A method and system for identifying stratum structure based on K-means unsupervised learning

Abstract

The invention discloses a stratigraphic structure recognition method and system based on K-means unsupervised learning, and relates to the technical field of stratigraphic structure recognition. The method comprises: step S100, obtaining data of various drilling parameters of a drilling rig during drilling through a drilling test device, and constructing an initial data set; step S200, normalizing the initial data set to obtain a normalized data set for K-means unsupervised learning; step S300, setting a category K of a prediction result, a maximum number of iterations, and a deviation; step S400, randomly selecting K data points as initial clustering centers; step S500, respectively calculating the distance from each data point in the normalized data set to the initial clustering center, and classifying points whose distance is less than or equal to a preset distance threshold into one category; step S600, reselecting K data points as updated clustering centers; step S700, repeating step S500 for iteration; step S800, outputting a classification result after reaching the maximum number of iterations in step S300, thereby completing stratigraphic structure recognition.

Description

Technical Field

[0001] The present invention belongs to the technical field of stratum structure recognition, and more specifically, relates to a stratum structure recognition method and system based on K-means unsupervised learning. Background Art

[0002] Accurately identifying the stratigraphic structure is crucial for geotechnical engineering projects such as mining, tunneling, petroleum and slope engineering. Traditional stratigraphic structure identification mainly relies on geological analysis after coring, a process that usually requires manual interpretation, which is not only time-consuming but also consumes a lot of financial and human resources, and has limited accuracy under complex stratigraphic conditions. As an innovative in-situ testing technology, test-while-drilling technology can invert the stratigraphic structure by monitoring the drilling parameters of the drill rig, providing a basis for rapid and accurate stratigraphic identification on site. However, at present, this technology usually relies on a single drilling parameter or a simple mechanical model for inversion. Due to the interference of various external factors, it is difficult to guarantee the accuracy of the results, which limits the accuracy improvement of stratigraphic identification methods based on test-while-drilling technology.

[0003] Machine learning is dedicated to studying how to enable computer systems to automatically improve performance through experience, automatically identify patterns and regularities in data through algorithms, and learn based on them, and make decisions or predictions. In recent years, the rapid development of computer science has made it possible to use drilling test data as a machine learning data set to accurately and automatically intelligently identify stratum structures based on drilling tests. For example, patent 202111338652.2 discloses a drilling intelligent identification method based on a multi-layer perceptron. Based on supervised learning, the algorithm learns the relationship between drilling parameters and rock and soil properties by labeling the strata, and constructs a rock structure intelligent identification model. Patent 202110784665.6, also based on supervised learning, discloses a tunnel advance lithology drilling identification method. This method is based on the advance drilling data of the tunnel face. After determining the lithology of the strata in the drilling area, a relationship model between drilling data and lithology is constructed to identify the lithology. Once the model is built, these methods can intelligently identify the strata. However, both Patent 202111338652.2 and Patent 202110784665.6 rely on supervised learning algorithms and require a large amount of labeled data to train the model. The cost of labeling geological data is extremely high, and accurate labeled data is difficult to obtain in complex or unknown stratigraphic environments, which limits the application scope and effectiveness of these machine learning methods. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a stratum structure recognition method based on K-means unsupervised learning, comprising:

[0005] Step S100, obtaining data of various drilling parameters of the drilling rig during drilling through a drilling test device, and constructing an initial data set;

[0006] Step S200, normalizing the initial data set to obtain a normalized data set for K-means unsupervised learning;

[0007] Step S300, setting the category K of the prediction result, the maximum number of iterations and the deviation;

[0008] Step S400, randomly selecting K data points as initial cluster centers;

[0009] Step S500, respectively calculating the distance between each data point in the normalized data set and the initial cluster center, and classifying the points whose distance is less than or equal to a preset distance threshold into one category;

[0010] Step S600, reselect K data points as updated cluster centers;

[0011] Step S700, repeating step S500 for iteration;

[0012] Step S800, outputting the classification result after reaching the maximum number of iterations in step S300, thereby completing the stratum structure identification.

[0013] Furthermore, before step S100, the method further includes: calibrating the sensor of the while drilling test equipment before the first operation.

