Rock stratum structure regularity while-drilling identification method and rock stratum structure regularity while-drilling identification system

Abstract

The invention discloses a rock stratum structure regularity while-drilling identification method and system. The identification method comprises the steps that while-drilling parameters in the drilling process are obtained; the thrust, the torque, the modulation specific energy and the rock drillability form four-dimensional while-drilling index parameters, and the while-drilling index parameters are input into a support vector machine model optimized through a genetic algorithm; bayes mutation detection is carried out on the four-dimensional while-drilling index parameters, and SLPgt in each index parameter sequence is extracted; the drilling displacement value and the SLP value corresponding to the significant abrupt change point of 0.2 are recorded; respectively calculating sample sequence number difference values of a high SLP value mutation point of a certain index parameter and high SLP value mutation points of other index parameters; selecting a count gt; 3, taking the mutation points as a structure mutation point sequence capable of effectively inverting a rock stratum structure; and calculating a drilling displacement difference value between the adjacent structure sudden change points, and judging the type of the rock stratum structure according to the drilling displacement difference value between the adjacent structure sudden change points and the rock stratum strength mean value.

Description

Technical Field

[0001] The present invention relates to the technical field of rock formation structure identification, and in particular to a strength-interface fusion rock formation structure regularity identification while drilling method and system. Background Art

[0002] As the main type of coal mine tunnel, composite layered roof generally has disorderly distributed structural weak surfaces such as weak interlayers, joints and fissures, and rock interfaces. These discontinuous structures significantly weaken the adhesion and friction between rock layers, making the roof prone to delamination and sliding or even roof collapse under stress disturbance. Existing technologies mainly use traditional methods such as drilling and core sampling to detect rock structure, which has the disadvantages of limited detection range, low efficiency and strong subjectivity. While drilling (MWD) technology inverts geological characteristics by collecting parameters such as drill rig thrust, torque, and speed in real time. It has the advantages of synchronous construction, continuous data, and low cost, and provides a new way for intelligent identification of rock structure.

[0003] However, existing methods for identifying rock formation structure while drilling focus on lithologic strength prediction, but are weak in identifying rock formation structure. Furthermore, existing methods often use a single indicator (such as torque mutation) for structural identification, which is susceptible to interference from drill vibration and sensor noise, resulting in a high rate of structural misjudgment. Furthermore, existing technologies fail to utilize methods and techniques such as multi-indicator parameter fusion, mutation point collaborative verification, and strength-structure coupling. As a result, existing rock formation structure detection methods are insensitive to subtle structural changes and cannot effectively distinguish between gradual lithologic changes and structural mutations. Furthermore, they lack a multi-source parameter cross-validation mechanism, resulting in a disconnect between structural identification and strength prediction. These factors severely restrict the application of while drilling detection technology in the field of intelligent and precise rock formation structure identification. Summary of the Invention

[0004] In response to the problems and needs raised above, this solution proposes a method and system for identifying rock structure regularity while drilling. By adopting the following technical features, it can achieve the above technical objectives and bring about many other technical effects.

[0005] An object of the present invention is to provide a method for identifying rock formation structural regularity while drilling, comprising the following steps:

[0006] S10: Acquire drilling parameters during the drilling process, specifically including thrust, torque, rotation speed, displacement, drilling speed, and vibration; and use the drilling parameters to solve two comprehensive indicators, modulation specific energy and rock drillability;

[0007] S20: Thrust, torque, modulation specific energy, and rock drillability are combined into four-dimensional while-drilling index parameters, which are then input into a support vector machine model optimized by a genetic algorithm. The model is trained with historical drilling data to establish a rock formation strength prediction, and outputs the rock formation strength in the tunnel roof in real time during the current drilling process.

[0008] S30: Perform Bayes mutation detection on the four-dimensional while-drilling index parameters respectively to obtain the drilling displacement and SLP corresponding to the mutation points in the data of each index parameter, where SLP is the mutation degree index, and extract the significant mutation points in each index parameter sequence that meet |SLP|>0.2, and record the corresponding drilling displacement value and SLP value;

[0009] S40: Calculate the sample sequence difference between the mutation point with high SLP value of a certain parameter and the mutation point with high SLP value of other parameters respectively. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0. The mutation point with count > 3 is selected as the structural mutation point sequence that can effectively invert the rock layer structure.

[0010] S50: Arrange the structural mutation points in ascending order according to the sample sequence number values, calculate the drilling displacement difference between adjacent structural mutation points, and determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0011] In addition, the method for identifying rock formation structural regularity while drilling according to the present invention may also have the following technical features:

[0012] In one example of the present invention, in step S10, the calculation formulas for solving the two comprehensive indicators SEM and RDA using the drilling parameters are as follows:

[0013]

[0014] Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

[0015] In an example of the present invention, step S30 specifically includes the following steps:

[0016] S31: Input each indicator parameter into the Bayes mutation detection algorithm to obtain the mutation point sequence of different indicator parameters and the SLP value corresponding to each mutation point;

[0017]

[0018] Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point;

[0019] S32: Setting the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening the mutation points whose |SLP| in each indicator parameter value is greater than the threshold, thereby forming a high SLP value mutation point sequence;

[0020]

[0021] Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.2.

