Micro drilling speed classification method, system and equipment based on logging data and medium

Abstract

The invention discloses a micro drilling speed classification method, system and equipment based on logging data and a medium, and relates to the technical field of petroleum and natural gas drilling engineering. The method comprises the following steps: constructing a drilling parameter time sequence database on the basis of logging data; calculating the micro-drilling speed corresponding to each group of data, constructing a time-drilling parameter-micro-drilling speed matrix, and numbering each group of data in the matrix; dividing the matrix into a drilling sample and a non-drilling sample, and extracting a bit pressure-micro drilling speed sample in the drilling sample to form a bit pressure-micro drilling speed curve; segmenting the bit pressure-micro drilling speed sample to form a normal drilling sample and an abnormal drilling sample; and according to the number of the segmentation position, a drilling parameter acquisition time period corresponding to the normal drilling sample is obtained, and the micro drilling speed and the drilling parameters are unified to the same time dimension. According to the method, the drilling parameters and the micro-drilling speed can be quickly unified on the time dimension, the mapping relation between the drilling parameters and the micro-drilling speed is more accurately established, and a basis is provided for optimizing the drilling parameters.

Description

Technical Field

[0001] This invention relates to the field of oil and gas drilling engineering technology, and in particular to a micro-drilling rate classification method, system, equipment and medium based on logging data. Background Technology

[0002] In oil and gas drilling operations, drilling parameters (such as rotational speed, torque, pump pressure, and displacement) are important factors affecting drilling efficiency and cost. Drilling speed is a direct reflection of drilling efficiency. Optimizing drilling parameters to improve drilling speed is a common method in oil and gas drilling operations. Traditional drilling parameter optimization methods mainly rely on human experience and post-analysis. This method is difficult to capture and adjust changes in the drilling process in real time, often leading to low drilling efficiency and even potential safety hazards.

[0003] With the development of sensing and data acquisition technologies, well logging data can be acquired in real time during drilling. However, currently, drilling parameters are acquired every few seconds in the form of time-series data, while drilling speed is acquired every tens of minutes or even an hour in the form of whole-meter data. This inconsistency in acquisition format and frequency makes it difficult to accurately analyze the relationship between the two. Furthermore, well logging data involves numerous drilling parameters, and the computational workload for optimizing these parameters is enormous, making real-time optimization impossible. Therefore, there is an urgent need for a technology that can quickly unify drilling parameters and drilling speed in the time dimension, thereby establishing a more accurate mapping relationship between the two to optimize drilling parameters in real time and achieve drilling speed improvement.

[0004] Chinese patent document CN114427430A, published on May 3, 2022, discloses a method and system for real-time collaborative drilling parameter optimization across multiple wells. The method includes: real-time acquisition of drilling data from target wells within the same block; extraction of real-time drilling parameter combinations and applied mechanical drilling rates (MRDs) for each target well; establishment of a real-time functional relationship between the MRD and the drilling parameters; optimization of the real-time functional relationship to obtain optimized combinations of drilling parameters for each drilling formation; establishment of historical functional relationships based on historical data from historical wells within the same block, followed by calculation of the expected MRD for each drilling formation; calculation of the actual MRD for each target well after applying the optimized drilling parameter combinations to each drilling formation, and comparison of this actual MRD with the expected MRD; using this ratio as a characterization parameter for abnormal drilling rates and excessive drill bit wear, and pushing corresponding handling solutions. This invention solves the problems of regional differences and difficulty in obtaining certain parameters and discrimination thresholds in existing methods, improving the applicability of the method model.

[0005] However, although the above technical solutions can establish a real-time functional relationship between the mechanical drilling rate and the drilling parameters for each drilling formation, establishing a real-time functional relationship between the mechanical drilling rate and the drilling parameters requires a large amount of calculation and model updates, which increases the consumption of computing resources. Furthermore, because the data contains noise and outliers, it is difficult to handle when reflecting the real-time functional relationship, which may affect the accuracy and stability of the model. More complex machine learning or data analysis algorithms are required, which increases the complexity and cost of technical implementation. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention provides a micro-drilling rate classification method, system, equipment, and medium based on logging data. This method unifies drilling parameters and drilling rate data to the same time dimension, classifies drilling rate samples using only a few drilling parameters such as drilling pressure to reduce computational complexity and achieve real-time optimization. It also distinguishes non-drilling parameter factors that affect drilling rate, obtaining normal drilling samples and stuck pipe risk analysis samples. This enables more precise optimization of drilling parameters at a finer-grained time scale, lays the foundation for subsequent complex analyses, reduces drilling costs, and improves the safety of drilling operations.

