Stratum analysis method and system based on TBM tunneling data

Abstract

The invention discloses a stratum analysis method and system based on TBM tunneling data. The method comprises the following steps: acquiring operation data of a TBM; segmenting according to a preset rotating speed so as to divide the rotating speed range of the cutterhead into a plurality of rotating speed intervals, and distributing a label for each rotating speed interval; segmenting the data in each interval according to the cutterhead torque, and clustering the propulsion speed in each segment to obtain a clustering calculation result; calculating the category of each batch of new operation data in the segmented clustering to obtain the distribution condition of the batch of data in the category so as to obtain clustering proportion distribution data; calculating a surrounding rock grade probability according to the clustering proportion distribution data by using normalized Gaussian distribution, and setting a surrounding rock grade with a continuous category representation; and based on the surrounding rock grade and the corresponding surrounding rock grade probability, performing weighted average calculation to obtain a grade fuzzy value of the stratum. According to the method, through segmented clustering and classified calculation of the cutterhead torque and the propulsion speed, the grade probability of the stratum surrounding rock is accurately judged.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geological exploration and tunnel construction, and particularly relates to a stratum analysis method and system based on TBM tunneling data. BACKGROUND

[0002] In geological exploration and tunnel construction, accurately determining the grade of stratum surrounding rock has an important influence on construction safety, efficiency and cost. Traditional stratum analysis methods often rely on manual experience and geological exploration data, and have the problems of strong subjectivity and insufficient accuracy. With the development of data science and machine learning technology, it is possible to determine the grade of stratum surrounding rock using data analysis methods. However, existing data analysis methods often lack fine processing of stratum data, resulting in inaccurate determination results. Therefore, a data analysis method capable of accurately determining the grade probability of stratum surrounding rock is needed. SUMMARY

[0003] The present application aims to at least solve one of the technical problems in the related art.

[0004] The present application proposes a stratum analysis method based on TBM tunneling data, which accurately determines the grade probability of stratum surrounding rock through segmented clustering and classification calculation of cutterhead torque and propulsion speed.

[0005] Another object of the present application is to propose a stratum analysis system based on TBM tunneling data.

[0006] To achieve the above-mentioned purpose, the present application proposes a stratum analysis method based on TBM tunneling data, comprising:

[0007] Obtaining the running data of the TBM tunneling machine; wherein the running data includes cutterhead torque, propulsion speed, cutterhead speed and propulsion force;

[0008] Segmenting according to the preset speed to divide the cutterhead speed range into multiple speed intervals, and assigning a label to each speed interval;

[0009] Segmenting the data in each interval according to the cutterhead torque, and clustering the propulsion speed in each segment to obtain a clustering calculation result;

[0010] According to the clustering calculation result, the class of each batch of new running data in the segmented clustering is calculated, and the distribution of the data in each class is obtained to obtain clustering proportion distribution data;

[0011] Using normalized Gaussian distribution to calculate the surrounding rock grade probability according to the clustering proportion distribution data, and setting the class to represent the continuous surrounding rock grade;

[0012] Based on the surrounding rock grade and the corresponding surrounding rock grade probability, a weighted average calculation is performed to obtain the grade fuzzy value of the stratum.

[0013] The stratum analysis method based on TBM tunneling data according to the embodiments of the application can further have the following additional technical features.

[0014] In an embodiment of the application, the advancing speed is clustered to obtain a clustering calculation result, including:

[0015] The cutter head torque on the x-axis and the advancing speed on the y-axis on the coordinate axis are segmented to obtain a plurality of groups, segmented according to the x-axis and clustered according to the y-axis, and the number of clusters is 2N*3+2, wherein N is the number of categories of surrounding rock.

[0016] The advancing speed of each group is kmeans clustered, and the number of clusters of each group is consistent.

[0017] A restriction condition is imposed on the data participating in the calculation: the advancing speed is less than 1.5 times the advancing speed Vmax, the cutter head torque is less than 1.5 times the cutter head torque Vmax, and the advancing force is greater than 0.

