Method for improving imaging resolution of borehole azimuthal gamma ray logging based on regularization processing

Abstract

The application discloses a method for improving the imaging resolution of a while-drilling azimuthal gamma logging device based on regularization processing, and relates to the technical field of oilfield logging, and comprises the following steps: acquiring azimuthal response sensitivity data of the while-drilling azimuthal gamma logging device; collecting multi-sector gamma imaging data of the azimuthal gamma logging at each depth according to the azimuthal response sensitivity data; and performing regularization processing on the multi-sector gamma imaging data to obtain new gamma data close to the real radioactivity intensity distribution of a formation, so that the imaging resolution of the while-drilling azimuthal gamma logging is improved. The application solves the problems of low imaging resolution and large measurement error of the existing logging.

Description

Technical Field

[0001] This invention relates to the field of oilfield logging technology, and in particular to a method for improving the resolution of azimuth gamma logging imaging based on regularization processing. Background Technology

[0002] Azimuth-based logging while drilling (AWDB) can not only determine clay content and assess formation permeability, but also provide crucial data for geosteering and accurate formation evaluation due to its 360° rotation and ability to measure azimuthally around the well. However, the imaging quality of AWDB is affected by various factors such as instrument focusing performance, formation radioactivity intensity, wellbore environment, and statistical errors. This results in low imaging resolution and an inability to accurately reflect the true distribution of radioactivity intensity, impacting the reliability of logging interpretation and evaluation. Therefore, processing the logging data is necessary to make it more closely resemble the actual formation model. Thus, accurate processing of azimuth-based logging data is key to improving data accuracy, i.e., imaging resolution.

[0003] Currently, due to limitations in downhole measurement signal transmission rates during drilling, azimuth gamma logging typically only uploads azimuth gamma data from 2 or 4 sectors in real time, while recording azimuth gamma data from 8 or 16 sectors in storage mode. Although the instrument incorporates a windowing design for the detector, to ensure a sufficiently high count rate, the measurement response does not reflect only a single azimuth but rather the combined contribution of all formations around the well. This significantly reduces the resolution of azimuth gamma imaging. Furthermore, the natural gamma intensity emitted by rocks in the formation is low, resulting in fewer gamma rays received by the detector. The more sectors involved, the lower the count, leading to a larger statistical error in azimuth gamma logging results. Summary of the Invention

[0004] To address the aforementioned shortcomings in existing technologies, the present invention provides a method for improving the resolution of azimuth gamma logging imaging based on regularization processing, which solves the problems of low imaging resolution and large measurement errors in existing logging methods.

[0005] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: a method for improving the resolution of azimuth gamma-ray logging imaging based on regularization processing, comprising the following steps:

[0006] S1: Acquire azimuth response sensitivity data of the drilling azimuth gamma logging instrument;

[0007] S2: Based on the azimuth response sensitivity data, collect multi-sector gamma imaging data of azimuth gamma logging at each depth;

[0008] S3: Regularize the multi-sector gamma imaging data to obtain new gamma data that closely approximates the true distribution of radioactivity intensity in the formation, thereby improving the resolution of azimuth gamma logging imaging while drilling.

[0009] The beneficial effects of the above solution are as follows: This invention provides a data processing method for improving the imaging resolution of gamma-ray logging while drilling based on regularization. It obtains the azimuth response sensitivity data of the logging instrument, and then processes the original gamma data through regularization. It continuously adjusts the relevant parameters to make the gamma data closer to the true value, thereby improving the imaging resolution and solving the problems of low imaging resolution and large measurement error in existing logging.