[0007] This invention is achieved through the following technical solution: The first aspect of this invention provides a method for classifying micro-drilling rates based on well logging data, comprising the following steps: Step 1: Construct a time-series database of drilling parameters based on logging data; Step 2: Based on the drilling parameter time series database, calculate the micro-drilling rate corresponding to each group of drilling parameters according to the time interval benchmark, form a drilling rate sample with time as the recording dimension, construct a time-drilling parameter-micro-drilling rate matrix, and number each group of data in the matrix sequentially as a unique identifier; the time interval benchmark refers to the longest time span in the drilling parameter time series database when the drilling pressure is greater than 0 and the drill bit position remains unchanged. Step 3: Divide the time-drilling parameter-micro-rate of drilling matrix into drilling samples and non-drilling samples, extract the bit pressure-micro-rate of drilling samples from the drilling samples, and form the bit pressure-micro-rate of drilling curve. Step 4: Based on image recognition methods, identify the abrupt change points of micro-drilling rate in the drill pressure-micro-drilling rate analysis sample, and segment the drill pressure-micro-drilling rate sample according to the abrupt change points to form normal drilling samples and abnormal drilling samples. Step 5: Automatically identify the unique identifier number corresponding to the segmentation position, and automatically determine the corresponding drilling parameter time series database acquisition time through the time-drilling parameter-micro-drilling rate matrix to obtain the drilling parameter acquisition time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters to the same time dimension.

[0008] The sampling frequency of the logging data is 1-10Hz.

[0009] In step 2, for the data group with unique identifier number i, its corresponding time t i Micro-drilling speed V ti The calculation formula is as follows: V ti =(X （ti-T） -X ti ) / T type 1 In Equation 1, V ti For t i Micro-drilling speed at any given moment; X （ti-T） For t i -Drill bit position at time T; X ti For t i The drill bit position at any given time; T is the time interval reference.

[0010] In step 3, dividing the time-drilling parameter-micro-drilling rate matrix into drilling samples and non-drilling samples means that the time-drilling parameter-micro-drilling rate matrix data that meets the conditions of drilling pressure greater than 0 and micro-drilling rate greater than or equal to 0 are divided into drilling samples, and the rest are divided into non-drilling samples.

[0011] In step 4, the image recognition method is as follows: Traverse all points on the curve, connect each interval point on the curve to form an interval line, and at the same time draw a perpendicular line through the midpoint to the line connecting the interval points to form a perpendicular line, until all points on the curve have been traversed. The lengths of the perpendiculars are denoted as M. １ M ２ M ３ …、M n The lengths of the lines connecting the interval points are denoted as N. １ N ２ N ３ …、N n (For example: There are three consecutive points A, B, and C. A and C are called interval points, B is the middle point, and line segment AC is the interval line. Draw a perpendicular line from point B to line segment AC, with the perpendicular point being D. Then line segment BD is the perpendicular line.) Calculate the median M of the perpendicular. Ｐ Mean M Ｑ The median N of the interval line Ｐ Mean N Ｑ ; M Ｑ =(M １ +M ２ +M ３ +……+M n ) / n formula 2 N Ｑ =(N１ +N ２ +N ３ +……+N n ) / n formula 3 In Equations 2 and 3, M Ｑ M is the average length of all perpendicular line segments; n The length of the perpendicular line corresponding to the data group with the unique identifier number n; N n The length of the interval line that uniquely identifies the data group with sequence number n; Calculate the standard deviation M of the vertical lines using the median as a reference. DP The standard deviation N of the interval line DP Calculate the standard deviation M of the vertical lines using the mean as a reference. DQ The standard deviation N of the interval line DQ Take M DP With M DQ The smaller value in is denoted as M. D0 N DP With N DQ The smaller value is denoted as N. D0 ; M DP =[(M1-M P ) 2 +(M2-M P ) 2 +(M3-M P ) 2 +……+(M n -M P ) 2 ] / n formula 4 N DP =[(N1-N P ) 2 +(N2-N P ) 2 +(N3-N P ) 2 +……+(N n -N P ) 2 ] / n formula 5 M DQ =[(M1-M Q ) 2 +(M2-M Q ) 2 +(M3-M Q ) 2 +……+(M n -M Q ) 2 ] / n formula 6 N DQ =[(N1-N Q )2 +(N2-N Q ) 2 +(N3-N Q ) 2 +……+(N n -N Q ) 2 ] / n formula 7 For a point on the curve with a unique identifier number i, if M i / N i >M D0 / N D0 If so, then the point is determined to be a mutation point, and the time t at that point is recorded. j If the condition is true, then the point is not considered a mutation point; Suppose there are j+1 mutation points, then there are j time intervals. Calculate the average value Δt across all time intervals: △t = (△t1 + ... + △t) j ) / j style 8 Equation 9: △t1=t2-t1 △t j =t j+1 -t j Formula 10 In Equations 8, 9, and 10, Δt is the average value of all time intervals; Δt1 is the first time interval; Δt j t is the j-th time interval; t1 is the time corresponding to the first mutation point; t j+1 This is the time corresponding to the (j+1)th mutation point; If the time interval between two adjacent abrupt change points is greater than the average value, then the time-drilling parameter-micro-drilling rate matrix within that time range is considered a normal drilling sample; otherwise, the curve is considered an abnormal drilling sample.