[0018] In an embodiment of the application, the advancing speed of each group is kmeans clustered, and the number of clusters of each group is consistent, including:

[0019] Randomly select k samples as initial centroids {μ1,μ2,...,μ k};

[0020] For each sample x∈X, calculate the distance from it to all centroids: d(x,μ i )=√Σ(x j -μ ij ) 2 ; Assign x to the nearest cluster: c(x)=argmin i ||x-μ i || 2 ;

[0021] For each cluster C i : Recalculate the centroid: μi=(1 / |C i |)Σx∈C i x;

[0022] If the centroid changes less than a threshold value ε or reaches a maximum number of iterations, stop;

[0023] Classify the data of each segment, and require that the label size of each segment is consistent in order;

[0024] Sort the cluster centroids of each group from small to large, and the smallest cluster number is 0 and the largest number is 2N*3+1, to obtain the ordered clusters of all data.

[0025] In one embodiment of the present application, the single batch of data is classified, and Vmax is calculated, including:

[0026] Sort all the data in a column to obtain a sorting result;

[0027] Divide the sorted data into 1000 packages, and the data amount in each package is the package length step;

[0028] Take the minimum value P_min and the maximum value P_max in each package, and calculate the slope P_k of each package = (P_max-P_min) / step;

[0029] Take the selected package and the previous and next 50 packages, and calculate the weighted average slope of the selected package, denoted as K500;

[0030] Multiply K500 by a preset multiple to obtain the threshold K_threshold for calculating Vmax;

[0031] From the maximum value point of the sorting result to the median value point of the sorting, iteratively calculate the slope between the maximum and minimum values in the window with a window size of step and a sliding step of 1, and if the slope is greater than the threshold

[0032] K_threshold, record it as Vmax', and until the smallest value, denoted as Vmax.

[0033] In one embodiment of the present application, the class of each batch of data in the segmented clustering is calculated according to the clustering calculation result, and the distribution of the batch of data in each class is calculated to obtain the clustering proportion distribution data, including:

[0034] Obtain a new batch of data, and search for the first belonging segment of each group of cutter head speed;

[0035] Calculate the second belonging segment of the cutter head torque of each group of data in the first belonging segment;

[0036] Kmeans clustering is performed on the second belonging segment, and the cluster number of the propelling speed is calculated;

[0037] Calculate the cluster number to which all data belongs, and calculate the number of data in each cluster number based on the cluster number to which all data belongs, to calculate the clustering proportion distribution data.

[0038] In one embodiment of the present application, the surrounding rock grade probability is calculated according to the clustering proportion distribution data by using the normalized Gaussian distribution, including:

[0039] A center of data distribution is calculated by using a weighted average method based on the cluster proportion distribution data, to represent the position of the concentration trend of the data distribution on the index;

[0040] An ideal position of each surrounding rock grade is obtained by corresponding each surrounding rock grade to an ideal position in the data space;

[0041] Two grade intervals in which the center is located are found, and a continuous magnitude value is obtained by linear interpolation;

[0042] It is assumed that the ideal position of each surrounding rock grade is a Gaussian distribution center, and a similarity between the center and the ideal position of each surrounding rock grade is calculated by using a Gaussian kernel function;

[0043] Probabilities of each surrounding rock grade are obtained by normalizing all the similarities.

[0044] To achieve the above purpose, another aspect of the present application provides a stratum analysis system based on TBM tunneling data, comprising:

[0045] An operation data acquisition module is configured to acquire operation data of a TBM tunneling machine; wherein the operation data includes cutter head torque, advancing speed, cutter head rotating speed and advancing force;

[0046] A segmented clustering module is configured to segment according to a preset rotating speed to divide a cutter head rotating speed range into multiple rotating speed intervals, and assign a label to each rotating speed interval; segment data in each interval according to cutter head torque, and cluster advancing speed in each segment to obtain a clustering calculation result;

[0047] A cluster proportion calculation module is configured to calculate a class of each batch of new operation data in the segmented clustering according to the clustering calculation result, to obtain a distribution of the data in each class, and obtain cluster proportion distribution data;

[0048] A grade probability calculation module is configured to calculate a surrounding rock grade probability according to the cluster proportion distribution data by using a normalized Gaussian distribution, and set the class to represent a continuous surrounding rock grade;

[0049] A grade ambiguity calculation module is configured to calculate a stratum grade ambiguity value by weighted average calculation based on the surrounding rock grade and the corresponding surrounding rock grade probability.

[0050] The stratum analysis method and system based on TBM tunneling data according to the embodiments of the present application can accurately judge the grade probability of the stratum surrounding rock by segmented clustering and classification calculation, improve the judgment accuracy, apply a limit condition to the data to ensure the effectiveness and accuracy of the data participating in the calculation, and calculate the surrounding rock grade probability by using a normalized Gaussian distribution to make the result more objective and reliable.