[0022] In an example of the present invention, step S40 specifically includes the following steps:

[0023] S41: Calculate the sample sequence number difference between a certain parameter high SLP value mutation point and other parameter high SLP value mutation points respectively. When the sample sequence number difference of a certain high SLP value mutation point is less than 200, the count is increased by 1, otherwise the count is increased by 0.

[0024]

[0025] In the formula, Sum represents the cumulative sum, and Count is the counting function;

[0026] S42: Select mutation points with a count greater than 3 as characteristic mutation points that can effectively invert the rock layer structure, and arrange the characteristic mutation points in ascending order according to the sample sequence value;

[0027] C I =(c I1 ,c I2 ,…c Ii ,…)Sum(c Ii )≥3&c Ii <c Ii+1

[0028] Where c Ii Characteristic mutation points are arranged in ascending order;

[0029] S43: Calculate the difference between the sample numbers of each mutation point in the characteristic mutation point sequence. When the difference between the sample numbers of two characteristic mutation points is less than 200, merge the two characteristic mutation points, and use the average of their sample numbers as the sample number of the merged mutation point, and the average of their SLPs as the SLP value of the merged mutation point, thereby obtaining a non-overlapping structural mutation point sequence related to the rock layer structure.

[0030] In one example of the present invention, in step S43, the expression of the sequence of structural mutation points related to the rock formation structure without overlap is:

[0031]

[0032] Where c Ik is the mean of the sample numbers of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the sample number sequence of the structural mutation point; s Ik It is the SLP mean of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the structural mutation point SLP value.

[0033] In an example of the present invention, step S50 specifically includes the following steps:

[0034] S51: Arrange the structural mutation points in ascending order according to the sample sequence numbers, and determine the drilling displacement corresponding to the structural mutation points according to the sample sequence numbers;

[0035]

[0036] In the formula, the sample number of each structural mutation point is c Ik The drilling displacement corresponding to each structural mutation point is expressed as d Ii ;

[0037] S51: Calculate the drilling displacement difference between adjacent structural mutation points;

[0038] l j =d i -d i-1

[0039] Where, l j is the distance between adjacent structural mutation points;

[0040] S52: Calculate the mean value of the rock formation strength predicted by the PSO-BP model between adjacent structural mutation points;

[0041]

[0042] Where R c is the structural mutation point c Ik and c IkThe rock strength sequence between p is the mean value of the rock formation strength sequence;

[0043] S53: Determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0044] In one example of the present invention, in step S53, determining the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength specifically includes the following steps:

[0045] ① Crack development: When 0mm <l j ≤50mm and [d i ,d i-1 ]In the range 0MPa≤R p When it is less than 5MPa, it means that the rock structure within the range of the two change points is fractured;

[0046]

[0047] ② Rock joints: when 50mm≤l j <100mm and [d i ,d i-1 ]In the range 0MPa≤R p When <5MPa, it means that the rock structure within the range of two change points is jointed;

[0048]

[0049] ③ Weak interlayer: when 100mm≤l j <1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak interlayer;

[0050]

[0051] ④ Weak rock layer: When l j ≥1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak rock layer;

[0052]

[0053] ⑤Hard rock layer: when l j ≥1000mm and [d i ,d i-1 ]R in the interval pWhen it is >20MPa, it means that the rock structure within the range of the two change points is hard rock;

[0054]

[0055] Another object of the present invention is to provide a system for identifying rock formation structural regularity while drilling, comprising:

[0056] The parameter acquisition and processing module is configured to acquire the drilling parameters during the drilling process, specifically including thrust, torque, rotational speed, displacement, drilling speed and vibration; and use the drilling parameters to solve the two comprehensive indicators of modulation specific energy and rock drillability;

[0057] A strength prediction module is configured to combine thrust, torque, modulation specific energy, and rock drillability into four-dimensional while-drilling index parameters, and input these while-drilling index parameters into a support vector machine model optimized by a genetic algorithm. The model is trained using historical drilling data to establish a rock formation strength prediction and outputs the rock formation strength in the tunnel roof in real time during the current drilling process;

[0058] A mutation point screening module is configured to perform Bayes mutation detection on the four-dimensional while-drilling indicator parameters, obtain the drilling displacement and SLP corresponding to the mutation points in each indicator parameter data, where SLP is the mutation degree index, and extract the significant mutation points in each indicator parameter sequence that meet |SLP|>0.2, and record the corresponding drilling displacement value and SLP value;