[0012] Step 5 specifically includes: Identify the unique identifier number corresponding to the segmentation position based on the mutation point; Based on the unique identifier of the segmentation location, extract the drilling parameters within the time period corresponding to the unique identifier from the time-drilling parameter-micro-drilling rate matrix constructed in step 2. Construct a dataset that combines micro-drilling rate and drilling parameters in the same time dimension under normal drilling conditions.

[0013] The non-drilling samples or abnormal drilling samples constitute a stuck drill risk analysis sample library, which is used for stuck drill risk analysis.

[0014] A second aspect of the present invention provides a micro-drilling rate classification system based on well logging data, including... A time-series database construction module that collects logging data and builds a drilling parameter time-series database based on the logging data; A micro-drilling rate calculation module is used to calculate the micro-drilling rate, construct a time-drilling parameter-micro-drilling rate matrix, and assign a unique identifier to each set of data in the matrix. The time-drilling parameter-micro-drilling rate matrix is ​​divided into drilling samples and non-drilling samples. Drilling pressure-micro-drilling rate samples are extracted from the drilling samples to form a sample division and curve generation module for drilling pressure-micro-drilling rate curves. Based on image recognition methods, the abrupt change points of micro-drilling rate in the drill pressure-micro-drilling rate analysis sample are identified. The drill pressure-micro-drilling rate sample is then segmented according to the abrupt change points to form a module for identifying abrupt change points and segmenting samples of normal drilling samples and abnormal drilling samples. The system automatically identifies the unique identifier corresponding to the mutation point, obtains the collection time in the corresponding time series database through the time-drilling parameter-micro-drilling rate matrix, and obtains the drilling parameter collection time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters into the same time dimension data unification module.

[0015] A third aspect of the present invention provides a computer device including a processor, an input device, an output device, and a memory, wherein the processor, the input device, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to perform some or all of the steps as described in the first aspect of the present invention.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in the first aspect of the present invention.

[0017] The beneficial effects of this invention are mainly reflected in the following aspects: 1. This invention can unify drilling parameters and drilling speed to the same time dimension in real time, and on this basis, establish a more accurate mapping relationship between drilling parameters and drilling speed, eliminating interference items such as outliers, thereby providing a basis for more accurate optimization of drilling parameters and achieving speed-up effect.

[0018] 2. This invention, by setting the sampling frequency range of logging data to 1-10Hz, can meet the high-frequency sampling requirements of the drilling process, which helps to capture minute data changes in a timely manner and makes the micro-drilling rate classification more accurate.

[0019] 3. The present invention uses the longest time span when the drilling pressure is greater than 0 and the drill bit position remains unchanged as the time interval benchmark for calculating the micro-drilling rate. This can eliminate the situation where the micro-drilling rate is 0 due to the sampling frequency being faster than the drill bit position change frequency, and ensure the integrity and validity of the micro-drilling rate sample.