[0051] Additional aspects and advantages of the present application will be better understood from the following descriptions, become apparent by the described embodiments, or learned from the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0052] The above and / or additional aspects and advantages of the present application will become apparent and be more readily understood from the following description, taken in conjunction with the accompanying drawings, in which:

[0053] Figure 1 is a flow chart of a stratum analysis method based on TBM tunneling data according to an embodiment of the present application;

[0054] Figure 2 is a running data chart of a TBM tunneling machine according to an embodiment of the present application;

[0055] Figure 3 is a structure chart of a stratum analysis system based on TBM tunneling data according to an embodiment of the present application. DETAILED DESCRIPTION

[0056] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0057] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative work should belong to the protection scope of the present application.

[0058] The stratum analysis method and system based on TBM tunneling data according to the embodiments of the present application will be described below with reference to the accompanying drawings.

[0059] Figure 1 is a flow chart of a stratum analysis method based on TBM tunneling data according to an embodiment of the present application, as shown in Figure 1 , including:

[0060] S1, obtaining running data of a TBM tunneling machine; wherein the running data includes cutterhead torque, advancing speed, cutterhead rotating speed and advancing force.

[0061] It can be understood that, in the process of tunnel construction, in order to better monitor and optimize the running state of TBM (full-face tunnel boring machine), it is necessary to obtain its running data in real time. These data mainly include the following key parameters: cutter torque, pushing speed, cutter rotating speed and pushing force. Cutter torque reflects the resistance that the cutter head bears during tunneling, which is an important indicator for evaluating rock hardness and tunneling efficiency; pushing speed is directly related to construction progress and efficiency; cutter rotating speed affects the cutting effect and wear of the cutter head; pushing force reflects the size of the pushing force required during tunneling, which is closely related to the compressive strength of the rock and the performance of the tunneling machine. Through real-time monitoring and analysis of these running data, tunneling parameters can be adjusted in time, construction scheme can be optimized, construction efficiency can be improved, and construction safety can be ensured.

[0062] As shown in Figure 2 , cutter torque and pushing speed, cutter rotating speed, and pushing force and other parameters can be directly read from the TBM PLC.

[0063] Optionally, before performing TBM tunneling data analysis, preprocessing of running data is a key step to ensure the accuracy and effectiveness of subsequent analysis.

[0064] Firstly, raw running data needs to be obtained from the control system or sensors of the TBM, which usually includes key parameters such as cutter torque, pushing speed, cutter rotating speed and pushing force. Since the actual collected data may have noise, outliers or missing values, it is necessary to clean and filter these data. For example, invalid data caused by equipment downtime or sensor failure is removed, missing values are filled or deleted, and data is normalized to eliminate the dimensional difference and order of magnitude difference between different parameters. In addition, according to the actual working conditions during tunneling, the data needs to be segmented for better analysis of the characteristics of different tunneling stages. Through preprocessing of running data, the quality and consistency of data can be ensured, providing a reliable basis for subsequent feature extraction, model establishment and construction parameter optimization.

[0065] S2, according to the preset rotating speed, the cutter rotating speed range is divided into multiple rotating speed intervals, the data in each interval is segmented according to the cutter torque, and the pushing speed is clustered in each segment to obtain a clustering calculation result.

[0066] Illustratively, the cutter rotating speed can be segmented by every 0.2 rotating speed. Each segment performs the following steps:

[0067] S21, segment clustering of cutter torque (x-axis) - pushing speed (y-axis) data, segmenting by x-axis and clustering by y-axis, with the number of clusters being 2N*3+2, where N is the number of surrounding rock categories, and the closer to the x-axis in the segment clustering, the higher the surrounding rock grade, and the farther away from the x-axis, the lower the surrounding rock grade;

[0068] S22, classify the data within each x-section range using the kmeans algorithm, requiring consistent label size ordering for each section;

[0069] S23, impose the following restrictions on the data involved in the calculation: the advance speed must be less than 1.5 times the advance speed Vmax, the cutterhead torque must be less than 1.5 times the cutterhead torque Vmax, and the thrust force must be greater than 0.