[0059] A mutation point calculation module is configured to calculate the sample sequence difference between the mutation point with a high SLP value of a certain parameter and the mutation point with a high SLP value of other parameters. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0; mutation points with a count greater than 3 are selected as the structural mutation point sequence that can effectively invert the rock structure;

[0060] The rock formation structure judgment module is configured to arrange the structural mutation points in ascending order according to the sample sequence number value, calculate the drilling displacement difference between adjacent structural mutation points, and judge the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0061] In one example of the present invention, the calculation formula for solving the two comprehensive indicators of SEM and RDA using the drilling parameters is as follows:

[0062]

[0063] Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

[0064] In one example of the present invention, the mutation point screening module includes:

[0065] a mutation point detection unit configured to input each indicator parameter into a Bayes mutation detection algorithm to obtain a mutation point sequence of different indicator parameters and an SLP value corresponding to each mutation point;

[0066]

[0067] Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point;

[0068] A mutation point sequence unit is configured to set the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening mutation points with |SLP| greater than the threshold in each indicator parameter value, thereby forming a high SLP value mutation point sequence;

[0069]

[0070] Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.2.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] Traditional methods mostly rely on a single downhole parameter (such as torque or drilling speed) for mutation detection, which is easily affected by noise and leads to misjudgment. The present invention innovatively uses the Bayes mutation detection algorithm to perform mutation detection on the four-dimensional indicator parameters of thrust, torque, SEM and RDA, and forms a structural mutation point sequence by counting and screening the difference in sample numbers across the indicator parameter mutation points (count>3). This multi-indicator parameter cross-validation mechanism effectively eliminates pseudo-mutations caused by abnormal fluctuations of a single indicator parameter by setting dual constraints of the mutation point SLP threshold (|SLP|>0.2) and the sample number difference (<200), solving the problem of traditional methods being insensitive to small rock layer structural responses. It can improve the spatial resolution of rock interface identification to the millimeter level, which is particularly suitable for the precise division of thin-layered or fractured rock masses.

[0073] In addition, the existing technology usually separates strength prediction and structure identification. The present invention integrates the rock formation strength prediction model and the Bayes mutation detection model, and uses the strength mean (R p ) and displacement difference (l j ) jointly constructs classification rules. This dual criterion of rock properties and geometric features breaks through the limitations of traditional rock structure classification, which relies primarily on displacement thresholds. This not only enables the technology in this invention to identify more rock types, but also effectively improves the recognition accuracy of rock structure. At the same time, to address the problem of overlap in detected mutation points, this invention proposes a dynamic merging strategy for characteristic mutation points. When the difference in sample numbers of adjacent mutation points is less than 200, the sample number mean and the SLP mean are merged to generate a sequence of structural mutation points without overlap.

[0074] Hereinafter, the best embodiment of the present invention will be described in more detail with reference to the accompanying drawings so that the features and advantages of the present invention can be easily understood. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings of the embodiments of the present invention. The drawings are only used to illustrate some embodiments of the present invention, but not to limit all embodiments of the present invention thereto.

[0076] Figure 1 Flowchart of a method for identifying rock formation structure regularity while drilling according to an embodiment of the present invention;

[0077] Figure 2 A schematic diagram of the structure of drilling and measurement equipment used in the method for identifying rock formation structure regularity while drilling according to an embodiment of the present invention;

[0078] Figure 3 A torque monitoring curve diagram of a roof rock drilling process according to an embodiment of the present invention;

[0079] Figure 4 2. This is a diagram showing the effect of detecting a sudden change in roof rock structure according to an embodiment of the present invention;

[0080] Figure 5 This is a peek at a drilling hole in the roof rock layer of a tunnel according to an embodiment of the present invention.

[0081] Reference numerals:

[0082] Single anchor drilling rig 1;

[0083] Drilling parameter measuring instrument 2;

[0084] Signal wireless transmission module 3;

[0085] Signal wireless receiving module 4;

[0086] Multi-channel data acquisition module 5;

[0087] Data storage and display terminal 6;

[0088] B19 hexagonal drill rod 7;

[0089] Rock layer 8. DETAILED DESCRIPTION

[0090] In order to make the purpose, technical solution and advantages of the technical solution of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of specific embodiments of the present invention. The same figure marks in the drawings represent the same parts. It should be noted that the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning understood by persons of ordinary skill in the field to which the invention belongs. The words "first", "second" and similar terms used in the patent application specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "a" or "an" do not necessarily indicate a quantity limitation. Words such as "include" or "comprising" mean that the elements or objects preceding the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0092] According to the first aspect of the present invention, a method for identifying rock formation structure regularity while drilling is proposed, such as Figure 1 As shown, the following steps are included:

[0093] S10: Acquire drilling parameters during the drilling process, specifically including thrust, torque, rotation speed, displacement, drilling speed, and vibration; and use the drilling parameters to solve two comprehensive indicators: specific energy of modulation (SEM) and rock drillability (RDA);

[0094] S20: Thrust, torque, SEM, and RDA are combined into four-dimensional while-drilling index parameters, which are then input into a support vector machine model optimized by a genetic algorithm (GA-SVM). This model is trained with historical drilling data to establish a rock formation strength prediction and outputs the rock formation strength in the tunnel roof in real time during the current drilling process.