[0020] 4. This invention uses image recognition to accurately identify abrupt changes in the drill pressure-micro-drilling rate curve. It does not rely on traditional manual analysis or complex algorithms. By determining abrupt changes through the ratio of the lengths of the interval lines and vertical lines, it can quickly screen out significant changes in drilling rate, helping to promptly detect drilling anomalies. Using the median and standard deviation of the mean of the vertical and interval lines to screen abrupt changes reduces the impact of random fluctuations and noise interference during drilling on the analysis results, ensuring the accuracy of abrupt change point identification. Furthermore, it requires no complex manual intervention; abrupt changes can be automatically detected through traversal and calculation, and the normal drilling samples and abnormal drilling samples can be distinguished, improving the automation level of data processing and reducing the time cost of manual analysis.

[0021] 5. This invention constructs a stuck pipe risk analysis sample library from non-drilling samples or abnormal drilling samples, which can provide data support for subsequent stuck pipe risk analysis, help improve the accuracy of subsequent stuck pipe risk prediction, and enhance the safety of the drilling process. Attached Figure Description

[0022] The present invention will now be further described in detail with reference to the accompanying drawings and specific embodiments: Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the time-drilling parameter-micro-drilling rate matrix of the present invention; Figure 3 This is a graph showing the drilling pressure-micro-drilling speed of the present invention. Detailed Implementation

[0023] Example 1 As the most basic embodiment of the present invention, this embodiment provides a micro-drilling rate classification method based on well logging data, including the following steps: Step 1: Construct a time-series database of drilling parameters based on logging data; Step 2: Based on the drilling parameter time series database, calculate the micro-rate of drilling for each group of drilling parameters with a duration no less than the time interval benchmark, forming a drilling rate sample with time as the recording dimension. Construct a time-drilling parameter-micro-rate of drilling matrix, and sequentially number each group of data in the matrix as a unique identifier. This matrix is ​​an array containing micro-rate of drilling and all drilling parameters based on time, and is sequentially numbered in chronological order as a unique identifier for each group of drilling parameters, thus establishing a one-to-one correspondence between drilling rate and drilling parameters. The time interval benchmark refers to the longest time span in the drilling parameter time series database when the drilling pressure is greater than 0 and the drill bit position remains unchanged. It is used as the time interval benchmark for calculating micro-rate of drilling. Its function is to eliminate micro-rate of drilling of 0 caused by the acquisition frequency being faster than the drill bit position change frequency, ensuring the integrity of the drilling rate sample. Step 3: Divide the time-drilling parameter-micro-rate of drilling matrix into drilling samples and non-drilling samples, extract the bit pressure-micro-rate of drilling samples from the drilling samples, and form the bit pressure-micro-rate of drilling curve; accurately identify the bit pressure-micro-rate of drilling conditions analysis samples in the drilling parameter time series database (logging data) that need to participate in the drilling parameter optimization analysis. Step 4: Based on mathematical or image recognition methods, determine the range of abrupt changes in micro-drilling rate in the drill pressure-micro-drilling rate analysis sample. Segment the drill pressure-micro-drilling rate sample according to the abrupt change points to form normal drilling samples and abnormal drilling samples. The normal drilling samples further eliminate interference items such as abnormal points in the drill pressure-micro-drilling rate analysis samples and are the basis for drilling parameter optimization. The abnormal drilling samples and non-drilling samples together constitute the stuck pipe risk analysis sample, which is the basis for subsequent stuck pipe risk analysis. Step 5: Automatically identify the unique identifier number corresponding to the segmentation position, and automatically determine the corresponding drilling parameter time series database acquisition time through the time-drilling parameter-micro-drilling rate matrix to obtain the drilling parameter acquisition time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters to the same time dimension.

[0024] Example 2 As a preferred embodiment of the present invention, this embodiment provides a micro-drilling rate classification method based on well logging data, including the following steps: The sampling frequency of the logging data is 1-10Hz.

[0025] In step 2, for the data group with unique identifier number i, its corresponding time t i Micro-drilling speed V ti The calculation formula is as follows: V ti =(X （ti-T） -X ti ) / T type 1 In Equation 1, V ti For t i Micro-drilling speed at any given moment; X （ti-T） For t i -Drill bit position at time T; X ti For t i The drill bit position at any given time; T is the time interval reference.