[0070] Specifically, in the analysis of TBM tunneling machine operation data, in order to more finely analyze the tunneling characteristics under different working conditions, the cutterhead speed is first segmented. Specifically, the cutterhead speed range is divided into multiple speed intervals, with each interval being 0.2 of the speed, and each interval is assigned a label (such as A, B, C, etc.) for subsequent analysis. Within each speed interval, the data is further processed and the following steps are performed:

[0071] According to the range of cutterhead torque, the data is segmented in the x-axis direction. For example, if the cutterhead torque range is 0 to 1000 kN·m, it can be segmented by every 100 kN·m as a step.

[0072] Within each cutterhead torque section, cluster analysis is performed on the advance speed (y-axis). The number of clusters is set to 2N×3+2, where N is the number of classes of surrounding rock. For example, if the surrounding rock is divided into 3 classes, the number of clusters is 20.

[0073] In the segmented clustering, the closer to the x-axis (smaller cutterhead torque) the class is, the higher the surrounding rock grade is (i.e., the harder the surrounding rock is); the farther away from the x-axis (larger cutterhead torque) the class is, the lower the surrounding rock grade is (i.e., the softer the surrounding rock is).

[0074] Using the K-means algorithm to classify the data within each x-section range: within each cutterhead torque section, the K-means algorithm is used to cluster the advance speed data, requiring consistent cluster result (label) size ordering for each section to facilitate subsequent analysis and comparison.

[0075] To ensure the validity and reliability of the data, the following restrictions are imposed on the data involved in the calculation:

[0076] The advance speed must be less than 1.5 times the advance speed Vmax.

[0077] The cutterhead torque must be less than 1.5 times the cutterhead torque Vmax.

[0078] The thrust force must be greater than 0 to exclude invalid data caused by equipment downtime or sensor failure.

[0079] Embodiments of the present application segment all data according to the torque of the cutter head by a certain step size to obtain several groups. Then, the push speed of each group is clustered by kmeans clustering, and the number of clusters of each group is consistent. The clustering steps are as follows:

[0080] 1. Initialization: randomly select k samples as initial centroids {μ1, μ2,..., μk}

[0081] 2. Iterate until convergence:

[0082] a. Assignment step:

[0083] For each sample x∈X:

[0084] Calculate its distance to all centroids: d(x, μ i ) = √Σ(x j - μ ij ) 2

[0085] Assign x to the nearest cluster: c(x) = argmin i ||x-μ i || 2

[0086] b. Update step:

[0087] For each cluster C i :

[0088] Recalculate the centroid: μi = (1 / |C i |)Σx∈C i x

[0089] c. Check termination condition:

[0090] If the centroid changes are less than a threshold ε or the maximum number of iterations is reached, stop

[0091] The kmeans algorithm is used to classify the data of each segment, and the label size of each segment is required to be consistent in order.

[0092] The cluster centroids of each group are sorted from small to large, and the smallest cluster number is 0 and the largest number is 2N*3+1, to obtain the ordered clusters of all data.

[0093] Further, the new running data of the single ring / batch is classified and calculated, and the data is processed according to the above Vmax requirements. The detailed calculation steps of Vmax are as follows:

[0094] Sort all data in a column to obtain the sorting results, such as the push speed.

[0095] The sorted data is divided into 1000 packages, and the data amount in each package is the package length step;

[0096] The minimum value P_min and the maximum value P_max in each package are taken, and the slope P_k of each package is calculated as (P_max-P_min) / step.

[0097] The slope of the selected package is calculated by weighted average calculation of the previous and next 50 packages of the package where the median value of all data is located, and is recorded as K500

[0098] K500 is multiplied by a multiple (such as 50) to obtain the threshold K_threshold for calculating Vmax

[0099] From the maximum value point of the sorting result to the median value point of the sorting result, the slope between the maximum and minimum values in the window is calculated, the sliding step is 1, and if the slope is greater than the threshold K_threshold, it is recorded as Vmax', until the smallest value, recorded as Vmax.

[0100] Through the above steps, the complex continuous data can be converted into segmented discrete data, and the tunneling characteristics under different surrounding rock conditions can be identified through clustering analysis. This method not only improves the accuracy of data analysis, but also provides a reliable basis for subsequent stratum classification and construction parameter optimization.

[0101] S3, according to the clustering calculation result, the category of each batch of new running data in the segmented clustering is calculated, and the distribution of the data in each category is obtained to obtain the clustering proportion distribution data.