[0095] S30: Perform Bayes mutation detection on the four-dimensional while-drilling index parameters respectively, obtain the drilling displacement and mutation level index (SLP) corresponding to the mutation points in the data of each index parameter, and extract the significant mutation points (called high SLP value mutation points) that meet |SLP|>0.2 in each index parameter sequence, and record the corresponding drilling displacement value and SLP value;

[0096] S40: Calculate the sample sequence difference between the mutation point with high SLP value of a certain parameter and the mutation point with high SLP value of other parameters respectively. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0. The mutation point with count > 3 is selected as the structural mutation point sequence that can effectively invert the rock layer structure.

[0097] S50: Arrange the structural mutation points in ascending order according to the sample sequence number values, calculate the drilling displacement difference between adjacent structural mutation points, and determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0098] Traditional methods often rely on a single while-drilling parameter (such as torque or drilling speed) for mutation detection, which is susceptible to noise interference and can lead to misjudgments. This identification method innovatively utilizes a Bayesian mutation detection algorithm to detect mutations in four-dimensional parameters: thrust, torque, SEM, and RDA. A sequence of structural mutation points is generated by counting and screening the difference in sample numbers across these parameter mutation points (counts > 3). This multi-parameter cross-validation mechanism, through the dual constraints of a mutation point SLP threshold (|SLP| > 0.2) and sample number difference (< 200), effectively eliminates spurious mutations caused by abnormal fluctuations in a single parameter. This addresses the traditional method's insensitivity to small rock layer structures and improves the spatial resolution of lithologic interface identification to the millimeter level, making it particularly suitable for the precise delineation of thin-layered or fractured rock masses.

[0099] In addition, the existing technology usually separates strength prediction and structure identification. This identification method integrates the rock formation strength prediction model and the Bayes mutation detection model, and calculates the strength mean (R p ) and displacement difference (l j ) jointly constructs classification rules. This dual criterion of rock properties and geometric features breaks through the limitations of traditional rock structure classification, which relies primarily on displacement thresholds. This not only enables the technology in this invention to identify more rock types, but also effectively improves the recognition accuracy of rock structure. At the same time, to address the problem of overlap in detected mutation points, this invention proposes a dynamic merging strategy for characteristic mutation points. When the difference in sample numbers of adjacent mutation points is less than 200, the sample number mean and the SLP mean are merged to generate a sequence of structural mutation points without overlap.

[0100] It is understood that the drilling equipment used in this method is Figure 3 As shown, it includes a single anchor drilling rig 1, a drilling parameter measuring instrument 2, a signal wireless transmission module 3, a signal wireless receiving module 4, a multi-channel data acquisition module 5, a data storage and display terminal 6 and a B19 hexagonal drill rod 7; wherein, the B19 hexagonal drill rod 7 is drilled into the rock formation 8, and the torque and displacement sensors are installed inside the drilling parameter measuring instrument 2. The lower end is connected to the single anchor drilling rig 1 using an external hexagonal shaft, and the upper end is connected to the B19 hexagonal drill rod 7.

[0101] The drilling rig operating parameters collected by the sensor first enter the signal wireless transmission module 3, and then are received by the signal wireless receiving module 4 and transmitted to the multi-channel data acquisition module 5. The multi-channel data acquisition module 5 is connected to the data storage and display terminal 6 for storage and display.

[0102] In one example of the present invention, in step S10, the calculation formulas for solving the two comprehensive indicators SEM and RDA using the drilling parameters are as follows:

[0103]

[0104] Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

[0105] In an example of the present invention, step S30 specifically includes the following steps:

[0106] S31: Input each indicator parameter into the Bayes mutation detection algorithm to obtain the mutation point sequence of different indicator parameters and the SLP value corresponding to each mutation point;

[0107]

[0108] Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point;

[0109] S32: Setting the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening the mutation points whose |SLP| in each indicator parameter value is greater than the threshold, thereby forming a high SLP value mutation point sequence;

[0110]

[0111] Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.2.

[0112] In an example of the present invention, step S40 specifically includes the following steps:

[0113] S41: Calculate the sample sequence number difference between a certain parameter high SLP value mutation point and other parameter high SLP value mutation points respectively. When the sample sequence number difference of a certain high SLP value mutation point is less than 200, the count is increased by 1, otherwise the count is increased by 0.