[0026] In step 3, dividing the time-drilling parameter-micro-drilling rate matrix into drilling samples and non-drilling samples means that the time-drilling parameter-micro-drilling rate matrix data that meets the conditions of drilling pressure greater than 0 and micro-drilling rate greater than or equal to 0 are divided into drilling samples, and the rest are divided into non-drilling samples.

[0027] In step 4, the image recognition method is as follows: Traverse all points on the curve, connect each interval point on the curve to form an interval line, and at the same time draw a perpendicular line through the midpoint to the line connecting the interval points to form a perpendicular line, until all points on the curve have been traversed. The lengths of the perpendiculars are denoted as M. １ M ２ M ３ …、M n The lengths of the lines connecting the interval points are denoted as N. １ N ２ N ３ …、N n (For example: There are three consecutive points A, B, and C. A and C are called interval points, B is the middle point, and line segment AC is the interval line. Draw a perpendicular line from point B to line segment AC, with the perpendicular point being D. Then line segment BD is the perpendicular line.) Calculate the median M of the perpendicular. Ｐ Mean M Ｑ The median N of the interval line Ｐ Mean N Ｑ ; M Ｑ =(M １ +M ２ +M ３ +……+M n ) / n formula 2 N Ｑ =(N １ +N ２ +N ３ +……+N n ) / n formula 3 In Equations 2 and 3, M Ｑ M is the average length of all perpendicular line segments; n The length of the perpendicular line corresponding to the data group with the unique identifier number n; N n The length of the interval line that uniquely identifies the data group with sequence number n; Calculate the standard deviation M of the vertical lines using the median as a reference. DP The standard deviation N of the interval line DP Calculate the standard deviation M of the vertical lines using the mean as a reference. DQ The standard deviation N of the interval line DQ Take M DP With M DQ The smaller value in is denoted as M. D0 N DP With N DQ The smaller value is denoted as N. D0 ; M DP =[(M1-M P ) 2 +(M2-M P )2 +(M3-M P ) 2 +……+(M n -M P ) 2 ] / n formula 4 N DP =[(N1-N P ) 2 +(N2-N P ) 2 +(N3-N P ) 2 +……+(N n -N P ) 2 ] / n formula 5 M DQ =[(M1-M Q ) 2 +(M2-M Q ) 2 +(M3-M Q ) 2 +……+(M n -M Q ) 2 ] / n formula 6 N DQ =[(N1-N Q ) 2 +(N2-N Q ) 2 +(N3-N Q ) 2 +……+(N n -N Q ) 2 ] / n formula 7 For a point on the curve with a unique identifier number i, if M i / N i >M D0 / N D0 If so, then the point is determined to be a mutation point, and the time t at that point is recorded. j If the condition is true, then the point is not considered a mutation point; Suppose there are j+1 mutation points, then there are j time intervals. Calculate the average value Δt across all time intervals: △t = (△t1 + ... + △t) j ) / j style 8 Equation 9: △t1=t2-t1 △t j =t j+1 -t j Formula 10 In Equations 8, 9, and 10, Δt is the average value of all time intervals; Δt1 is the first time interval; Δt j t is the j-th time interval; t1 is the time corresponding to the first mutation point; t j+1 This is the time corresponding to the (j+1)th mutation point; If the time interval between two adjacent abrupt change points is greater than the average value, then the time-drilling parameter-micro-drilling rate matrix within that time range is considered a normal drilling sample; otherwise, the curve is considered an abnormal drilling sample.

[0028] Step 5 specifically includes: Identify the unique identifier number corresponding to the segmentation position based on the mutation point; Based on the unique identifier of the segmentation location, extract the drilling parameters within the time period corresponding to the unique identifier from the time-drilling parameter-micro-drilling rate matrix constructed in step 2. Construct a dataset that combines micro-drilling rate and drilling parameters in the same time dimension under normal drilling conditions.

[0029] The non-drilling samples or abnormal drilling samples constitute a stuck drill risk analysis sample library, which is used for stuck drill risk analysis.

[0030] This embodiment can unify drilling parameters and drilling speed to the same time dimension in real time, and on this basis, establish a more accurate mapping relationship between drilling parameters and drilling speed, eliminating interference items such as outliers. This provides a basis for more accurate optimization of drilling parameters, enabling more precise optimization of drilling parameters at a finer time scale. Abnormal drilling samples and non-drilling samples together constitute the stuck pipe risk analysis sample, which is the basis for subsequent stuck pipe risk analysis, reducing drilling costs and improving the safety of drilling operations.

