Shale lithofacies identification method based on infrared spectrum

Abstract

The invention relates to the technical field of petroleum geological logging, and discloses a shale lithofacies identification method based on infrared spectroscopy, and the method comprises the following steps: pretreating a collected shale sample to make the collected shale sample become a standard sample meeting measurement requirements; performing infrared spectrum measurement on the standard sample to obtain infrared spectrum data of the standard sample; based on the infrared spectrum data of the standard sample, establishing a shale lithofacies identification model for identifying shale mineral components by utilizing a machine learning algorithm; and utilizing the shale lithofacies identification model to analyze mineral components of shale rock to be detected. The shale lithofacies identification method disclosed by the invention is simple to operate, low in cost, accurate in lithofacies identification and wide in application range.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of petroleum geological logging, and in particular to a shale lithofacies identification method based on infrared spectroscopy. Background Art

[0002] Shale lithofacies classification is one of the important contents of the evaluation of the "double sweet spots" of unconventional shale formations. At present, the means of identifying lithofacies mainly rely on laboratory core XRD testing, which has obvious limitations: (1) It is difficult to obtain cores, and cores are generally not taken during the development stage, especially in the horizontal section; (2) It has poor timeliness and high cost, and cannot realize lithofacies identification while drilling horizontal wells; (3) The commonly used XRD test is generally used for rock mineral classification and quantification, but it has problems such as complex operation, long time consumption, large differences in the results of measurements by different XRD instruments, and poor classification effect.

[0003] Therefore, there is an urgent need to develop more convenient, rapid, consistent and effective rock and mineral classification technologies to evaluate lithofacies types in order to meet exploration-related application needs such as on-site rapid lithofacies identification. Summary of the invention

[0004] The purpose of the present invention is to provide at least a shale lithofacies identification method based on infrared spectroscopy, which can at least solve the technical problems of poor timeliness, high cost, and inability to realize lithofacies identification while drilling horizontal wells in the prior art, and at least achieve a technical effect of convenient, fast and accurate identification.

[0005] To solve the above technical problems, at least one embodiment of the present invention provides a shale lithofacies identification method based on infrared spectroscopy, wherein the method comprises:

[0006] Pre-process the collected shale samples to make them standard samples that meet measurement requirements;

[0007] Performing infrared spectrum measurement on the standard sample to obtain infrared spectrum data of the standard sample;

[0008] Based on the infrared spectrum data of the standard sample, a shale lithofacies identification model for identifying shale mineral components is established using a machine learning algorithm;

[0009] The shale lithofacies identification model is used to analyze the mineral components of the shale rock to be tested.

[0010] In some optional embodiments, the analyzing the mineral components of the shale rock to be tested by using the shale lithofacies identification model includes:

[0011] Pre-process the shale rock to be tested so that it becomes a sample to be tested that meets the measurement requirements;.

[0012] Performing infrared spectrum measurement on the sample to be tested to obtain infrared spectrum data of the sample to be tested;

[0013] The infrared spectrum data of the sample to be tested is input into the shale lithofacies identification model to obtain the mineral components of the sample to be tested.

[0014] In some optional embodiments, the collected shale samples are preprocessed to make them standard samples that meet measurement requirements, including:

[0015] The shale sample is ground so that the particle size of the ground standard sample is greater than a preset threshold and the particle size is uniform.

[0016] In some optional embodiments, the step of performing infrared spectrum measurement on the standard sample to obtain infrared spectrum data of the standard sample includes:

[0017] Performing attenuated total emission infrared spectroscopy measurement and / or diffuse reflectance infrared spectroscopy measurement on the standard sample;

[0018] Different infrared spectrum databases are established for different types of infrared spectrum testing methods.

[0019] In some optional embodiments, the shale lithofacies identification model for identifying shale mineral components is established based on the infrared spectrum data of the standard sample using a machine learning algorithm, including:

[0020] Based on the infrared spectrum data of the standard sample, calculating the data of the second-order derivative spectrum of the infrared spectrum of the standard sample;

[0021] A shale lithofacies identification model for identifying shale mineral components is established using a machine learning algorithm, and the data of the second-order derivative spectrum is used to complete the training and verification of the shale lithofacies identification model.

[0022] In some optional embodiments, the calculating of the data of the second-order derivative spectrum of the infrared spectrum of the standard sample comprises:

[0023] Based on the Savitzky Golay algorithm, the data of the second-order derivative spectrum of the infrared spectrum of the standard sample is calculated.

