CN-122014218-A - Fracture surface density parameter prediction method and device based on array induction logging

CN122014218ACN 122014218 ACN122014218 ACN 122014218ACN-122014218-A

Abstract

The application discloses a fracture surface density parameter prediction method and device based on array induction logging, wherein the method comprises the steps of obtaining at least three initial resistivity curves of a target well by adopting an array induction mode; the method comprises the steps of preprocessing at least three initial resistivity curves to obtain at least three target resistivity curves, determining initial fracture surface characteristic vectors of a target well based on the at least three target resistivity curves, normalizing the initial fracture surface characteristic vectors to obtain target fracture surface characteristic vectors, inputting the target fracture surface characteristic vectors into a target prediction model for analysis to obtain fracture surface density parameters of the target well, and accordingly solving the technical problem that the fracture surface density parameters are difficult to stably predict under the condition of high-angle or vertical fracture.

Inventors

LUO YANG
LI JIE

Assignees

中国地质大学（武汉）

Dates

Publication Date: 20260512
Application Date: 20260318

Claims (10)

1. A fracture surface density parameter prediction method based on array induction logging, the method comprising: Acquiring at least three initial resistivity curves of a target well in an array induction mode, wherein the resistivity in the initial resistivity curves is used for representing the conductivity of fluid in a stratum, and the three initial resistivity curves are respectively used for representing the resistivity change conditions of different detection depths; Preprocessing at least three initial resistivity curves to obtain at least three target resistivity curves, wherein data noise in the target resistivity curves is smaller than data noise in the initial resistivity curves; Determining an initial fracture surface feature vector of the target well based on at least three target resistivity curves, and carrying out normalization processing on the initial fracture surface feature vector to obtain a target fracture surface feature vector; Inputting the target fracture surface feature vector into a target prediction model for analysis to obtain fracture surface density parameters of the target well, wherein the target prediction model is obtained by training an initial prediction model through historical data of the target fracture surface feature vector and historical data of the fracture surface density parameters, and the initial prediction model is obtained through neural network construction.
2. The method of claim 1, wherein preprocessing at least three of the initial resistivity curves to obtain at least three target resistivity curves comprises: Performing logarithmic transformation processing on at least three initial resistivity curves respectively to obtain at least three first resistivity curves; denoising at least three first resistivity curves to obtain at least three second resistivity curves; Performing background separation processing on at least three second resistivity curves to obtain at least three third resistivity curves; at least three of the target resistivity curves are determined based on at least three of the first resistivity curves and at least three of the second resistivity curves.
3. The method of claim 1, wherein the at least three target resistivity curves comprise a first target resistivity curve, a second target resistivity curve, and a third target resistivity curve, wherein the first target resistivity curve has a detection depth less than a detection depth of the second target resistivity curve, the second target resistivity curve has a detection depth less than a detection depth of the third target resistivity curve, and determining an initial fracture surface feature vector for the target well based on the at least three target resistivity curves comprises: Determining a radial differential parameter for the target well based on the third target resistivity curve and the first target resistivity curve; determining a radial curvature parameter of the target well based on the first, second, and third target resistivity curves; Determining a first average value corresponding to the first target resistivity curve, a second average value corresponding to the second target resistivity curve, and a third average value corresponding to the third target resistivity curve, respectively, based on the first target resistivity curve, the second target resistivity curve, and the third target resistivity curve; determining a radial uniformity parameter for the target well based on the first target resistivity curve, the second target resistivity curve, the third target resistivity curve, the first mean, the second mean, and the third mean; determining boundary enhancement parameters of the target well based on the third target resistivity curve, and performing scale smoothing on the third target resistivity curve to obtain scale band-pass parameters; And constructing the initial fracture surface feature vector based on the radial difference parameter, the radial curvature parameter, the radial consistency parameter, the boundary enhancement parameter and the scale band-pass parameter.
4. A method according to claim 3, wherein performing scale smoothing on the third target resistivity curve to obtain scale bandpass parameters comprises: Performing short-scale smoothing on the third target resistivity curve to obtain a first smooth curve, and performing long-scale smoothing on the third target resistivity curve to obtain a second smooth curve; the scale bandpass parameter is determined based on the first smoothing curve and the second smoothing curve.
5. The method of claim 3, wherein determining the radial differential parameter for the target well based on the third target resistivity curve and the first target resistivity curve by the formula comprises: DR L =R far_proc R near_proc Wherein DR L is used to represent the radial difference parameter, R far_proc is used to represent the third target resistivity curve, and R near_proc is used to represent the first target resistivity curve.
6. The method of claim 3, wherein determining the radial curvature parameter of the target well based on the first, second, and third target resistivity curves by the following formula comprises: C=R far_proc 2R mid_proc +R near_proc Wherein C is used to represent a radial curvature parameter and R mid_proc is used to represent the second target resistivity curve.
7. The method of claim 3, wherein determining the radial uniformity parameter for the target well based on the first target resistivity curve, the second target resistivity curve, the third target resistivity curve, the first average, the second average, and the third average by the formula comprises: Wherein RING is used to represent the radial compliance parameter, N is used to represent an integer, and N is greater than or equal to 3, For representing the ith target resistivity curve, For representing the mean of the ith target resistivity curve.
8. The method of claim 3, wherein determining the boundary enhancement parameters for the target well based on the third target resistivity curve by the following formula comprises: Wherein EDGE M is used to represent the boundary enhancement parameters, For representing the numerical derivative/first order difference.
9. A fracture surface density parameter prediction device based on array induction logging, the device comprising: the first acquisition unit is used for acquiring at least three initial resistivity curves of the target well in an array induction mode, wherein the resistivity in the initial resistivity curves is used for representing the conductivity of fluid in a stratum, and the three initial resistivity curves are respectively used for representing the resistivity change conditions of different detection depths; The second acquisition unit is used for preprocessing at least three initial resistivity curves to obtain at least three target resistivity curves, wherein data noise in the target resistivity curves is smaller than data noise in the initial resistivity curves; the determining unit is used for determining an initial fracture surface feature vector of the target well based on at least three target resistivity curves, and normalizing the initial fracture surface feature vector to obtain a target fracture surface feature vector; The third obtaining unit is used for inputting the target fracture surface feature vector into a target prediction model for analysis to obtain the fracture surface density parameter of the target well, wherein the target prediction model is obtained by training an initial prediction model through historical data of the target fracture surface feature vector and historical data of the fracture surface density parameter, and the initial prediction model is obtained through neural network construction.
10. A computer device comprising a storage medium, a processor and a computer program stored on the storage medium and executable on the processor, characterized in that the processor implements the method of any one of claims 1 to 8 when executing the computer program.