[0102] Specifically, according to the set size, the speed of each ring or batch of new data can be segmented (such as A, B, C…), and classified and calculated according to the above cutter torque segmentation and advancing speed clustering (for example, if it is A segment, the cutter torque and advancing speed clustering results in the A segment are selected for classification and calculation), the category of each data in the segmented clustering is calculated, and the distribution of the data in each category, i.e. the proportion in each cluster, can be obtained after all data is calculated. The specific steps can be as follows:

[0103] S31, a batch of new data is obtained, and each group of cutter speed belonging to the segment is found;

[0104] S32, the cutter torque belonging to the segment in each group of data in the speed segment is calculated;

[0105] S33, find the kmeans cluster of the belonging segment, calculate the cluster number where the advancing speed is located, repeat S31 and S32 to calculate the cluster number where all data belong to;

[0106] S34, based on the result of S33, the number of data in each cluster number can be calculated, so as to calculate the probability of its proportion (normalized probability).

[0107] Specifically, first, a batch of new operation data is obtained from the control system or sensor of the TBM tunneling machine, including parameters such as cutter torque, advancing speed, cutter rotating speed and advancing force.

[0108] According to the preset rotating speed interval (such as 0.2r / min per step), the cutter rotating speed range is divided into multiple rotating speed intervals, and a label (such as A, B, C, etc.) is assigned to each interval. For each group of data, determine the rotating speed interval to which the cutter rotating speed belongs.

[0109] In each rotating speed interval, segment according to the range of cutter torque. For each group of data, determine the torque segment where the cutter torque is located.

[0110] In each cutter torque segment, K-means algorithm is used for cluster analysis of the advancing speed data.

[0111] For each group of data, find the cluster number to which it belongs.

[0112] For all data, repeat the above steps to calculate the cluster number of each group of data. Count the number of data in each cluster number. Calculate the proportion probability of each cluster number, that is, the number of data in the cluster number divided by the total data amount. Normalize all proportion probabilities so that the sum is 1.

[0113] Through the above steps, the cluster number of each group of data in different rotating speed and torque segments and the proportion probability of each cluster number can be obtained. These results can be used for further surrounding rock grade judgment and stratum classification, and provide a scientific basis for TBM tunneling machine construction parameter selection and stratum classification.

[0114] S4, using the normalized Gaussian distribution to calculate the surrounding rock grade probability according to the cluster proportion distribution data, and setting the category representation as continuous surrounding rock grade.

[0115] Specifically, the present application constructs a grade probability classification algorithm, the input is the probability value of each cluster, and the output is the probability of each grade of surrounding rock grade. The calculation idea is: the trend of the weighted center positioning distribution is located, and linear interpolation and Gaussian modeling are combined with ideal grade position to output the probability belonging to each discrete grade. The steps are as follows:

[0116] S41, calculate the "center" of the data distribution using weighted average method: center is the location of the central tendency of the entire distribution on the index (may be a floating point number).

[0117] First, according to each cluster number and its corresponding probability value, the "center" of the data distribution is calculated using the weighted average method, and the formula is:

[0118]

[0119] where c i is the i-th cluster number, p i is the probability value of the i-th cluster.

[0120] "center" is the location of the central tendency of the entire distribution on the index, which may be a floating point number.

[0121] S42, map each level to the "ideal position" in the data space.

[0122] Assume that each surrounding rock level corresponds to an ideal position, which can be determined based on historical data or prior knowledge.

[0123] For example, if the surrounding rock level is divided into 3, its ideal position can be μ1, μ2, μ3.

[0124] S43, find the two level intervals where the center is located, and perform linear interpolation to obtain a continuous "magnitude" value.

[0125] Determine the two level intervals where the "center" is located, such as μi and μi+1.

[0126] Use the linear interpolation formula to calculate a continuous "magnitude" value:

[0127]

[0128] S44, assume that the ideal position of each level is a Gaussian distribution center. Calculate the similarity (using Gaussian kernel) of center to each level_position.

[0129] Assume that the ideal position of each surrounding rock level is a Gaussian distribution center, and use the Gaussian kernel function to calculate the similarity of "center" to each level ideal position:

[0130]

[0131] where σ is the bandwidth parameter of the Gaussian kernel.

[0132] S45, normalize all the probabilities of the grades so that the sum is 1.

[0133] Specifically, all the similarity values S i are normalized so that the sum is 1:

[0134]

[0135] where N is the total number of surrounding rock grades.

