A method and system for stratigraphic division based on multiple well logging curves

Abstract

The present invention relates to the technical field of well logging curve drawing, and specifically to a method and system for stratigraphic division based on multiple well logging curves; the method comprises: drawing well logging curves according to different well logging phases; obtaining the inflection point on each well logging curve, clustering the depth values ​​of the inflection points on all well logging curves to obtain cluster clusters; in response to the difference between the maximum value and the minimum value in the cluster cluster being greater than a preset threshold, determining the optimal stratification point according to the depth value in the cluster cluster. That is, the scheme of the present invention further obtains the optimal stratification point by drawing multiple well logging curves, referring to multiple well logging phases, and clustering the stratification points of the stratigraphic formation, thereby reducing the influence of a single factor on the division result and improving the accuracy of stratigraphic division.

Description

Technical Field

[0001] The present invention relates to the technical field of well logging curve drawing, and more specifically, to a method and system for dividing strata based on multiple well logging curves. Background Art

[0002] The well logging curve is drawn based on the data of different parameters measured during the drilling process that change with depth, such as resistivity, natural gamma rays, acoustic wave velocity, formation density, etc., where the depth is the vertical distance from the surface to the measuring point. The well logging curve reflects the lithology, physical properties and fluid properties of the reservoir, and is usually used to analyze the geological structure and rock properties passed by the wellbore.

[0003] Stratigraphic division refers to the division of rock layers in a stratigraphic section of a region and the establishment of stratigraphic sequence, which is a very important basic work in geological research. Generally, for a stratigraphic section of a region, the rock stratigraphy is first divided according to the lithology and lithofacies characteristics, and then the biostratigraphy is divided according to the fossils collected systematically, and then the chronostratigraphic sequence is established.

[0004] The Chinese patent application document with publication number CN112211628A discloses a method for stratigraphic division based on multiple logging curves, which uses the difference between neutron porosity and density porosity to characterize the shale content, and builds a correlation with the natural gamma logging data that characterizes the radioactivity intensity. The stratigraphic division can be performed based on the difference in correlation. A method for dividing stratigraphic division using the difference in correlation between radioactivity intensity and shale content between different stratigraphic formations is proposed.

[0005] The above scheme divides the strata by mapping the mud content with gamma rays. The division elements are relatively simple and the division results are greatly affected by the single element data, which makes the final results have large errors and poor accuracy. Summary of the invention

[0006] The present invention proposes a stratum division method and system based on multiple well logging curves, which is used to solve the problem of large errors when using a single element to divide strata in the prior art. To this end, the present invention provides solutions in the following two aspects.

[0007] In a first aspect, the present invention provides a method for stratigraphic division based on multiple well logging curves, comprising the following steps:

[0008] Draw logging curves according to different logging phases, including resistivity, acoustic wave velocity, formation density, and gamma ray;

[0009] Obtain the inflection point on each logging curve, and cluster the depth values ​​of the inflection points on all logging curves to obtain multiple clusters;

[0010] In response to the difference between the maximum value and the minimum value in the cluster being greater than a preset threshold, determining an optimal layering point according to the depth value in the cluster;

[0011] Among them, the method for determining the optimal stratification point is:

[0012] Set the initial parameters, use the absolute value of the difference between the initial parameters and each depth value in the cluster to construct a difference sequence, and use the response intensity of the inflection point corresponding to the depth value in the cluster to construct a standard response intensity sequence;

[0013] The Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is calculated, and the initial parameters are updated until the Pearson correlation coefficient is less than the threshold value to obtain the optimal parameters, which are used as the best stratification points of the logging curve.

[0014] By drawing multiple logging curves, referring to multiple logging phases, and clustering the stratification points of the formation, the optimal stratification points can be further obtained, reducing the influence of a single factor on the division results and improving the accuracy of stratigraphic division.

[0015] Preferably, the method for constructing a standard response intensity sequence is:

[0016] The mean of the response intensity on the logging curve is calculated, and the ratio of the response intensity of the inflection point to the mean of the corresponding logging curve is taken as the standard response intensity of the inflection point. The standard response intensities of all inflection points in the cluster form a standard response intensity sequence.