Description

Fracture surface density parameter prediction method and device based on array induction logging Technical Field The application relates to the technical field of oil and gas exploration and development, in particular to a fracture surface density parameter prediction method and device based on array induction logging. Background In the process of crack identification and quantitative evaluation of tight sandstone and mudstone reservoirs, the prior art has obvious limitations. The existing scheme mainly comprises three types, namely a direct evaluation method based on imaging logging and a rock core, which can intuitively reflect geometric characteristics of cracks, but has high cost and limited implementation conditions, and is difficult to comprehensively popularize, and an empirical or comprehensive index evaluation method based on conventional logging, which generally builds crack evaluation indexes based on abnormal response or combined characteristics of conventional logging curves such as resistivity, sound waves and the like, indirectly characterizes the development degree of the cracks, but has weak response under the conditions of compact sand shale and high-angle cracks, is easily influenced by factors such as lithology, invasion effect and the like, and has quantitative relation depending on experience calibration, and a data-driven crack prediction method, which directly utilizes the original logging curves or simple statistical characteristics to build a prediction model, but has insufficient stability and interpretability, and has higher requirements on the integrity of logging data. In summary, the prior art has the technical problem that the fracture surface density parameter is difficult to stably predict under the condition of high-angle or vertical fracture. Disclosure of Invention In view of the above, the embodiment of the application provides a method and a device for predicting the fracture surface density parameter based on array induction logging, so as to at least solve the technical problem that the fracture surface density parameter is difficult to stably predict under the condition of high angle or vertical fracture. According to one aspect of the application, a fracture surface density parameter prediction method based on array induction logging is provided, and the method comprises the steps of obtaining at least three initial resistivity curves of a target well in an array induction mode, wherein the resistivity in the initial resistivity curves is used for representing the conductivity of fluid in a stratum, the three initial resistivity curves are respectively used for representing the resistivity change conditions of different detection depths, preprocessing the at least three initial resistivity curves to obtain at least three target resistivity curves, wherein data noise in the target resistivity curves is smaller than data noise in the initial resistivity curves, determining initial fracture surface feature vectors of the target well based on the at least three target resistivity curves, normalizing the initial fracture surface feature vectors to obtain target fracture surface feature vectors, inputting the target fracture surface feature vectors into a target prediction model to analyze to obtain fracture surface density parameters of the target well, training the initial prediction model through historical data of the target fracture surface feature vectors and historical data of the fracture surface density parameters, and constructing the initial prediction model through a neural network. Optionally, preprocessing at least three initial resistivity curves to obtain at least three target resistivity curves, wherein the preprocessing comprises the steps of performing logarithmic transformation processing on the at least three initial resistivity curves to obtain at least three first resistivity curves, denoising the at least three first resistivity curves to obtain at least three second resistivity curves, performing background separation processing on the at least three second resistivity curves to obtain at least three third resistivity curves, and determining the at least three target resistivity curves based on the at least three first resistivity curves and the at least three second resistivity curves. Optionally, the at least three target resistivity curves comprise a first target resistivity curve, a second target resistivity curve and a third target resistivity curve, wherein the detection depth of the first target resistivity curve is smaller than the detection depth of the second target resistivity curve, the detection depth of the second target resistivity curve is smaller than the detection depth of the third target resistivity curve, the initial fracture surface feature vector of the target well is determined based on the at least three target resistivity curves, the method comprises the steps of determining a radial difference parameter of the target we