[0136] S5, based on the surrounding rock grades and the corresponding surrounding rock grade probabilities, a weighted average is calculated to obtain the grade fuzzy value of the stratum.

[0137] Specifically, in the analysis of TBM tunneling machine operation data, after obtaining the probability of each surrounding rock grade through the previous steps, the fuzzy value of the surrounding rock grade can be further calculated. This fuzzy value is calculated by a weighted average method, which integrates the probabilities of all surrounding rock grades and the corresponding grade values. The specific steps are as follows:

[0138] First, determine the surrounding rock grades and their probabilities: let the surrounding rock grades be w1, w2, w3, …, w N , and the corresponding probabilities be p1, p2, p3, …, p N .

[0139] Calculate the fuzzy value of the surrounding rock grade:

[0140] Use the weighted average formula to calculate the fuzzy value of the surrounding rock grade:

[0141]

[0142] Here w i is the value of the i-th surrounding rock grade, and p i is the probability of the i-th surrounding rock grade.

[0143] The fuzzy value is a value between the minimum surrounding rock grade value and the maximum surrounding rock grade value, which reflects the concentration trend of the probability distribution of all surrounding rock grades. The closer the fuzzy value is to a specific surrounding rock grade value, the greater the likelihood of that surrounding rock grade.

[0144] The stratum analysis method based on TBM tunneling data according to the embodiment of the present application can effectively determine the stratum currently being tunneled by the TBM, and can assist the main driver in selecting tunneling parameters and accurately determining the grade probability of the surrounding rock of the stratum.

[0145] To implement the above-mentioned embodiments, as Figure 3 shown, the present embodiment further provides a stratum analysis system 10 based on TBM tunneling data, comprising:

[0146] The running data acquisition module 100 is configured to acquire running data of the TBM tunneling machine; wherein the running data comprises a cutterhead torque, a propulsion speed, a cutterhead rotating speed and a propulsion force;

[0147] The segmented clustering module 200 is configured to segment according to a preset rotating speed to divide the cutterhead rotating speed range into a plurality of rotating speed intervals, segment the data in each interval according to the cutterhead torque, and cluster the propulsion speed in each segment to obtain a clustering calculation result;

[0148] The cluster proportion calculation module 300 is configured to calculate the category of each batch of new running data in the segmented clustering according to the clustering calculation result, to obtain the distribution of the data in each category, and to obtain cluster proportion distribution data;

[0149] The grade probability calculation module 400 is configured to calculate a surrounding rock grade probability according to the cluster proportion distribution data by using a normalized Gaussian distribution, and to set the category to represent a continuous surrounding rock grade;

[0150] The grade fuzzy calculation module 500 is configured to perform a weighted average calculation based on the surrounding rock grade and the corresponding surrounding rock grade probability to obtain a grade fuzzy value of the stratum.

[0151] Further, the segmented clustering module 200 is further configured to:

[0152] segment the cutterhead torque on the x-axis and the propulsion speed on the y-axis on the coordinate axis to obtain a plurality of groups, segment according to the x-axis and cluster according to the y-axis, and the number of clusters is 2N*3+2, wherein N is the number of categories of the surrounding rock;

[0153] perform kmeans clustering on the propulsion speed of each group, and the number of clusters of each group is consistent;

[0154] apply a restriction condition to the data participating in the calculation: the propulsion speed is less than 1.5 times the propulsion speed Vmax, the cutterhead torque is less than 1.5 times the cutterhead torque Vmax, and the propulsion force is greater than 0.

[0155] Further, the kmeans clustering is performed on the propulsion speed of each group, and the number of clusters of each group is consistent, including:

[0156] randomly select k samples as initial centroids {μ1, μ2,..., μk};

[0157] for each sample x∈X, calculate the distance from x to all centroids: d(x, μ i )=√Σ(x j -μ ij ) 2 ; and assign x to the nearest cluster: c(x)=argmin i ||x-μ i ||2 ;

[0158] For each cluster C i : Recalculate the centroid: μi = (1 / |C i |)Σx∈C i x;

[0159] If the centroid changes less than a threshold ε or reaches a maximum number of iterations, stop.

[0160] Classify each segment of data, requiring the label size of each segment to be consistent in order;

[0161] Sort the cluster centroids of each group from small to large, with the smallest cluster number being 0 and the largest number being 2N*3+1, to obtain the ordered clusters of all data.