[0010] Furthermore, S1 includes the following sub-steps:

[0011] S1-1: Based on the structure of the azimuth logging instrument, construct the corresponding three-dimensional model of the instrument and the geological environment;

[0012] S1-2: Based on the instrument and the three-dimensional geological model, the Monte Carlo simulation method is used to perform numerical simulation of the azimuth gamma logging while drilling to obtain the azimuth response data under the current geological conditions;

[0013] S1-3: The radioactivity intensity and density of the strata in each of the 1st to 16th azimuth sectors were changed sequentially, and multiple numerical simulations were performed to obtain azimuth response data under different geological conditions;

[0014] S1-4: Process the azimuth response data under different geological conditions to obtain the response sensitivity data of the 1st to 16th azimuths suitable for the selected instruments and geological conditions.

[0015] The beneficial effects of the above-mentioned further scheme are as follows: By using the above technical scheme, the drilling process of azimuth gamma logging is simulated using the Monte Carlo simulation method. The Monte Carlo simulation method can simulate the individual response of each azimuth formation and has high calculation accuracy, which is conducive to accurately obtaining azimuth response data under the current geological conditions.

[0016] Furthermore, S2 includes the following sub-steps:

[0017] S2-1: Based on the detector counting principle, obtain the gamma logging data N0 for a single sector during drilling, using the following formula:

[0018]

[0019] Where n is the number of sectors within the detector's detection range, f i For the i-th sector, F r Here are the azimuth response sensitivity data, where d is the integral and N is the number of azimuth responses. i Count the gamma detectors in the i-th sector;

[0020] S2-2: Based on gamma logging data from a single sector, the detector is used to move in a circular motion around the borehole center to obtain gamma data from multiple sectors;

[0021] S2-3: Based on gamma data from multiple sectors, the detector is moved downwards to obtain multi-sector gamma imaging data of azimuth gamma logging at each depth.

[0022] The beneficial effect of the above-mentioned further solution is that, through the above technical solution, the detector can perform circular and downward motion to obtain gamma data from multiple sectors at different depths, which helps to improve the accuracy of the data.

[0023] Furthermore, S3 includes the following sub-steps:

[0024] S3-1: Construct a regularization processing model, and input the azimuth response sensitivity data and the azimuth gamma logging multi-sector gamma imaging data at each depth into the regularization processing model;

[0025] S3-2: The azimuth response of multi-sector gamma-ray imaging data at each depth in azimuth logging is processed using a regularization model. The calculation formula is constructed using the L2 regularization method:

[0026]

[0027] Where L is the loss function, m is the size of the training set, l is the number of selected training set samples, and y l This represents the actual value of the gamma data. Here, w represents the predicted value of the gamma data, λ is the regularization parameter, and w is the predicted value of the gamma data. l These are weight parameters;

[0028] S3-3: The formula for calculating weight updates is as follows:

[0029]

[0030] Where w(t+1) is the weight at time t+1, η is the learning rate, w(t) is the weight at time t, L0 is the original loss function, and w is the weight. For partial derivatives;

[0031] S3-4: Continuously adjust the regularization parameter λ and weight w, compare the different new gamma data obtained with the data obtained by fast forward modeling, make the error less than 5%, and obtain the true value of the processed gamma data;

[0032] S3-5: Use the actual values ​​of the processed gamma data to create a graph, show the imaging effect, and improve the resolution of azimuth logging imaging while drilling.

[0033] The beneficial effects of the above-mentioned further solutions are: by performing regularization processing on the original gamma data and continuously adjusting the regularization parameters and weights, gamma data close to the true value can be obtained, thereby improving the imaging resolution. Attached Figure Description

[0034] Figure 1 The flowchart shows a method for improving the resolution of azimuth gamma logging imaging while drilling based on regularization.

[0035] Figure 2 This is a schematic diagram of a stratigraphic sector.

[0036] Figure 3 This is a comparison chart of the original data and the azimuth gamma logging curves and imaging after regularization. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0038] like Figure 1 As shown, a method for improving the resolution of azimuth logging imaging based on regularization processing includes the following steps:

[0039] S1: Acquire azimuth response sensitivity data of the drilling azimuth gamma logging instrument;

[0040] S2: Based on the azimuth response sensitivity data, collect multi-sector gamma imaging data of azimuth gamma logging at each depth;

[0041] S3: Regularize the multi-sector gamma imaging data to obtain new gamma data that closely approximates the true distribution of radioactivity intensity in the formation, thereby improving the resolution of azimuth gamma logging imaging while drilling.