[0024] In some optional embodiments, the use of the second-order derivative spectrum data to complete the training and verification of the shale lithofacies identification model includes:

[0025] Data within a specified band range is selected from the second-order derivative spectrum of the infrared spectrum of the standard sample to train and test the shale lithofacies identification model.

[0026] In some optional embodiments, the machine learning algorithm includes at least one supervised machine learning algorithm selected from the group consisting of partial least squares discriminant algorithm, principal component analysis, support vector machine, and convolutional neural network.

[0027] At least one embodiment of the present invention further provides an electronic device, including:

[0028] at least one processor; and,

[0029] a memory communicatively connected to the at least one processor; wherein,

[0030] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the shale lithofacies identification method based on infrared spectroscopy as described above.

[0031] At least one embodiment of the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the shale lithofacies identification method based on infrared spectroscopy as described above.

[0032] At least one embodiment of the present invention further provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the steps of the shale lithofacies identification method based on infrared spectroscopy as described above are implemented.

[0033] An embodiment of the present invention provides a shale lithofacies identification method based on infrared spectroscopy, the method comprising: pre-processing the collected shale samples to make them standard samples that meet the measurement requirements; performing infrared spectral measurement on the standard samples to obtain infrared spectral data of the standard samples; using a machine learning algorithm to establish a shale lithofacies identification model for identifying shale mineral components based on the infrared spectral data of the standard samples; and using the shale lithofacies identification model to analyze the mineral components of the shale rock to be measured. This method can overcome the current technical difficulties that shale lithofacies identification is heavily dependent on core and laboratory analysis, has poor timeliness, and is expensive. It innovatively proposes a rock mineral classification and identification method based on Fourier transform infrared spectroscopy, realizes the rapid distinction of shale formation lithofacies, and can especially meet the application needs related to exploration such as rapid on-site identification of lithofacies. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0035] Figure 1 is a flow chart of a shale lithofacies identification method provided by an embodiment of the present invention;

[0036] Figure 2 is a schematic diagram of a DRIFTS spectrum of carbonate shale according to Example 2 of the present invention;

[0037] Figure 3 is a schematic diagram of a DRIFTS spectrum of clay shale according to Example 2 of the present invention;

[0038] Figure 4 is a schematic diagram of a DRIFTS spectrum of felsic shale in Example 2 of the present invention;

[0039] Figure 5 It is a variation of the prediction accuracy of the prediction set of an embodiment of the present invention with the number of partial least squares latent variables. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0041] In order to solve the above technical problems, the present invention proposes a shale lithofacies identification method based on infrared spectroscopy.

[0042] like Figure 1 As shown, the main core idea of ​​this method is to perform infrared spectral measurements on the standard shale sample and the sample to be tested respectively, and then, based on the infrared spectral data of the standard sample, a shale lithofacies identification model is established (including training and verification) using a machine learning algorithm. Finally, the infrared spectral data of the sample to be tested is input into the trained shale lithofacies identification model to obtain the lithofacies identification result of the sample to be tested.

[0043] The implementation details of the above method are specifically described below through examples. The following content is only for the convenience of understanding the implementation details and is not necessary for implementing this solution.

[0044] Embodiment 1:

[0045] This embodiment provides a shale lithofacies identification method based on infrared spectroscopy, which mainly includes the following steps:

[0046] S10, pre-treating the collected shale samples to make them standard samples that meet the measurement standards;

[0047] S20, performing infrared spectrum measurement on the standard sample to obtain infrared spectrum data of the standard sample;

[0048] S30, based on the infrared spectrum data of the standard sample, a shale lithofacies identification model for identifying shale mineral components is established using a machine learning algorithm;

[0049] S40, using the shale lithofacies identification model to analyze the shale rock to be tested (i.e. Figure 1 The mineral composition of the unknown sample).

[0050] In step S40, the mineral components of the shale rock to be tested are analyzed using the shale lithofacies identification model, further comprising:

[0051] S40.1, pre-treat the shale rock to be tested so that it becomes a test sample that meets the measurement requirements;.

[0052] S40.2, performing infrared spectrum measurement on the sample to be tested to obtain infrared spectrum data of the sample to be tested;

[0053] S40.3, inputting the infrared spectrum data of the sample to be tested into a shale lithofacies identification model to obtain the mineral components of the sample to be tested.

[0054] Of course, in the specific implementation stage, steps S40.1 and S40.2 can also be performed simultaneously with steps S10 and S20 in the initial stage of implementation, and their execution order is not limited.