[0017] By standardizing the response intensity of the inflection point to obtain the standard response intensity, the situation where the response intensity values ​​have large differences is reduced, which facilitates the subsequent calculation of the standard response intensity and improves the accuracy of the calculation results.

[0018] Preferably, the method for obtaining the inflection point on each logging curve is:

[0019] The moving average algorithm is used to smooth the logging curve, the second-order derivative of the smoothed curve is calculated, and the point where the second-order derivative is zero is taken as the inflection point.

[0020] By smoothing the logging curve, it is easy to calculate the inflection point of the smooth curve, which provides great convenience for dividing the formation.

[0021] Preferably, in response to the difference between the maximum value and the minimum value in the cluster being less than or equal to a preset threshold, the median of the depth values ​​in the cluster or the mean of the depth values ​​is used as the stratification point on the corresponding logging curve.

[0022] The mean or median of the depth values ​​is used as the stratification point, which reduces the error in determining the position of the stratification point and improves the position accuracy of the stratification point.

[0023] Preferably, data of each logging phase is obtained, mean shift clustering is performed on the values ​​of the same logging phase data at different depths, data points in the clustering results that do not belong to any cluster are taken as outliers, and then the outliers are deleted.

[0024] Preferably, the depth values ​​of the inflection points are clustered using the DBSCAN algorithm.

[0025] Preferably, the initial parameter is between the maximum and minimum depth values ​​in the cluster.

[0026] By determining the numerical range of the initial parameters, it is convenient to update the initial parameters to obtain the optimal stratification point.

[0027] In a second aspect, the present invention further provides a well logging curve automatic drawing system, comprising:

[0028] processor;

[0029] A memory stores computer instructions for safety early warning of photovoltaic power stations based on artificial intelligence. When the computer instructions are executed by the processor, the system executes the stratigraphic division method based on multiple well logging curves.

[0030] The beneficial effects of the present invention are:

[0031] The solution of the present invention can further obtain the best stratification point by drawing multiple logging curves, referring to multiple logging phases, and clustering the stratification points of the formation, thereby reducing the influence of a single factor on the division result and improving the accuracy of stratum division. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0033] Figure 1 A flowchart schematically shows a method for dividing a formation based on multiple well logging curves in this embodiment;

[0034] Figure 2 The structural block diagram of a stratigraphic division system based on multiple well logging curves in this embodiment is schematically shown. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention. The specific implementation methods of the present invention are described in detail in combination with the drawings.

[0036] Figure 1 A flowchart of the steps of a method for stratigraphic division based on multiple well logging curves in this embodiment is schematically shown.

[0037] Specifically, Figure 1 As shown, a stratigraphic division method based on multiple well logging curves in this embodiment includes the following steps:

[0038] S1. Draw logging curves according to different logging phases.

[0039] The logging phase includes resistivity, acoustic wave velocity, formation density, and gamma ray. The data of each logging phase is obtained, and the values ​​of the same logging phase data at different depths are clustered by mean shift. The data points that do not belong to any cluster in the clustering result are regarded as outliers, and then the outliers are deleted. For the data points with deleted outliers, missing value processing is performed. Specifically, the data fitting model is used by the random forest algorithm to predict the values ​​of the data points, and the predicted values ​​are used to replace the outliers, thereby improving the accuracy of the acquired logging phase data. Data fitting is a way of representing the existing data by mathematical methods into a formula. It is an existing technology, and the specific principle is not repeated here. Select a suitable model for the preprocessed logging data to draw different logging curves. Eliminate the abnormal data in the logging phase data, and draw logging curves for resistivity, acoustic wave velocity, formation density, and gamma ray. The ordinate of the logging curve is the depth value, and the abscissa is the corresponding response intensity of the logging phase. The logging curve reflects the change of the response intensity with the depth value. For example, in a resistivity logging curve, the resistivity value of a data point is the response intensity value, and in a sonic velocity logging curve, the sonic velocity value of a data point is the response intensity value.

[0040] S2. Obtain the inflection point on each logging curve, and cluster the depth values ​​of the inflection points on all logging curves to obtain clusters.