[0114]

[0115] In the formula, Sum represents the cumulative sum, and Count is the counting function;

[0116] S42: Select mutation points with a count greater than 3 as characteristic mutation points that can effectively invert the rock layer structure, and arrange the characteristic mutation points in ascending order according to the sample sequence value;

[0117] C I =(c I1 ,c I2 ,…c Ii ,…)Sum(c Ii )≥3&c Ii <c Ii+1

[0118] Where c Ii Characteristic mutation points are arranged in ascending order;

[0119] S43: Calculate the difference between the sample numbers of each mutation point in the characteristic mutation point sequence. When the difference between the sample numbers of two characteristic mutation points is less than 200, merge the two characteristic mutation points, and use the average of their sample numbers as the sample number of the merged mutation point, and the average of their SLPs as the SLP value of the merged mutation point, thereby obtaining a non-overlapping structural mutation point sequence related to the rock layer structure.

[0120] In one example of the present invention, in step S43, the expression of the sequence of structural mutation points related to the rock formation structure without overlap is:

[0121]

[0122] Where c Ik is the mean of the sample numbers of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the sample number sequence of the structural mutation point; s Ik It is the SLP mean of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the structural mutation point SLP value.

[0123] In an example of the present invention, step S50 specifically includes the following steps:

[0124] S51: Arrange the structural mutation points in ascending order according to the sample sequence numbers, and determine the drilling displacement corresponding to the structural mutation points according to the sample sequence numbers;

[0125]

[0126] In the formula, the sample number of each structural mutation point is c Ik The drilling displacement corresponding to each structural mutation point is expressed as d Ii ;

[0127] S51: Calculate the drilling displacement difference between adjacent structural mutation points;

[0128] l j =d i -d i-1

[0129] Where, l j is the distance between adjacent structural mutation points;

[0130] S52: Calculate the mean value of the rock formation strength predicted by the PSO-BP model between adjacent structural mutation points;

[0131]

[0132] Where R c is the structural mutation point c Ik and c Ik The rock strength sequence between p is the mean value of the rock formation strength sequence;

[0133] S53: Determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0134] In one example of the present invention, in step S53, determining the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength specifically includes the following steps:

[0135] ① Crack development: When 0mm <l j ≤50mm and [d i ,d i-1 ]In the range 0MPa≤R p When it is less than 5MPa, it means that the rock structure within the range of the two change points is fractured;

[0136]

[0137] ② Rock joints: when 50mm≤l j <100mm and [d i ,d i-1 ]In the range 0MPa≤R p When <5MPa, it means that the rock structure within the range of two change points is jointed;

[0138]

[0139] ③ Weak interlayer: when 100mm≤l j <1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak interlayer;

[0140]

[0141] ④ Weak rock layer: When l j ≥1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak rock layer;

[0142]

[0143] ⑤Hard rock layer: when l j≥1000mm and [d i ,d i-1 ]R in the interval p When it is >20MPa, it means that the rock structure within the range of the two change points is hard rock;

[0144]

[0145] According to a second aspect of the present invention, a system for identifying rock formation structural regularity while drilling comprises:

[0146] The parameter acquisition and processing module is configured to acquire the drilling parameters during the drilling process, specifically including thrust, torque, rotational speed, displacement, drilling speed, and vibration; and use the drilling parameters to solve the two comprehensive indicators of modulation specific energy (SEM) and rock drillability (RDA);

[0147] A strength prediction module is configured to combine thrust, torque, SEM, and RDA into four-dimensional while-drilling parameters, and input these parameters into a support vector machine model optimized by a genetic algorithm (GA-SVM). The model is trained using historical drilling data to establish a rock formation strength prediction and outputs the rock formation strength in the tunnel roof in real time during the current drilling process;

[0148] The mutation point screening module is configured to perform Bayes mutation detection on the four-dimensional while-drilling index parameters, obtain the drilling displacement and mutation level index (SLP) corresponding to the mutation points in each index parameter data, extract the significant mutation points (called high SLP value mutation points) that meet |SLP|>0.2 in each index parameter sequence, and record the corresponding drilling displacement value and SLP value;

[0149] A mutation point calculation module is configured to calculate the sample sequence difference between the mutation point with a high SLP value of a certain parameter and the mutation point with a high SLP value of other parameters. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0; mutation points with a count greater than 3 are selected as the structural mutation point sequence that can effectively invert the rock structure;

[0150] The rock formation structure judgment module is configured to arrange the structural mutation points in ascending order according to the sample sequence number value, calculate the drilling displacement difference between adjacent structural mutation points, and judge the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

[0151] Traditional methods often rely on a single while-drilling parameter (such as torque or drilling speed) for mutation detection, which is susceptible to noise interference and misjudgment. This identification system innovatively utilizes a Bayesian mutation detection algorithm to detect mutations in four-dimensional parameters: thrust, torque, SEM, and RDA. It then forms a sequence of structural mutation points by counting and screening the difference in sample numbers across these parameter mutation points (counts > 3). This multi-parameter cross-validation mechanism, by setting dual constraints of the mutation point SLP threshold (|SLP| > 0.2) and the sample number difference (< 200), effectively eliminates spurious mutations caused by abnormal fluctuations in a single parameter. This addresses the traditional method's insensitivity to small rock layer structures and improves the spatial resolution of lithologic interface identification to the millimeter level, making it particularly suitable for the precise delineation of thin-layered or fractured rock masses.