[0042] S1 includes the following steps:

[0043] S1-1: Based on the structure of the azimuth logging instrument, construct the corresponding instrument and geological 3D model, including setting various instrument and geological parameters such as drill collar, detector, wellbore size and formation size, and radioactivity intensity;

[0044] S1-2: Based on the instrument and the three-dimensional geological model, the Monte Carlo simulation method is used to perform numerical simulation of the azimuth gamma logging while drilling to obtain the azimuth response data under the current geological conditions;

[0045] S1-3: The radioactivity intensity and density of the strata in each of the 1st to 16th azimuth sectors were changed sequentially, and multiple numerical simulations were performed to obtain azimuth response data under different geological conditions;

[0046] S1-4: Process the azimuth response data under different geological conditions to obtain the response sensitivity data of the 1st to 16th azimuths suitable for the selected instruments and geological conditions.

[0047] S2 includes the following steps:

[0048] S2-1: Based on the detector counting principle, obtain the gamma logging data N0 for a single sector during drilling, using the following formula:

[0049]

[0050] Where n is the number of sectors within the detector's detection range, f i For the i-th sector, F r Here are the azimuth response sensitivity data, where d is the integral and N is the number of azimuth responses. i Count the gamma detectors in the i-th sector;

[0051] S2-2: Based on gamma logging data from a single sector, the detector is used to move in a circular motion around the borehole center to obtain gamma data from multiple sectors;

[0052] S2-3: Based on gamma data from multiple sectors, the detector is moved downwards to obtain multi-sector gamma imaging data of azimuth gamma logging at each depth.

[0053] like Figure 2 As shown, the detector's response in each orientation is affected by different orientation sectors. The figure includes 16 formation sectors. In this embodiment, data processing can be performed on sector data of different numbers such as 2, 4, and 6.

[0054] S3 includes the following steps:

[0055] S3-1: Construct a regularization processing model, and input the azimuth response sensitivity data and the azimuth gamma logging multi-sector gamma imaging data at each depth into the regularization processing model;

[0056] S3-2: The azimuth response of multi-sector gamma-ray imaging data at each depth in azimuth logging is processed using a regularization model. The calculation formula is constructed using the L2 regularization method:

[0057]

[0058] Where L is the loss function, m is the size of the training set, l is the number of selected training set samples, and y l This represents the actual value of the gamma data. Here, w represents the predicted value of the gamma data, λ is the regularization parameter, and w is the predicted value of the gamma data. l These are weight parameters;

[0059] S3-3: The formula for calculating weight updates is as follows:

[0060]

[0061] Where w(t+1) is the weight at time t+1, η is the learning rate, w(t) is the weight at time t, L0 is the original loss function, and w is the weight. For partial derivatives;

[0062] S3-4: Continuously adjust the regularization parameter λ and weight w, compare the different new gamma data obtained with the data obtained by fast forward modeling, make the error less than 5%, and obtain the true value of the processed gamma data;

[0063] S3-5: Use the actual values ​​of the processed gamma data to create a graph, show the imaging effect, and improve the resolution of azimuth logging imaging while drilling.

[0064] In one embodiment of the present invention, such as Figure 3 The figure shows a comparison of the original data and the azimuth gamma logging curves and imaging after regularization. It can be seen from the figure that when the detector enters another radioactive formation, the contribution of different azimuth sectors to the detector count varies, and the gamma-ray flux changes with the depth of detection. Furthermore, after regularization, the separation of the upper and lower gamma curves is more obvious, the gamma imaging resolution is improved, and the accurate formation information can be reflected more clearly, significantly improving the resolution of thin layers. Compared with the original data, the gamma logging imaging after regularization is accurate and distortion-free, with reduced blurring. Therefore, the method of this invention can effectively improve imaging resolution and meet engineering requirements.