[0055] In addition, in the specific implementation stage, a suitable machine learning algorithm can be selected to establish a shale petrographic phase identification model for identifying shale mineral components according to the characteristics of the infrared spectrum data of the collected samples, such as the dimension of the data, the amount of data, and the accuracy of the data. In this embodiment, the machine learning algorithm is preferably a supervised machine learning algorithm, such as a partial least squares discriminant algorithm, or an algorithm combining principal component analysis and support vector machine, or a convolutional neural network, etc.

[0056] The above-mentioned shale lithofacies identification method based on infrared spectroscopy adopts a machine learning algorithm model based on multivariate statistics and uses infrared spectral data to achieve rapid and accurate distinction of shale lithofacies. It has a high prediction accuracy and meets the current application needs of shale formation evaluation while drilling.

[0057] Embodiment 2

[0058] The following is a specific example to further illustrate the technical solution of the present invention and its beneficial effects. This example uses diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) to test and analyze three types of shale facies, namely carbonate shale, clay shale and felsic shale. The shale facies identification method based on infrared spectroscopy provided in Example 1 is roughly the same, and the implementation steps of the whole process are as follows:

[0059] (1) Rock sample preparation

[0060] First, prepare a standard shale rock sample. In this step, the collected shale sample is pre-processed to make it a standard sample that meets the measurement standard, including: grinding the shale sample so that the particle size of the ground standard sample is greater than a preset threshold (e.g., 100 mesh) and the particle size is uniform. The ground standard sample can be directly used for measurement without other complicated pre-processing.

[0061] Then, mixed samples of three types of shale phases are prepared, among which: carbonate shale is composed of clay minerals, quartz, calcite and dolomite mixed in different proportions, wherein the sum of calcite and dolomite is not less than 50%; clay shale is composed of clay minerals, quartz, calcium feldspar, potassium feldspar and calcite mixed in different proportions, wherein the clay minerals account for not less than 50%; felsic shale is composed of clay minerals, quartz, potassium feldspar, plagioclase, calcite, dolomite and ankerite mixed in different proportions, wherein the sum of quartz and feldspar minerals account for not less than 50%.

[0062] (2) Infrared spectroscopy measurement and database establishment

[0063] The mixed sample was measured by ALPHA II Fourier transform infrared spectrometer, using attenuated total emission infrared spectroscopy or diffuse reflectance infrared spectroscopy. To improve data quality, multiple scans can be averaged, the sampling number can be 16 to 64 times, and the resolution can be selected to be 4 cm -1 , the scanning time can be selected as 3min, the background scanning time can be selected as 3min, and the saved band range is 4000cm -1 ~400cm -1 The data was obtained, and the resulting spectra were presented in the form of Kubelka Munk (KM). Before measuring each group of samples, the background needs to be scanned to ensure the accuracy of the sample measurement. The final database contains about 20 samples in each category, and each sample is measured 2 to 3 times.

[0064] Collect infrared spectrum data of standard shale samples and establish infrared spectrum database. Different infrared spectrum databases should be established for infrared spectrum data of different test types (attenuated total emission or diffuse reflection, etc.).

[0065] In the specific implementation stage, the second-order derivative spectra of different infrared spectra can be calculated based on the Savitzky Golay algorithm, and the data of the second-order derivative spectra can be used to train and test the model, wherein the data in the band range of 4000cm-1 to 900cm-1 is preferred. Figure 2 to Figure 4 The infrared spectra of different types of shales in this embodiment are shown in FIG. 1 , wherein the left side of each figure is the original spectrum and the right side is the corresponding second-order derivative spectrum; spectra of different colors represent samples with different proportions, and spectra of the same color represent repeated measurements of the same sample.

[0066] (3) Establishment of algorithm model

[0067] In this embodiment, the algorithm model is trained based on the partial least squares discriminant algorithm, 4 / 5 of the spectral data in the database are randomly extracted to construct a training set for training the algorithm model, and the remaining 1 / 5 of the spectral data in the database are extracted to construct a prediction set for testing the algorithm model. Finally, the prediction accuracy of the algorithm model is analyzed. Figure 5 The figure shows that the average value and error of the prediction accuracy were calculated after ten repetitions.

[0068] Of course, in the specific implementation stage, the available machine learning models include but are not limited to partial least squares discriminant (PLS-DA), and can also be data dimension reduction + classification models, such as principal component analysis + support vector machine, or convolutional neural network, etc. Afterwards, different models can be cross-validated or compared to determine the most suitable model and appropriate model parameters.

[0069] (4) Lithofacies prediction of unknown samples

[0070] For the shale rock to be tested (i.e. Figure 1 For unknown samples), the infrared spectrum data of the unknown samples are collected using methods (1) and (2), and the lithology of the unknown samples is predicted based on the algorithm model (3).