[0041] In the drawn logging curve, the logging curve is not smooth, and direct inflection point detection will produce large errors. Therefore, the moving average algorithm is used to smooth the logging curve. The moving average algorithm has a smoothing or smoothing effect on the data sequence, weakening the ups and downs of the data sequence. The moving average algorithm is an existing technology, and the specific process is not repeated here. Calculate the second-order derivative of the smoothed curve, and take the point where the second-order derivative is zero as the inflection point. Each inflection point corresponds to a depth value and a response intensity value. The depth values ​​of all inflection points on all logging curves are clustered using DBSCAN (density-based clustering algorithm) to obtain multiple clusters. The second-order derivative represents the rate of change of the slope of the smooth curve. The inflection point of the logging curve means that the rate of change of the response intensity of the formation at the corresponding depth reaches an extreme value, the concavity of the logging curve changes, and the formation is likely to be stratified. By obtaining the inflection point, it is easy to determine the depth value of the stratification point.

[0042] S3. In response to the difference between the maximum value and the minimum value in the cluster being greater than a preset threshold, determining an optimal layering point according to the depth value in the cluster.

[0043] The standard response intensity sequence is constructed using the response intensity of the inflection point corresponding to the depth value in the cluster: the mean of the response intensity on the logging curve is calculated respectively, and the ratio of the response intensity of the inflection point to the mean of the corresponding logging curve is taken as the standard response intensity of the inflection point; the standard response intensity of all inflection points constitutes a standard response intensity sequence.

[0044] For example, for the resistivity logging curve, first obtain the response intensity of the inflection point depth value on the resistivity logging curve , the response intensity is standardized to obtain the standard response intensity, and the expression of the standard response intensity is:

[0045]

[0046] In the formula, is the standard response intensity of the inflection point, is the average value of the response intensity on the logging curve, is the response intensity of the inflection point. Standardizing the logging curve can make the logging curve smoother, making it easier to find the inflection point. It should be noted here that the response value of the same logging curve in different formations will vary greatly, and even the response value will suddenly change. After smoothing, the place where the response value suddenly changes is where the inflection point is.

[0047] The method for determining the optimal stratification point is as follows: setting initial parameters, the values ​​of which are between the maximum and minimum values ​​in the cluster; constructing a difference sequence using the absolute value of the difference between the initial parameters and each depth value in the cluster, and constructing a standard response intensity sequence using the response intensity of the inflection point corresponding to the depth value in the cluster; calculating the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence, updating the initial parameters until the Pearson correlation coefficient is less than a threshold, and obtaining the optimal parameters, using the optimal parameters as the optimal stratification point of the logging curve, and the threshold is artificially set according to the actual situation, and exemplarily, the threshold is -0.8.

[0048] The calculation method of the initial parameters is:

[0049] The minimum value in the cluster is used as the initial parameter to obtain the difference sequence, and the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is calculated; the initial parameter is iteratively updated until the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is less than a threshold.

[0050] For example, let the initial parameters corresponding to the cluster be The depth of the optimal stratigraphic stratification point is , is between the maximum and minimum values ​​in the cluster. and the depth value in the cluster The distance is , For the The depth value on the well logging curve is also the depth value in the cluster; construct the difference sequence and standard response intensity sequence, where the difference sequence is , the standard response intensity sequence is , iteratively update the initial parameters The value of is used to calculate the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence until the Pearson correlation coefficient is less than the threshold. At this time, the value of the initial parameter is the optimal stratification point of the stratum. It should be noted that the optimal stratification point is the dividing point between strata. The threshold is set according to the actual situation, for example, the threshold is -0.8.

[0051] The initial parameters are updated iteratively The method for calculating the value of is as follows: Assume that the minimum value in the cluster is 10.0 and the maximum value is 15.0. The initial parameters The value of is 10.0, which is substituted into the difference sequence to calculate the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence. If the Pearson correlation coefficient is greater than -0.8, the value of the initial parameter is updated to 10.5, which is substituted into the difference sequence to calculate the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence. If the Pearson correlation coefficient is greater than -0.8, the value of the initial parameter is updated to 11, which is substituted into the difference sequence to calculate the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence. The initial parameters are updated in this way. The values ​​of are 10.0, 10.5, 11.0, 11.5, 12.0, ..., until the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is less than -0.8. It should be noted that the initial parameters The value of is always less than the maximum value of 15.0.