[0162] Further, classify the single batch of data and calculate Vmax, including:

[0163] Sort all data in a column to obtain the sorting result;

[0164] Divide the sorted data into 1000 packages, and the data amount in each package is the package length step;

[0165] Take the minimum value P_min and the maximum value P_max in each package to calculate the slope P_k of each package = (P_max-P_min) / step;

[0166] Take the selected package and the previous and next 50 packages, and calculate the slope of the selected package by weighted average, denoted as K500;

[0167] Multiply K500 by a predetermined multiple to obtain the threshold K_threshold for calculating Vmax;

[0168] From the maximum value point of the sorting result to the median value point of the sorting, iteratively calculate the slope between the maximum and minimum values in the window with a window size of step and a sliding step of 1, and if the slope is greater than the threshold

[0169] K_threshold, record it as Vmax', until the smallest value, denoted as Vmax.

[0170] The stratum analysis system based on TBM tunneling data according to the embodiments of the present application can effectively determine the stratum of the current TBM tunneling, can assist the main driver in selecting tunneling parameters, and can accurately determine the grade probability of the stratum surrounding rock.

[0171] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the description of the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples, without contradiction.

[0172] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise specifically limited.

Claims

1. A geological analysis method based on TBM tunneling data, characterized in that, include: Acquire the operating data of the TBM tunneling machine; wherein the operating data includes cutterhead torque, feed speed, cutterhead rotation speed and feed force; The cutter head speed range is divided into multiple speed intervals based on a preset speed, and a label is assigned to each speed interval. The data within each interval is segmented according to the cutterhead torque, and the feed speed is clustered within each segment to obtain the clustering calculation results; Based on the clustering calculation results, calculate the category of each batch of new running data in the segmented clustering, and obtain the distribution of the data in each category to obtain the clustering proportion distribution data; The probability of the surrounding rock grade is calculated based on the cluster proportion distribution data using a normalized Gaussian distribution, and the category is defined as a continuous surrounding rock grade. Based on the surrounding rock grade and the corresponding surrounding rock grade probability, a weighted average is used to calculate the fuzzy value of the stratum grade.

2. The method according to claim 1, characterized in that, Clustering the propulsion speeds yields the following clustering results: The cutterhead torque on the x-axis and the feed speed on the y-axis are segmented to obtain several groups. The segments are divided according to the x-axis and clustered according to the y-axis. The number of clusters is 2N*3+2, where N is the number of surrounding rock categories. Perform k-means clustering on the advancement speed of each group, with the same number of clusters in each group; Constraints are imposed on the data used in the calculation: the propulsion speed is less than 1.5 times the propulsion speed Vmax, the cutterhead torque is less than 1.5 times the cutterhead torque Vmax, and the propulsion force is greater than 0.

3. The method according to claim 1, characterized in that, For each group's propulsion speed, perform k-means clustering, ensuring the same number of clusters per group, including: Randomly select k samples as the initial centroids {μ1,μ2,...,μ k }; For each sample x∈X, calculate its distance to all centroids: d(x,μ) i )=√Σ(x j -μ ij ) 2 Assign x to the nearest cluster: c(x) = argmin i ||x-μ i || 2 ; For each cluster C i Recalculate the centroid: μi = (1 / |C i |)Σx∈C i x; If the change in centroid is less than the threshold ε or the maximum number of iterations is reached, then stop. Classify the data in each segment, ensuring that the label size and order are consistent for each segment; Sort the centroids of each cluster in ascending order, with the smallest cluster numbered 0 and the largest numbered 2N*3+1, to obtain the ordered clusters of all data.

4. The method according to claim 1, characterized in that, Classify the data in a single batch and calculate Vmax, including: Sort all the data in one column to get the sorted result; The sorted data is divided into 1000 packages, and the amount of data in each package is the package length step; Take the minimum value P_min and the maximum value P_max in each package, and calculate the slope P_k = (P_max - P_min) / step for each package; Take the 50 packages before and after the package containing the median value from all data, calculate the slope of the selected package using a weighted average, and denot it as K500; Multiply K500 by a preset factor to obtain the threshold K_threshold for calculating Vmax; Iterative calculations begin from the maximum value in the sorted results to the median value. The window size is step, and the sliding step size is 1. The slope between the maximum and minimum values ​​within the window is calculated. If the slope is greater than the threshold K_threshold, it is recorded as Vmax'. This process continues until the minimum value is reached, which is then recorded as Vmax.