[0014] Furthermore, it also includes: step S900, if the formation structure to be identified is in the same formation, and only the formation at a certain depth changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, the formation is regarded as consistent with the original formation.

[0015] Furthermore, the method further includes: step S1000, performing weighted average processing on the data of the drilling parameters according to the classification result.

[0016] Furthermore, the method further includes: calculating anomaly measurement values ​​of data points in the classification results, and deleting data points whose anomaly measurement values ​​are greater than a preset anomaly measurement threshold from the categories corresponding thereto.

[0017] The present invention also proposes a stratum structure recognition system based on K-means unsupervised learning, comprising:

[0018] The normalization module is used to obtain the data of various drilling parameters of the drilling rig during drilling through the drilling test equipment, construct an initial data set, and normalize the initial data set to obtain a normalized data set for K-means unsupervised learning.

[0019] The clustering module is used to set the category K of the prediction result, the maximum number of iterations and the deviation, randomly select K data points as the initial clustering center, calculate the distance from each data point in the normalized data set to the initial clustering center, and classify the points with a distance less than or equal to a preset distance threshold into one category, reselect K data points as updated clustering centers, repeat the iteration until the maximum number of iterations is reached, and output the classification result, thereby completing the stratigraphic structure identification.

[0020] Furthermore, it also includes a calibration module for calibrating the sensor of the while drilling test equipment before the first operation.

[0021] Furthermore, it also includes: if the formation structure to be identified is in the same formation, and only the formation at a certain depth changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, the formation is regarded as consistent with the original formation.

[0022] Furthermore, the method further includes: performing weighted average processing on the data of the drilling parameters according to the classification results.

[0023] Furthermore, the method further includes: calculating anomaly measurement values ​​of data points in the classification results, and deleting data points whose anomaly measurement values ​​are greater than a preset anomaly measurement threshold from the categories corresponding thereto.

[0024] Compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0025] 1. The method of the present invention first constructs a data set through data collected by high-precision downhole testing equipment, and normalizes the data set. The data obtained by the high-precision downhole equipment provides a reliable guarantee for accurate identification. The normalization process is intended to avoid the dominance of certain features in the model training process due to their large value range. In addition, since the dimensions of downhole parameters vary, normalization can ensure that different parameters are compared and evaluated on the same scale. Normalization also helps to avoid numerical calculation problems such as overflow or underflow that may be caused by the inevitable extreme maximum and minimum values ​​in the data set, thereby laying the foundation for the accuracy of subsequent analysis results.

[0026] 2. The method of the present invention introduces the K-means unsupervised learning algorithm to effectively address the challenge of requiring a large amount of labeled data in supervised learning. Given that the labeling of geological data is costly and time-consuming, and that it is difficult to obtain accurately labeled data in complex or unknown stratigraphic environments, this method can significantly reduce costs and quickly and economically identify stratigraphic structures based on drilling data.

[0027] 3. The method of the present invention effectively reduces classification errors and the impact caused by changes in microscopic geological conditions or other random factors through a self-designed decision module. The decision module can filter out noise or errors that may be caused by random factors in the classification results, thereby improving the quality and practicality of stratum classification and ensuring the accuracy and reliability of the results.

[0028] 4. The method of the present invention can accurately identify key layers that are not shown in the geological profile, provided that the large-scale results are consistent with the geological profile. This feature provides a strong guarantee for the identification of broken and weak rock layers and their strength parameters, as well as the accurate construction of geological models. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0030] Figure 2 is a system structure diagram of Embodiment 2 of the present invention;

[0031] Figure 3 This is a graph of the original depth sequence of the borehole in Example 1 of the present invention;

[0032] Figure 4 It is a schematic diagram of the comparison between the recognition result of the K-means unsupervised learning algorithm and the original geological profile in Example 1 of the present invention. DETAILED DESCRIPTION

[0033] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0034] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0035] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0036] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0037] The display is used to show the interactive cross-section of each application.

[0038] All subscripts in the formulas of the present invention are only used to distinguish parameters and have no actual meaning.