[0065] This invention analyzes the generation principle of the sensitivity factor in azimuth logging while drilling (AWDB). Monte Carlo numerical simulation is used to establish instrument and formation models to obtain azimuth response sensitivity data for AWDB azimuth logging instruments. After analyzing the response patterns, the data generated by rapid forward modeling is stored and imaged. A regularization formula is then established to process the original gamma data, adjusting the calculation parameters to make the gamma data closer to the true values. Finally, the original and processed data are compared to examine and analyze the practical application effects. This invention can effectively improve the data accuracy and imaging resolution of azimuth logging while drilling, providing support for conventional logging applications such as geological steering and thin-layer identification.

[0066] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the invention.

Claims

1. A method for improving the resolution of azimuth logging imaging based on regularization processing, characterized in that, Includes the following steps: S1: Acquire azimuth response sensitivity data of the drilling azimuth gamma logging instrument; S2: Based on the azimuth response sensitivity data, collect multi-sector gamma imaging data of azimuth gamma logging at each depth; S2 includes the following sub-steps: S2-1: Based on the detector counting principle, obtain gamma logging data for a single sector during drilling. The formula is: in, The number of sectors within the detector's detection range. For the first Each sector, This is the azimuth response sensitivity data. For integration, For the first Count of gamma detectors in each sector; S2-2: Based on gamma logging data from a single sector, the detector is used to move in a circular motion around the borehole center to obtain gamma data from multiple sectors; S2-3: Based on gamma data from multiple sectors, the detector is moved downwards to obtain multi-sector gamma imaging data of azimuth gamma logging at each depth; S3: Regularize the multi-sector gamma imaging data to obtain new gamma data that closely approximates the true distribution of radioactivity intensity in the formation, thereby improving the resolution of azimuth gamma logging imaging while drilling. S3 includes the following sub-steps: S3-1: Construct a regularization processing model, and input the azimuth response sensitivity data and the azimuth gamma logging multi-sector gamma imaging data at each depth into the regularization processing model; S3-2: The azimuth response of multi-sector gamma-ray imaging data at each depth in azimuth logging is processed using a regularization model. The calculation formula is constructed using the L2 regularization method: in, For loss function, The size of the training set samples. These are the selected training set samples. This represents the actual value of the gamma data. For gamma data predictions, For regularization parameters, These are weight parameters; S3-3: The formula for calculating weight updates is as follows: in, for Weight of time, For learning rate, for Weight of time, The original loss function, As weight, For partial derivatives; S3-4: Continuously adjust the regularization parameters and weight The different new gamma data obtained are compared with the data obtained by fast forward modeling, and the error is made less than 5% to obtain the true value of the processed gamma data. S3-5: Use the actual values ​​of the processed gamma data to create a graph, show the imaging effect, and improve the resolution of azimuth logging imaging while drilling.

2. The method for improving the resolution of azimuth logging imaging based on regularization processing according to claim 1, characterized in that, S1 includes the following sub-steps: S1-1: Based on the structure of the azimuth logging instrument, construct the corresponding three-dimensional model of the instrument and the geological environment; S1-2: Based on the instrument and the three-dimensional geological model, the Monte Carlo simulation method is used to perform numerical simulation of the azimuth gamma logging while drilling to obtain the azimuth response data under the current geological conditions; S1-3: The radioactivity intensity and density of the strata in each of the 1st to 16th azimuth sectors were changed sequentially, and multiple numerical simulations were performed to obtain azimuth response data under different geological conditions; S1-4: Process the azimuth response data under different geological conditions to obtain the response sensitivity data of the 1st to 16th azimuths suitable for the selected instruments and geological conditions.

Images