[0152] In addition, the existing technology usually separates strength prediction and structure recognition. This identification system integrates the rock strength prediction model and the Bayes mutation detection model, and calculates the strength mean (R p ) and displacement difference (l j ) jointly constructs classification rules. This dual criterion of rock properties and geometric features breaks through the limitations of traditional rock structure classification, which relies primarily on displacement thresholds. This not only enables the technology in this invention to identify more rock types, but also effectively improves the recognition accuracy of rock structure. At the same time, to address the problem of overlap in detected mutation points, this invention proposes a dynamic merging strategy for characteristic mutation points. When the difference in sample numbers of adjacent mutation points is less than 200, the sample number mean and the SLP mean are merged to generate a sequence of structural mutation points without overlap.

[0153] In one example of the present invention, the calculation formula for solving the two comprehensive indicators of SEM and RDA using the drilling parameters is as follows:

[0154]

[0155] Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

[0156] In one example of the present invention, the mutation point screening module includes:

[0157] a mutation point detection unit configured to input each indicator parameter into a Bayes mutation detection algorithm to obtain a mutation point sequence of different indicator parameters and an SLP value corresponding to each mutation point;

[0158]

[0159] Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point;

[0160] A mutation point sequence unit is configured to set the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening mutation points with |SLP| greater than the threshold in each indicator parameter value, thereby forming a high SLP value mutation point sequence;

[0161]

[0162] Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.2.

[0163] It should be noted that the system for identifying rock formation structure regularity while drilling of the present invention can also perform any processing in the method for identifying rock formation structure regularity while drilling described previously, and the specific details are not repeated here.

[0164] Specific cases

[0165] Identification of rock strata and rock sequence in tunnel roof while drilling

[0166] Figure 3 and Figure 4 The following is a torque curve for the on-site tunnel roof drilling process. The figure shows that the overall rotational torque range for drilling depths of 0 to 1.5 m is between 4.5 and 10.0 N·m, with an average value of 7.2 N·m. The rotational torque range for drilling depths of 1.5 to 5.2 m is concentrated between 10.0 and 16.0 N·m, with an average value of approximately 14.6 N·m. The rotational torque range for drilling depths of 5.2 to 7.0 m is between 14.0 and 22.0 N·m, with an average value of approximately 18.2 N·m.

[0167] The data during the drilling stop period are eliminated, and only the torque data during the rock formation drilling period are retained. The torque data are substituted into the GA-SVM model for prediction to obtain the roof rock formation strength at the drilling location. Then, the Bayes mutation detection algorithm is used to detect the mutation points in the rock formation strength data sequence to obtain Figure 3 and Figure 4 The test results are shown in the figure. Based on on-site statistics, the average drilling speed for each borehole was approximately 15 mm / s, and the data sampling frequency was 100 samples / s. Therefore, the drill depth at a specific moment can be calculated based on the sample number. The following table shows the rock structure identification process and process parameters.

[0168] Rock structure mutation detection process and parameters

[0169]

[0170]

[0171] During the rock formation structure identification process, a Bayesian algorithm was first used to detect 12 mutation points with SLP absolute values ​​greater than 0.2. The sample number differences between adjacent mutation points were then calculated. The sample number difference between mutation points 16851 and 17076 was found to be 225, and the sample number difference between mutation points 42218 and 42260 was 42. Therefore, mutation points with sample number differences less than 300 were merged into 16964 and 42239, with SLP values ​​of 0.9 and -5.4. The sample number differences between adjacent mutation points after merging were calculated, and the displacement spacing between adjacent mutation points was determined based on the drilling speed and sampling frequency. The mean of the rock formation strength series between adjacent mutation points was then calculated. Finally, the rock formation structure was determined based on the displacement spacing and mean strength between the mutation points.

[0172] Figure 5 The rock formation information shown in the drilling identification results are compared with the drilling identification results. Figure 5 The comparison of the ZK2 borehole peek results is shown below. The rock structure identification characteristics of the two are compared as follows:

[0173] ① Rock stratification characteristics.

[0174] The results of the rock structure identification while drilling (LWD) show that the roof depth range of 0.0 to 1.3 meters (corresponding to sample numbers 0 to 8430) is sandy mudstone, the range of 1.3 to 5.1 meters (corresponding to sample numbers 8430 to 33700) is mainly coarse-grained sandstone, and the range of 5.1 to 7.0 meters (corresponding to sample numbers 33700 to 47100) is fine-grained sandstone. Borehole observation results show that the rock lithologies within the roof borehole range can be divided into three categories, and the rock interface is highly consistent with the results obtained by the LWD rock structure identification method.

[0175] ② Characteristics of interlayer and fissure development.