[0039] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0040] Example 1

[0041] like Figure 1 As shown, the present invention proposes a stratum structure recognition method based on K-means unsupervised learning, comprising:

[0042] Step S100: Before the first operation, the sensor of the drilling test equipment needs to be calibrated to ensure the accuracy of the drilling test data. Then, the high-precision drilling test equipment is used to obtain the data of various drilling parameters such as drilling pressure, torque, drilling speed, rotation speed, etc. during the drilling of the drilling rig (such as Figure 3 As shown in Figure 3), construct the initial dataset.

[0043] Step S200: Normalize the initial data set to obtain a normalized data set for the K-means unsupervised algorithm. The normalization method can be selected according to the actual situation, mainly including the following methods:

[0044] 1. Min-Max normalization:

[0045] x' = (x - min(x)) / (max(x) - min(x)) (1),

[0046] After normalization, the range is [0, 1], where min(x) and max(x) exclude the influence of extreme values.

[0047] 2. Mean normalization:

[0048] x' = (x - mean(x)) / (max(x) - min(x)) (2),

[0049] The normalized range is [-1, 1], where mean(x) represents the mean value of the data.

[0050] 3. Z-scores normalization:

[0051] x' = (x - μ) / σ (3),

[0052] The normalized range is the entire set of real numbers, where μ and σ represent the mean and standard deviation of the data, respectively.

[0053] Step S300: Set the category K, maximum number of iterations and deviation of the prediction results according to the requirements. The number of categories K mainly depends on the diversity of rock types and the requirements for classification accuracy. Assuming that there are three main strata in a specific area, such as granite, basalt and limestone, K can be set to 3. If the preliminary analysis shows that there may be more strata, the value of K can be adjusted according to actual needs. The maximum number of iterations is used to limit the execution time of the K-means algorithm to prevent the algorithm from failing to converge quickly in some cases. The common maximum number of iterations is set between 100 and 300. For example, setting the maximum number of iterations to 200 means that the algorithm will run a maximum of 200 iterations, and stop early if it has converged before then. The deviation (or tolerance threshold) is used to determine whether the algorithm has converged. Generally speaking, if the position change of the centroid in two consecutive iterations is less than the set deviation value, the algorithm will stop. A commonly used deviation value may be 0.0001 or 0.001. For example, setting the bias to 0.0001 means that if the position of the centroid changes by less than 0.0001, the algorithm is considered to have converged and stops iterating.

[0054] Step S400: Randomly select K data points as initial cluster centers.

[0055] Step S500: Calculate the distance from each point to the center respectively, and classify the points close to the center into one category.

[0056] For the i-th type of stratum, the calculation formula of the intra-cluster sum of squares is shown in formula (4):

[0057] ,

[0058] The calculation formula for the sum of the intra-cluster square sums of all strata is shown in formula (5):

[0059] ,

[0060] In the formula represents the set of data points of the i-th type of stratum, is the data point of type j stratum, is the center point of the i-th type of stratum, and K represents the number of stratum categories.

[0061] Step S600: Reassign and update the center point.

[0062] Step S700: Repeat step S500 for iteration.

[0063] Step S800: Output the classification result after reaching the maximum iteration step in S300 (such as Figure 4 as shown).

[0064] Step S900: The decision module adopts a judgment standard, that is, if in the same stratum, only a certain depth stratum changes or the number of consecutive changes is less than the decision threshold n, these strata are regarded as consistent with the original stratum. The value of n needs to be comprehensively considered in combination with actual geological conditions and engineering requirements.

[0065] Step S1000: The final prediction and recognition result is obtained and output, and the drilling parameters are weighted averaged according to the prediction result (different rock formations have different geological and mechanical properties, so the importance of drilling parameters (such as drilling speed, torque, drilling pressure, etc.) in different rock formations will also be different. By weighted averaging different rock formations, the impact of each rock formation on the drilling process can be more accurately reflected.), the weighted average is shown in the following formula:

[0066] (6),

[0067] Where w i is the proportion of the i-th stratum in the total stratum.

[0068] In order to make the classification result more accurate, this embodiment calculates the abnormality metric value of the data point in the classification result, and deletes the data point whose abnormality metric value is greater than a preset abnormality metric threshold from the corresponding category.

