. The method for logging and evaluating reservoir parameters based on meshing of claim 1, wherein: and (1) determining that the input logging curve data is in the effective numerical value distribution range, and dividing the curve according to a linear rule or a logarithmic rule according to the abundance and the modeling precision of the actual data so as to establish a one-dimensional or multi-dimensional independent variable grid.

The technical field is as follows:

the invention relates to the field of reservoir parameter evaluation methods in the field of well logging in petroleum exploration and development industries, and provides a reservoir parameter well logging evaluation method based on grid division.

Background art:

reservoir parameter calculation is an important component in reservoir comprehensive evaluation, a logging response equation is generally established on the basis of a logging volume model and an Archie formula for reservoir parameter calculation, but a complex reservoir is influenced by various factors such as lithology and heterogeneity, the model is too simple and cannot adapt to the situation of an actual stratum, interpretation parameters in the model are related to the experience of interpreters and sometimes cannot be selected, in recent years, a mathematical statistical method is adopted for establishing the relationship between logging data and core analysis data, and then the relationship is applied for quantitative interpretation, but when statistical regression modeling is applied, the problems in two aspects are mainly faced, on one hand, due to the characteristics of the regression method, sample points participating in regression are enabled to have the minimum integral error of samples used in a certain specific formula form, so that under certain conditions, reservoir parameters calculated by applying a statistical regression method to establish a model often have a certain variation trend on an intersection graph of a core analysis result, and are not distributed on a 45-degree line. On the other hand, for reservoir parameters with complex relationships, the rule that the reservoir parameters change along with independent variables cannot be accurately described by applying a single regression relationship. It is therefore almost impossible to accurately characterize reservoir parameters of complex reservoirs by applying a single interpretation equation.

The grid division modeling technology has the core idea that independent variable information such as well logging curves and the like is divided in the distribution range according to values determined by the abundance of actual data and modeling precision, on the basis of grid division of the independent variable parameters, the distribution characteristics of data obtained by core analysis in each grid are counted, then the values which can represent the data distribution characteristics in a certain grid most are determined by a counting method to serve as parameter values in the grid, and finally a grid division model is established. The method can well describe the change rule of the reservoir parameter data and solve the problem of data modeling with complex correlation.

The invention content is as follows:

the invention aims to solve the technical problem that the interpretation parameters of a complex oil and gas reservoir cannot be accurately described by applying a single interpretation equation, provides a method for establishing a reservoir parameter model by applying a grid division technology, improves the calculation precision of reservoir parameters, predicts and describes the change rule of reservoir data, and solves the problem of data modeling with complex correlation; the technical scheme is as follows:

in order to achieve the purpose, the invention provides a reservoir parameter logging evaluation method by using grid division, which comprises the following specific steps:

a reservoir parameter logging evaluation method based on grid division comprises the following steps:

step (1), determining logging curve data serving as independent variable parameters, and establishing an independent variable grid;

step (2), projecting the core analysis data into grids, counting and calculating distribution characteristic values such as the average value, the maximum value, the minimum value and the peak value of the core analysis data in each grid;

step (3), determining a numerical value which can represent the most reservoir parameter distribution characteristics in a certain grid as a parameter value in the grid by using a statistical method;

step (4), calculating a grid characteristic value close to no data point by using an interpolation method, and establishing a grid model;

and (5) predicting reservoir parameters, extracting logging curve related information, projecting the logging curve related information into a grid, and predicting reservoir parameter values by adopting a triangle-based cubic equation interpolation method.

Further, the step (1) is to determine that the input logging curve data is in the effective numerical value distribution range, and divide the curve according to a linear rule or a logarithmic rule according to the abundance and the modeling precision of the actual data, so as to establish a one-dimensional or multi-dimensional grid.

The invention has the beneficial effects that: the method is used for evaluating reservoir parameters of the complex reservoir, so that the calculation precision of the reservoir parameters is improved, the variation trend of data points is reflected, and multiple practices prove that the grid division modeling method can better solve the data modeling problem with complex correlation.

Drawings

FIG. 1 is a flow chart of an embodiment of the present invention.

Fig. 2 is a schematic diagram of a one-dimensional meshing method.

FIG. 3 is a schematic diagram of acoustic moveout and density model meshing.

FIG. 4 is a plot of the porosity distribution within the grid;

FIG. 5 is a representation of a two-dimensional porosity model created using sonic moveout and density logs.

Detailed Description

The invention will be described in detail below (taking porosity model as an example) with reference to the drawings and examples.

Referring to the attached figure 1, a reservoir parameter logging evaluation method based on grid division comprises the following steps:

determining logging curve data serving as independent variable parameters, determining values in an effective value distribution range according to the abundance of actual data and modeling precision, and dividing the curve according to a linear rule or a logarithmic rule.

FIG. 2 is a schematic diagram of a one-dimensional grid division method, wherein the value of the modeling input independent variable parameter density logging data determined according to the abundance of the actual data and the modeling precision is 1.7-2.8g/cm in the effective value range³The inner part is divided into N parts.

FIG. 3 is a schematic diagram of the meshing of the acoustic wave time difference and density model, according to the one-dimensional mesh dividing method, the acoustic wave time difference curve data is divided by M equally within the effective numerical range of 50-150 μ s/ft, and thus a two-dimensional mesh for calculating the porosity model is established.

And (2) projecting data points of the core analysis into grids on the basis of performing two-dimensional grid division on the independent variable parameter curve, respectively analyzing the distribution characteristics of the data points in each grid, and calculating the distribution characteristic value of the core analysis data in each grid.

Fig. 4 is a distribution characteristic diagram of the porosity in the grid, and the distribution characteristic of the porosity in each grid obtained by the core analysis is counted, and characteristic values such as an average value, a maximum value, a minimum value, a peak value and the like in the grid are calculated in a statistical manner.

And (3) determining that the average value of the porosity can best represent the distribution characteristics of the storage layer parameters in each grid by using a statistical method.

Establishing a grid division model by using the average value of the porosity;

fig. 5 is a diagram showing a two-dimensional porosity model established by using acoustic time difference and density logging data, and it can be seen from the diagram that a trend surface of a core analysis data point can be established after two-dimensional gridding modeling is applied, and the acoustic time difference and density logging data and the core analysis porosity are not simple linearly increased relations, so the trend surface is not a simple plane but has height fluctuation and curvature together, the high value part of the porosity is slightly convex, and the low value part is slightly concave, which is an "S" type feature.

And (5) respectively extracting sound wave and density curve data points to project into a grid, firstly finding out 3 points around an interpolation point according to a Delaunay method by adopting a triangle-based cubic equation interpolation method to form a triangle, wherein the interpolation point is in the triangle, ensuring each group of adjacent sample data in the fitting process, and fitting a curve between the sample data points by adopting a cubic polynomial. In order to ensure the uniqueness of the fitting result, the first-order derivative and the second-order derivative are used for constraint at the sampling position of the cubic polynomial, and the data between the interpolation sample points by applying the method are ensured to satisfy the continuity of the first-order derivative and the second-order derivative, so that the curved surface formed by the interpolation result is smoother.