[0176] The results of while-drilling rock structure identification revealed that the roof range of 2.56-2.89 m contained a hard interlayer with a thickness of approximately 0.33 m. Weak interlayers were found in the ranges of 4.1-5.0 m and 6.3-6.7 m, respectively, with thicknesses of 0.9 m and 0.4 m. Furthermore, a fissure developed at 6.75 m in the roof, with an opening of 0.12 m. Borehole observation results showed that the differences in rock structure detection results between while-drilling detection and borehole observation were minimal, further demonstrating the effectiveness of the proposed method for while-drilling rock structure identification.

[0177] The exemplary implementation schemes of the method and system for identifying rock structure regularity while drilling proposed in the present invention are described in detail above with reference to preferred embodiments. However, it will be understood by those skilled in the art that, without departing from the concept of the present invention, various modifications and variations can be made to the above-mentioned specific embodiments, and various technical features and structures proposed in the present invention can be combined in various ways without exceeding the scope of protection of the present invention, which is determined by the appended claims.

Claims

1. A method for identifying rock formation structural regularity while drilling, characterized in that: The steps include: S10: Acquire drilling parameters during the drilling process, specifically including thrust, torque, rotation speed, displacement, drilling speed, and vibration; and use the drilling parameters to solve two comprehensive indicators, modulation specific energy and rock drillability; S20: Thrust, torque, modulation specific energy, and rock drillability are combined into four-dimensional while-drilling index parameters, which are then input into a support vector machine model optimized by a genetic algorithm. The model is trained with historical drilling data to establish a rock formation strength prediction, and outputs the rock formation strength in the tunnel roof in real time during the current drilling process. S30: Perform Bayes mutation detection on the four-dimensional while-drilling index parameters respectively to obtain the drilling displacement and SLP corresponding to the mutation points in the data of each index parameter, where SLP is the mutation degree index, and extract the significant mutation points in each index parameter sequence that meet |SLP|>0.2, and record the corresponding drilling displacement value and SLP value; S40: Calculate the sample sequence difference between the mutation point with high SLP value of a certain parameter and the mutation point with high SLP value of other parameters respectively. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0. The mutation point with count > 3 is selected as the structural mutation point sequence that can effectively invert the rock layer structure. S50: Arrange the structural mutation points in ascending order according to the sample sequence number values, calculate the drilling displacement difference between adjacent structural mutation points, and determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

2. The method for identifying rock formation structural regularity while drilling according to claim 1, characterized in that: In step S10, the calculation formulas for the two comprehensive indicators SEM and RDA are solved using the drilling parameters as follows: Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

3. The method for identifying rock formation structural regularity while drilling according to claim 1, characterized in that: The step S30 specifically includes the following steps: S31: Input each indicator parameter into the Bayes mutation detection algorithm to obtain the mutation point sequence of different indicator parameters and the SLP value corresponding to each mutation point; Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point; S32: Setting the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening the mutation points whose |SLP| in each indicator parameter value is greater than the threshold, thereby forming a high SLP value mutation point sequence; Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.

2.

4. The method for identifying rock formation structural regularity while drilling according to claim 1, characterized in that: The step S40 specifically includes the following steps: S41: Calculate the sample sequence number difference between a certain parameter high SLP value mutation point and other parameter high SLP value mutation points respectively. When the sample sequence number difference of a certain high SLP value mutation point is less than 200, the count is increased by 1, otherwise the count is increased by 0. In the formula, Sum represents the cumulative sum, and Count is the counting function; S42: Select mutation points with a count greater than 3 as characteristic mutation points that can effectively invert the rock layer structure, and arrange the characteristic mutation points in ascending order according to the sample sequence value; C I =(c I1 ,c I2 ,…c Ii ,…)Sum(c Ii )≥3&c Ii <c Ii+1 Where c Ii Characteristic mutation points are arranged in ascending order; S43: Calculate the difference between the sample numbers of each mutation point in the characteristic mutation point sequence. When the difference between the sample numbers of two characteristic mutation points is less than 200, merge the two characteristic mutation points, and use the average of their sample numbers as the sample number of the merged mutation point, and the average of their SLPs as the SLP value of the merged mutation point, thereby obtaining a non-overlapping structural mutation point sequence related to the rock layer structure.

5. The method for identifying rock formation structural regularity while drilling according to claim 4, characterized in that: In step S43, the expression of the sequence of structural mutation points related to the rock formation structure without overlap is: Where c Ik is the mean of the sample numbers of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the sample number sequence of the structural mutation point; s Ik It is the SLP mean of the i-th characteristic mutation point and the j-th characteristic mutation point detected by thrust, also known as the structural mutation point SLP value.