[0069] Calculating anomaly metrics includes:

[0070] ,

[0071] in, is the data point in the classification result The abnormality measure of is the data point in the classification result After projection, is the mean of the projected data points, is the standard deviation of the projected data points, is the adjustment factor of the projected data point, is the distance adjustment factor, is the dimension of the projected data point, The classification result is Data points The adjacent Data points distance.

[0072] Data points in the classification results The projected data points Specifically:

[0073] ,

[0074] in, The classification result is Data points The first adjustment factor is The classification result is Data points The second adjustment factor is The classification result is Data points The third adjustment factor is The classification result is Data points The fourth adjustment factor is The classification result is Data points The fifth adjustment factor, The classification result is Data points The sixth adjustment factor is The classification result is Data points The seventh adjustment factor.

[0075] The classification results Data points The adjacent Data points Distance Specifically:

[0076] ,

[0077] in, The classification result is Data points The adjacent Data points The first adjustment factor for the distance, The classification result is Data points The adjacent Data points The second adjustment factor for the distance, The classification result is Data points The adjacent Data points The third adjustment factor for the distance, The classification result is Data points The adjacent Data points The fourth adjustment factor for the distance, The classification result is Data points The adjacent Data points The fifth adjustment factor of the distance, The classification result is Data points The adjacent Data points The sixth adjustment factor for the distance, The classification result is Data points The adjacent Data points The seventh adjustment factor for the distance.

[0078] The preset anomaly measurement thresholds include:

[0079] ,

[0080] in, For time The preset abnormal measurement threshold when is the initial preset abnormal measurement threshold, is the time adjustment factor.

[0081] Example 2

[0082] like Figure 2 As shown, the embodiment of the present invention also proposes a stratum structure recognition system based on K-means unsupervised learning, including:

[0083] The calibration module is used to calibrate the sensors of the test while drilling equipment before the first operation.

[0084] A normalization module is used to obtain data of various drilling parameters of the drilling rig during drilling through a drilling test device, construct an initial data set, and normalize the initial data set to obtain a normalized data set for K-means unsupervised learning;

[0085] The clustering module is used to set the category K of the prediction result, the maximum number of iterations and the deviation, randomly select K data points as the initial clustering center, calculate the distance from each data point in the normalized data set to the initial clustering center, and classify the points with a distance less than or equal to a preset distance threshold into one category, reselect K data points as updated clustering centers, repeat the iteration until the maximum number of iterations is reached, and output the classification result, thereby completing the stratigraphic structure identification.

[0086] Specifically, it also includes: if the formation structure to be identified is in the same formation, and only a certain depth formation changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, the formation is regarded as consistent with the original formation.

[0087] Specifically, it also includes: performing weighted average processing on the data of the drilling parameters according to the classification results.

[0088] Specifically, it also includes: calculating the abnormal measurement value of the data point in the classification result, and deleting the data point whose abnormal measurement value is greater than a preset abnormal measurement threshold from the category corresponding to it.

[0089] Example 3

[0090] The embodiment of the present invention further provides a storage medium storing a plurality of instructions, wherein the instructions are used to implement the method for identifying stratum structure based on K-means unsupervised learning.

[0091] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0092] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: calibrating sensors of the test while drilling equipment before the first operation;

[0093] Step S100, obtaining data of various drilling parameters of the drilling rig during drilling through a drilling test device, and constructing an initial data set;

[0094] Step S200, normalizing the initial data set to obtain a normalized data set for K-means unsupervised learning;

[0095] Step S300, setting the category K of the prediction result, the maximum number of iterations and the deviation;

[0096] Step S400, randomly selecting K data points as initial cluster centers;

[0097] Step S500, respectively calculating the distance between each data point in the normalized data set and the initial cluster center, and classifying the points whose distance is less than or equal to a preset distance threshold into one category;

[0098] Step S600, reselect K data points as updated cluster centers;

[0099] Step S700, repeating step S500 for iteration;

[0100] Step S800, outputting the classification result after reaching the maximum number of iterations in step S300, thereby completing the stratum structure identification.

[0101] Step S900: If the formation structure to be identified is in the same formation, and only a certain depth formation changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, the formation is considered to be consistent with the original formation.