[0031] Example 3 This embodiment provides a micro-drilling rate classification system based on well logging data, including... A time-series database construction module that collects logging data and builds a drilling parameter time-series database based on the logging data; A micro-drilling rate calculation module is used to calculate the micro-drilling rate, construct a time-drilling parameter-micro-drilling rate matrix, and assign a unique identifier to each set of data in the matrix. The time-drilling parameter-micro-drilling rate matrix is ​​divided into drilling samples and non-drilling samples. Drilling pressure-micro-drilling rate samples are extracted from the drilling samples to form a sample division and curve generation module for drilling pressure-micro-drilling rate curves. Based on image recognition methods, the abrupt change points of micro-drilling rate in the drill pressure-micro-drilling rate analysis sample are identified. The drill pressure-micro-drilling rate sample is then segmented according to the abrupt change points to form a module for identifying abrupt change points and segmenting samples of normal drilling samples and abnormal drilling samples. The system automatically identifies the unique identifier corresponding to the mutation point, obtains the collection time in the corresponding time series database through the time-drilling parameter-micro-drilling rate matrix, and obtains the drilling parameter collection time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters into the same time dimension data unification module.

[0032] Example 4 This embodiment provides a computer device, including a processor, an input device, an output device, and a memory. The processor, input device, output device, and memory are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to call the program instructions to execute the steps described in Embodiment 1 or Embodiment 2 of this invention.

[0033] Example 5 This embodiment provides a computer-readable storage medium storing a computer program for electronic data interchange. The computer program includes program instructions that, when executed by a processor, cause the processor to perform the steps described in Embodiment 1 or Embodiment 2.

Claims

1. A micro-drilling rate classification method based on well logging data, characterized in that, Includes the following steps: Step 1: Construct a time-series database of drilling parameters based on logging data; Step 2: Based on the drilling parameter time series database, calculate the micro-drilling rate corresponding to each group of drilling parameters according to the time interval benchmark, form a drilling rate sample with time as the recording dimension, construct a time-drilling parameter-micro-drilling rate matrix, and number each group of data in the matrix sequentially as a unique identifier; the time interval benchmark refers to the longest time span in the drilling parameter time series database when the drilling pressure is greater than 0 and the drill bit position remains unchanged. Step 3: Divide the time-drilling parameter-micro-rate of drilling matrix into drilling samples and non-drilling samples, extract the bit pressure-micro-rate of drilling samples from the drilling samples, and form the bit pressure-micro-rate of drilling curve. Step 4: Based on image recognition methods, identify the abrupt change points of micro-drilling rate in the drill pressure-micro-drilling rate analysis sample, and segment the drill pressure-micro-drilling rate sample according to the abrupt change points to form normal drilling samples and abnormal drilling samples. Step 5: Automatically identify the unique identifier number corresponding to the segmentation position, and automatically determine the corresponding drilling parameter time series database acquisition time through the time-drilling parameter-micro-drilling rate matrix to obtain the drilling parameter acquisition time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters to the same time dimension.

2. The micro-drilling rate classification method based on logging data according to claim 1, characterized in that: The sampling frequency of the logging data is 1-10Hz.

3. The micro-drilling rate classification method based on logging data according to claim 1, characterized in that: In step 2, for the data group with unique identifier number i, its corresponding time t i Micro-drilling speed V ti The calculation formula is as follows: V ti =(X （ti-T） -X ti ) / T type 1 In Equation 1, V ti For t i Micro-drilling speed at any given moment; X （ti-T） For t i -Drill bit position at time T; X ti For t i The drill bit position at any given time; T is the time interval reference.

4. The micro-drilling rate classification method based on logging data according to claim 1, characterized in that: In step 3, dividing the time-drilling parameter-micro-drilling rate matrix into drilling samples and non-drilling samples means that the time-drilling parameter-micro-drilling rate matrix data that meets the conditions of drilling pressure greater than 0 and micro-drilling rate greater than or equal to 0 are divided into drilling samples, and the rest are divided into non-drilling samples.