[0052] At the stratum level, the change rate and response intensity of the logging curve response intensity have changed greatly, and the difference sequence and the standard response intensity sequence are negatively correlated. When the Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is less than the threshold, it indicates that the initial parameters The data in the difference sequence can be understood as the distance between the calculated inflection point and the optimal stratification point.

[0053] In other embodiments, a genetic algorithm or a particle swarm algorithm may also be used to calculate the value of the initial parameter when the Pearson correlation coefficient of two sequences is less than a threshold.

[0054] It should be noted that calculating the Pearson correlation coefficient of two sequences is a prior art in the art, and the calculation process thereof will not be described in detail here.

[0055] In one embodiment, in response to the difference between the maximum value and the minimum value in the cluster being less than or equal to a preset threshold, the median of the depth values ​​in the cluster is used as a stratification point on the corresponding well logging curve.

[0056] In other embodiments, in response to the difference between the maximum value and the minimum value in the cluster being less than or equal to a preset threshold, the mean value of the depth values ​​in the cluster is used as the stratification point on the well logging curve.

[0057] The present invention draws multiple logging curves, refers to multiple logging phases, and clusters the stratification points of the formation, thereby further obtaining the optimal stratification point, reducing the influence of a single factor on the division result, and improving the accuracy of stratum division.

[0058] Figure 2 The structural block diagram of a well logging curve automatic drawing system in this embodiment is schematically shown.

[0059] The present invention also provides a well logging curve automatic drawing system. Figure 2 As shown, the automatic logging curve drawing system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, a stratigraphic division method based on multiple logging curves according to the present invention is implemented.

[0060] The automatic logging curve drawing system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and the configuration and functions of these components are known in the art, so they will not be described in detail here.

[0061] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0062] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for stratigraphic division based on multiple well logging curves, characterized in that: The following steps are involved: Draw logging curves according to different logging phases, including resistivity, acoustic wave velocity, formation density, and gamma ray; Obtain the inflection point on each logging curve, and cluster the depth values ​​of the inflection points on all logging curves to obtain multiple clusters; In response to the difference between the maximum value and the minimum value in the cluster being greater than a preset threshold, determining the best stratification point according to the depth value in the cluster, otherwise, taking the median or mean of the depth value in the cluster as the stratification point on the corresponding logging curve; Among them, the method for determining the optimal stratification point is: Set the initial parameters, use the absolute value of the difference between the initial parameters and each depth value in the cluster to construct a difference sequence, and use the response intensity of the inflection point corresponding to the depth value in the cluster to construct a standard response intensity sequence; The Pearson correlation coefficient between the difference sequence and the standard response intensity sequence is calculated, and the initial parameters are updated until the Pearson correlation coefficient is less than the threshold value to obtain the optimal parameters, which are used as the best stratification points of the logging curve.

2. A method for stratigraphic division based on multiple well logging curves according to claim 1, characterized in that: The method to construct the standard response intensity sequence is: The mean of the response intensity on the logging curve is calculated, and the ratio of the response intensity of the inflection point to the mean of the corresponding logging curve is taken as the standard response intensity of the inflection point. The standard response intensities of all inflection points in the cluster form a standard response intensity sequence.

3. The method for stratigraphic division based on multiple well logging curves according to claim 1, characterized in that: The method for obtaining the inflection point on each logging curve is: The moving average algorithm is used to smooth the logging curve, the second-order derivative of the smoothed curve is calculated, and the point where the second-order derivative is zero is taken as the inflection point.

4. The method for stratigraphic division based on multiple well logging curves according to claim 1, characterized in that: Before drawing the well logging curve, it also includes: The data of each logging phase is obtained, and the values ​​of the same logging phase data at different depths are clustered by mean shift. The data points that do not belong to any cluster in the clustering results are regarded as outliers, and then the outliers are deleted.

5. The method for stratigraphic division based on multiple well logging curves according to claim 1, characterized in that: The DBSCAN algorithm is used to cluster the depth values ​​of the inflection points.

6. The method for stratigraphic division based on multiple well logging curves according to claim 1, characterized in that: The initial parameters are between the maximum and minimum depth values ​​in the cluster.

7. A stratigraphic division system based on multiple well logging curves, characterized in that: include: processor; A memory storing computer program instructions, which, when executed by the processor, enables the system to execute a stratigraphic division method based on multiple well logging curves according to any one of claims 1 to 6.

Images