[0102] Step S1000: performing weighted average processing on the data of the drilling parameters according to the classification results.

[0103] Specifically, it also includes: calculating the abnormal measurement value of the data point in the classification result, and deleting the data point whose abnormal measurement value is greater than a preset abnormal measurement threshold from the category corresponding to it.

[0104] Example 4

[0105] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores multiple instructions, and the instructions can be loaded and executed by the processor so that the processor can execute a stratigraphic structure identification method based on K-means unsupervised learning.

[0106] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0107] Among them, the storage medium can be used to store software programs and modules, such as a stratigraphic structure identification method based on K-means unsupervised learning in an embodiment of the present invention, and corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned stratigraphic structure identification method based on K-means unsupervised learning. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0108] The processor may call the information and application program stored in the storage medium through the transmission system to execute the steps of: calibrating the sensor of the test while drilling equipment before the first operation;

[0109] Step S100, obtaining data of various drilling parameters of the drilling rig during drilling through a drilling test device, and constructing an initial data set;

[0110] Step S200, normalizing the initial data set to obtain a normalized data set for K-means unsupervised learning;

[0111] Step S300, setting the category K of the prediction result, the maximum number of iterations and the deviation;

[0112] Step S400, randomly selecting K data points as initial cluster centers;

[0113] Step S500, respectively calculating the distance between each data point in the normalized data set and the initial cluster center, and classifying the points whose distance is less than or equal to a preset distance threshold into one category;

[0114] Step S600, reselect K data points as updated cluster centers;

[0115] Step S700, repeating step S500 for iteration;

[0116] Step S800, outputting the classification result after reaching the maximum number of iterations in step S300, thereby completing the stratum structure identification.

[0117] Step S900: If the formation structure to be identified is in the same formation, and only a certain depth formation changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, the formation is considered to be consistent with the original formation.

[0118] Step S1000: performing weighted average processing on the data of the drilling parameters according to the classification results.

[0119] Specifically, it also includes: calculating the abnormal measurement value of the data point in the classification result, and deleting the data point whose abnormal measurement value is greater than a preset abnormal measurement threshold from the category corresponding to it.

[0120] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0121] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0122] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0123] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0125] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) 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, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0126] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for identifying stratum structure based on K-means unsupervised learning, characterized in that: include: Step S100, obtaining data of various drilling parameters of the drilling rig during drilling through a drilling test device, and constructing an initial data set; Step S200, normalizing the initial data set to obtain a normalized data set for K-means unsupervised learning; Step S300, setting the category K of the prediction result, the maximum number of iterations and the deviation; Step S400, randomly selecting K data points as initial cluster centers; Step S500, respectively calculating the distance between each data point in the normalized data set and the initial cluster center, and classifying the points whose distance is less than or equal to a preset distance threshold into one category; Step S600, reselect K data points as updated cluster centers; Step S700, repeating step S500 for iteration; Step S800, outputting the classification result after reaching the maximum number of iterations in step S300, thereby completing the stratum structure identification; Calculate the abnormality metric value of the data point in the classification result, and delete the data point whose abnormality metric value is greater than a preset abnormality metric threshold from the category corresponding to it; Calculating anomaly metrics includes: ， in, is the data point in the classification result The abnormality measure of is the data point in the classification result After projection, is the mean of the projected data points, is the standard deviation of the projected data points, is the adjustment factor of the projected data point, is the distance adjustment factor, is the dimension of the projected data point, The classification result is Data points The adjacent Data points distance; Data points in the classification results The projected data points Specifically: ， in, The classification result is Data points The first adjustment factor is The classification result is Data points The second adjustment factor is The classification result is Data points The third adjustment factor is The classification result is Data points The fourth adjustment factor is The classification result is Data points The fifth adjustment factor, The classification result is Data points The sixth adjustment factor is The classification result is Data points The seventh adjustment factor of The classification results Data points The adjacent Data points Distance Specifically: ， in, The classification result is Data points The adjacent Data points The first adjustment factor for the distance, The classification result is Data points The adjacent Data points The second adjustment factor for the distance, The classification result is Data points The adjacent Data points The third adjustment factor for the distance, The classification result is Data points The adjacent Data points The fourth adjustment factor for the distance, The classification result is Data points The adjacent Data points The fifth adjustment factor of the distance, The classification result is Data points The adjacent Data points The sixth adjustment factor for the distance, The classification result is Data points The adjacent Data points The seventh adjustment factor of the distance; The preset anomaly measurement thresholds include: ， in, For time The preset abnormal measurement threshold when is the initial preset abnormal measurement threshold, is the time adjustment factor.