5. The micro-drilling rate classification method based on logging data according to claim 1, characterized in that: In step 4, the image recognition method is as follows: Traverse all points on the curve, connect each interval point on the curve to form an interval line, and at the same time draw a perpendicular line through the midpoint to the line connecting the interval points to form a perpendicular line, until all points on the curve have been traversed. The lengths of the perpendiculars are denoted as M1 and M2 respectively. ２ M ３ …、M n The lengths of the lines connecting the interval points are denoted as N1 and N2, respectively. ２ N ３ …、N n ; Calculate the median M of the perpendicular. Ｐ Mean M Ｑ The median N of the interval line Ｐ Mean N Ｑ ; M Ｑ = (M1 + M ２ + M ３ + …… + M n ) / n Equation 2 N Ｑ = (N1 + N ２ + N ３ + …… + N n ) / n Equation 3 In Equations 2 and 3, M Ｑ M is the average length of all perpendicular line segments; n The length of the perpendicular line corresponding to the data group with the unique identifier number n; N n The length of the interval line that uniquely identifies the data group with sequence number n; Calculate the standard deviation M of the vertical lines using the median as a reference. DP The standard deviation N of the interval line DP Calculate the standard deviation M of the vertical lines using the mean as a reference. DQ The standard deviation N of the interval line DQ Take M DP With M DQ The smaller value in is denoted as M. D0 N DP With N DQ The smaller value is denoted as N. D0 ; For a point on the curve with a unique identifier number i, if M i / N i >M D0 / N D0 If so, then the point is determined to be a mutation point, and the time t at that point is recorded. j If the condition is true, then the point is not considered a mutation point; Suppose there are j+1 mutation points, then there are j time intervals. Calculate the average value Δt across all time intervals: △t = (△t1 + ... + △t) j ) / j style 8 Equation 9: △t1=t2-t1 △t j =t j+1 -t j Formula 10 In Equations 8, 9, and 10, Δt is the average value of all time intervals; Δt1 is the first time interval; Δt j t is the j-th time interval; t1 is the time corresponding to the first mutation point; t j+1 This is the time corresponding to the (j+1)th mutation point; If the time interval between two adjacent abrupt change points is greater than the average value, then the time-drilling parameter-micro-drilling rate matrix within that time range is considered a normal drilling sample; otherwise, the curve is considered an abnormal drilling sample.

6. The micro-drilling rate classification method based on logging data according to claim 1, characterized in that: Step 5 specifically includes: Identify the unique identifier number corresponding to the segmentation position based on the mutation point; Based on the unique identifier of the segmentation location, extract the drilling parameters within the time period corresponding to the unique identifier from the time-drilling parameter-micro-drilling rate matrix constructed in step 2. Construct a dataset that combines micro-drilling rate and drilling parameters in the same time dimension under normal drilling conditions.

7. A micro-drilling rate classification method based on logging data according to claim 4 or 5, characterized in that: Non-drilling samples or abnormal drilling samples are compiled into a stuck drill risk analysis sample library for use in stuck drill risk analysis.

8. A micro-drilling rate classification system based on well logging data, characterized in that: This includes a time-series database construction module that collects logging data and builds a drilling parameter time-series database based on the logging data; A micro-drilling rate calculation module is used to calculate the micro-drilling rate, construct a time-drilling parameter-micro-drilling rate matrix, and assign a unique identifier to each set of data in the matrix. The time-drilling parameter-micro-drilling rate matrix is ​​divided into drilling samples and non-drilling samples. Drilling pressure-micro-drilling rate samples are extracted from the drilling samples to form a sample division and curve generation module for drilling pressure-micro-drilling rate curves. Based on image recognition methods, the abrupt change points of micro-drilling rate in the drill pressure-micro-drilling rate analysis sample are identified. The drill pressure-micro-drilling rate sample is then segmented according to the abrupt change points to form a module for identifying abrupt change points and segmenting samples of normal drilling samples and abnormal drilling samples. The system automatically identifies the unique identifier corresponding to the mutation point, obtains the collection time in the corresponding time series database through the time-drilling parameter-micro-drilling rate matrix, and obtains the drilling parameter collection time period corresponding to the normal drilling sample, thereby unifying the micro-drilling rate and drilling parameters into the same time dimension data unification module.

9. A computer device, characterized in that: The system includes a processor, an input device, an output device, and a memory, which are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1-7.

Images