6. The method for identifying rock formation structural regularity while drilling according to claim 1, characterized in that: The step S50 specifically includes the following steps: S51: Arrange the structural mutation points in ascending order according to the sample sequence numbers, and determine the drilling displacement corresponding to the structural mutation points according to the sample sequence numbers; In the formula, the sample number of each structural mutation point is c Ik The drilling displacement corresponding to each structural mutation point is expressed as d Ii ; S52: Calculate the drilling displacement difference between adjacent structural mutation points; l j =d i -d i-1 Where, l j is the distance between adjacent structural mutation points; S52: Calculate the mean value of the rock formation strength predicted by the PSO-BP model between adjacent structural mutation points; Where R c is the structural mutation point c Ik and c Ik The rock strength sequence between p is the mean value of the rock formation strength sequence; S53: Determine the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

7. The method for identifying rock formation structural regularity while drilling according to claim 6, characterized in that: In step S53, judging the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength specifically includes the following steps: ① Crack development: When 0mm <l j ≤50mm and [d i ,d i-1 ]In the range 0MPa≤R p When it is less than 5MPa, it means that the rock structure within the range of the two change points is fractured; ② Rock joints: when 50mm≤l j <100mm and [d i ,d i-1 ]In the range 0MPa≤R p When <5MPa, it means that the rock structure within the range of two change points is jointed; ③ Weak interlayer: when 100mm≤l j <1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak interlayer; ④ Weak rock layer: When l j ≥1000mm and [d i ,d i-1 ]5MPa within the range <R p When ≤20MPa, it means that the rock structure within the range of the two change points is a weak rock layer; ⑤Hard rock layer: when l j ≥1000mm and [d i ,d i-1 ]R in the interval p When it is >20MPa, it means that the rock structure within the range of the two change points is hard rock; 8. A system for identifying rock formation structural regularity while drilling, characterized in that: include: The parameter acquisition and processing module is configured to acquire the drilling parameters during the drilling process, specifically including thrust, torque, rotational speed, displacement, drilling speed and vibration; and use the drilling parameters to solve the two comprehensive indicators of modulation specific energy and rock drillability; A strength prediction module is configured to combine thrust, torque, modulation specific energy, and rock drillability into four-dimensional while-drilling index parameters, and input these while-drilling index parameters into a support vector machine model optimized by a genetic algorithm. The model is trained using historical drilling data to establish a rock formation strength prediction and outputs the rock formation strength in the tunnel roof in real time during the current drilling process; A mutation point screening module is configured to perform Bayes mutation detection on the four-dimensional while-drilling indicator parameters, obtain the drilling displacement and SLP corresponding to the mutation points in each indicator parameter data, where SLP is the mutation degree index, and extract the significant mutation points in each indicator parameter sequence that meet |SLP|>0.2, and record the corresponding drilling displacement value and SLP value; A mutation point calculation module is configured to calculate the sample sequence difference between the mutation point with a high SLP value of a certain parameter and the mutation point with a high SLP value of other parameters. When the difference is less than 200, the count is increased by 1, otherwise the count is increased by 0; mutation points with a count greater than 3 are selected as the structural mutation point sequence that can effectively invert the rock structure; The rock formation structure judgment module is configured to arrange the structural mutation points in ascending order according to the sample sequence number value, calculate the drilling displacement difference between adjacent structural mutation points, and judge the rock formation structure type based on the drilling displacement difference between adjacent structural mutation points and the average rock formation strength.

9. The rock formation structure regularity identification while drilling system according to claim 8, characterized in that: The calculation formulas for the two comprehensive indicators SEM and RDA using the drilling parameters are as follows: Where w is used to enhance the weak signal in the drilling specific energy sequence; c specifies the switching point of the logistic function; a controls the transition interval of the SEM curve when the lithology changes; k a and C are constants respectively; λ, α and β are important parameters that control the functional relationship between rock strength and drilling parameters.

10. The rock formation structure regularity identification while drilling system according to claim 8, characterized in that: The mutation point screening module includes: a mutation point detection unit configured to input each indicator parameter into a Bayes mutation detection algorithm to obtain a mutation point sequence of different indicator parameters and an SLP value corresponding to each mutation point; Where c Fi is the sample number of the i-th mutation point detected by thrust, s Fi is the SLP value corresponding to the i-th mutation point of thrust, c Mi is the sample number of the i-th mutation point detected by torque, s Mi is the SLP value corresponding to the i-th torque mutation point; A mutation point sequence unit is configured to set the absolute value of the SLP threshold of each indicator parameter to 0.2, thereby screening mutation points with |SLP| greater than the threshold in each indicator parameter value, thereby forming a high SLP value mutation point sequence; Where c HFi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the thrust mutation points; s HFi is the SLP value corresponding to the i-th mutation point where the SLP absolute value is greater than 0.2 among the thrust mutation points; c HMi is the sample number of the i-th mutation point whose SLP absolute value is greater than 0.2 among the torque mutation points; s HMi is the SLP value corresponding to the i-th mutation point among the torque mutation points where the SLP absolute value is greater than 0.2.

Images