2. A method for identifying stratum structure based on K-means unsupervised learning as claimed in claim 1, characterized in that: Before step S100, the method further includes: calibrating the sensor of the while drilling test equipment before the first operation.

3. A method for identifying stratum structure based on K-means unsupervised learning as claimed in claim 1, characterized in that: It also includes: step S900, if the formation structure to be identified is in the same formation, and only a certain depth formation changes or the number of sequences of data of various types of drilling parameters that change continuously is less than a decision threshold, then the formation is regarded as consistent with the original formation.

4. A method for identifying stratum structure based on K-means unsupervised learning as claimed in claim 1, characterized in that: The method further includes: step S1000, performing weighted average processing on the data of the drilling parameters according to the classification result.

5. A stratum structure recognition system based on K-means unsupervised learning, characterized in that: include: A normalization module is used to obtain data of various drilling parameters of the drilling rig during drilling through a drilling test device, construct an initial data set, and normalize the initial data set to obtain a normalized data set for K-means unsupervised learning; The clustering module is used to set the category K of the prediction result, the maximum number of iterations and the deviation, randomly select K data points as the initial clustering center, calculate the distance from each data point in the normalized data set to the initial clustering center, and classify the points whose distance is less than or equal to the preset distance threshold into one category, reselect K data points as updated clustering centers, repeat the iteration until the maximum number of iterations is reached, and output the classification result, thereby completing the stratigraphic structure identification; Calculate the abnormality metric value of the data point in the classification result, and delete the data point whose abnormality metric value is greater than a preset abnormality metric threshold from the category corresponding to it; Calculating anomaly metrics includes: ， in, is the data point in the classification result The abnormality measure of is the data point in the classification result After projection, is the mean of the projected data points, is the standard deviation of the projected data points, is the adjustment factor of the projected data point, is the distance adjustment factor, is the dimension of the projected data point, The classification result is Data points The adjacent Data points distance; Data points in the classification results The projected data points Specifically: ， in, The classification result is Data points The first adjustment factor is The classification result is Data points The second adjustment factor is The classification result is Data points The third adjustment factor is The classification result is Data points The fourth adjustment factor is The classification result is Data points The fifth adjustment factor, The classification result is Data points The sixth adjustment factor is The classification result is Data points The seventh adjustment factor of The classification results Data points The adjacent Data points Distance Specifically: ， in, The classification result is Data points The adjacent Data points The first adjustment factor for the distance, The classification result is Data points The adjacent Data points The second adjustment factor for the distance, The classification result is Data points The adjacent Data points The third adjustment factor for the distance, The classification result is Data points The adjacent Data points The fourth adjustment factor for the distance, The classification result is Data points The adjacent Data points The fifth adjustment factor of the distance, The classification result is Data points The adjacent Data points The sixth adjustment factor for the distance, The classification result is Data points The adjacent Data points The seventh adjustment factor of the distance; The preset anomaly measurement thresholds include: ， in, For time The preset abnormal measurement threshold when is the initial preset abnormal measurement threshold, is the time adjustment factor.

6. A stratum structure recognition system based on K-means unsupervised learning as claimed in claim 5, characterized in that: It also includes a calibration module for calibrating the sensors of the while drilling test equipment before the first operation.

7. A stratum structure recognition system based on K-means unsupervised learning as claimed in claim 5, characterized in that: Also includes: If the formation structure to be identified is in the same formation, and only the formation at a certain depth changes or the number of sequences of data of various types of drilling parameters that change continuously is less than the decision threshold, the formation is considered to be consistent with the original formation.

8. The stratum structure recognition system based on K-means unsupervised learning as claimed in claim 5, characterized in that: Also includes: The data of the drilling parameters are processed by weighted average according to the classification results.

Images