[0071] Figure 5 The figure shows how the prediction accuracy of the algorithm model of this embodiment changes with the number of latent variables of the partial least squares method. As can be seen from the figure, the prediction effect of the algorithm model that extracts more than or equal to 2 latent variables is relatively close. Taking the extraction of 5 latent variables as an example, the prediction accuracy of unknown samples is 88±4%, and the highest can reach 92%.

[0072] The above-mentioned shale rock mineral classification and identification method based on Fourier transform infrared spectroscopy, combined with standardized sample preparation and data processing methods based on machine learning algorithms, can achieve rapid differentiation of common shale lithofacies. The shale lithofacies prediction model established by this method has a high prediction accuracy rate and can accurately identify shale lithofacies, overcoming the technical difficulties that the current shale lithofacies identification mainly relies on core laboratory analysis. The present invention adopts low-cost and widely used Fourier transform infrared spectroscopy technology, which is simple to operate, low-cost, accurate in lithofacies identification, and has a wide range of applications.

[0073] Embodiment three:

[0074] Another embodiment of the present invention relates to an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the shale lithofacies identification method in the above-mentioned embodiments.

[0075] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0076] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0077] Embodiment 4:

[0078] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program, which implements the shale lithofacies identification method of the above embodiment when executed by a processor.

[0079] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0080] Embodiment 5

[0081] Another embodiment of the present invention relates to a computer program product, including a computer program, which implements the steps of the shale lithofacies identification method of the above embodiment when executed by a processor.

[0082] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present invention, and that in actual applications, various changes may be made in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A shale lithofacies identification method based on infrared spectroscopy, characterized in that: include: Pre-process the collected shale samples to make them standard samples that meet measurement requirements; Performing infrared spectrum measurement on the standard sample to obtain infrared spectrum data of the standard sample; Based on the infrared spectrum data of the standard sample, a shale lithofacies identification model for identifying shale mineral components is established using a machine learning algorithm; The shale lithofacies identification model is used to analyze the mineral components of the shale rock to be tested.

2. The shale lithofacies identification method based on infrared spectroscopy according to claim 1 is characterized in that: The method of analyzing the mineral components of the shale rock to be tested by using the shale lithofacies identification model includes: Pre-process the shale rock to be tested so that it becomes a sample to be tested that meets the measurement requirements;. Performing infrared spectrum measurement on the sample to be tested to obtain infrared spectrum data of the sample to be tested; The infrared spectrum data of the sample to be tested is input into the shale lithofacies identification model to obtain the mineral components of the sample to be tested.

3. The shale lithofacies identification method based on infrared spectroscopy according to claim 1, characterized in that: The step of measuring the infrared spectrum of the standard sample to obtain infrared spectrum data of the standard sample includes: Performing attenuated total emission infrared spectroscopy measurement and / or diffuse reflectance infrared spectroscopy measurement on the standard sample; Different infrared spectrum databases are established for different types of infrared spectrum measurement methods.

4. The shale lithofacies identification method based on infrared spectroscopy according to claim 1, characterized in that: The method of establishing a shale lithofacies identification model for identifying shale mineral components based on the infrared spectrum data of the standard sample using a machine learning algorithm includes: Based on the infrared spectrum data of the standard sample, calculating the data of the second-order derivative spectrum of the infrared spectrum of the standard sample; A shale lithofacies identification model for identifying shale mineral components is established using a machine learning algorithm, and the data of the second-order derivative spectrum is used to complete the training and verification of the shale lithofacies identification model.

5. The shale lithofacies identification method based on infrared spectroscopy according to claim 4 is characterized in that: The data for calculating the second-order derivative spectrum of the infrared spectrum of the standard sample includes: Based on the Savitzky Golay algorithm, the data of the second-order derivative spectrum of the infrared spectrum of the standard sample is calculated.

6. The shale lithofacies identification method based on infrared spectroscopy according to claim 4 is characterized in that: The method of using the data of the second-order derivative spectrum to complete the training and testing of the shale lithofacies identification model includes: Data within a specified band range is selected from the second-order derivative spectrum of the infrared spectrum of the standard sample to train and test the shale lithofacies identification model.

7. The shale lithofacies identification method based on infrared spectroscopy according to any one of claims 1 to 6, characterized in that: The machine learning algorithm includes at least one supervised machine learning algorithm among partial least squares discriminant algorithm, principal component analysis, support vector machine and convolutional neural network.

8. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the shale lithofacies identification method based on infrared spectroscopy as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the infrared spectroscopy-based shale lithofacies identification method according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the shale lithofacies identification method based on infrared spectroscopy described in any one of claims 1 to 7 are implemented.

Images