5. The method according to claim 1, characterized in that, Based on the clustering calculation results, the category of each batch of data in the segmented clustering is calculated, and the distribution of all data in each category is calculated to obtain the clustering proportion distribution data, including: Acquire a new batch of data and search for the first segment of each group of cutterhead rotation speeds; Calculate the second segment where the cutter head torque is located within the first segment for each set of data; Perform k-means clustering on the second segment and calculate the cluster number where the advancement speed belongs; Calculate the cluster number to which all data belong, and calculate the number of data in each cluster number based on the cluster number to which all data belong, so as to calculate the cluster proportion distribution data.

6. The method according to claim 1, characterized in that, The probability of surrounding rock grade is calculated using a normalized Gaussian distribution based on the cluster proportion distribution data, including: The center of gravity of the data distribution is calculated using a weighted average method based on the cluster distribution data, in order to represent the position of the central tendency of the data distribution on the index. The ideal position for each rock mass grade is obtained by mapping each rock mass grade to an ideal position in the data space; Find the two level intervals where the center is located, and perform linear interpolation to obtain a continuous magnitude value; Assuming the ideal location for each rock grade is a Gaussian distribution center, the similarity between the center and the ideal location for each rock grade is calculated using a Gaussian kernel function; Normalize all similarities to obtain the probability of each surrounding rock grade.

7. A geological analysis system based on TBM tunneling data, characterized in that, include: The operation data acquisition module is used to acquire the operation data of the TBM tunneling machine; wherein, the operation data includes cutterhead torque, feed speed, cutterhead rotation speed and feed force; The segmented clustering module is used to divide the cutter head speed range into multiple speed intervals according to a preset speed, and assign a label to each speed interval; the data in each interval is segmented according to the cutter head torque, and the feed speed is clustered within each segment to obtain the clustering calculation result; The clustering percentage calculation module is used to calculate the category of each batch of new running data in the segmented clustering based on the clustering calculation results, and to obtain the distribution of the data in each category, so as to obtain the clustering percentage distribution data. The grade probability calculation module is used to calculate the surrounding rock grade probability based on the cluster proportion distribution data using a normalized Gaussian distribution, and to set the category representation as a continuous surrounding rock grade; The fuzzy classification calculation module is used to calculate the fuzzy classification value of the strata by weighted averaging based on the surrounding rock grade and the corresponding surrounding rock grade probability.

8. The system according to claim 7, characterized in that, The segmented clustering module is also used for: The cutterhead torque on the x-axis and the feed speed on the y-axis are segmented to obtain several groups. The segments are divided according to the x-axis and clustered according to the y-axis. The number of clusters is 2N*3+2, where N is the number of surrounding rock categories. Perform k-means clustering on the advancement speed of each group, with the same number of clusters in each group; Constraints are imposed on the data used in the calculation: the propulsion speed is less than 1.5 times the propulsion speed Vmax, the cutterhead torque is less than 1.5 times the cutterhead torque Vmax, and the propulsion force is greater than 0.

9. The system according to claim 7, characterized in that, For each group's propulsion speed, perform k-means clustering, ensuring the same number of clusters per group, including: Randomly select k samples as the initial centroids {μ1,μ2,...,μ k }; For each sample x∈X, calculate its distance to all centroids: d(x,μ) i )=√Σ(x j -μ ij ) 2 Assign x to the nearest cluster: c(x) = argmin i ||x-μ i || 2 ; For each cluster C i Recalculate the centroid: μ i =(1 / |C i |)Σx∈C i x; If the change in centroid is less than the threshold ε or the maximum number of iterations is reached, then stop. Classify the data in each segment, ensuring that the label size and order are consistent for each segment; Sort the centroids of each cluster in ascending order, with the smallest cluster numbered 0 and the largest numbered 2N*3+1, to obtain the ordered clusters of all data.

10. The system according to claim 7, characterized in that, Classify the data in a single batch and calculate Vmax, including: Sort all the data in one column to get the sorted result; The sorted data is divided into 1000 packages, and the amount of data in each package is the package length step; Take the minimum value P_min and the maximum value P_max in each package, and calculate the slope P_k = (P_max - P_min) / step for each package; Take the 50 packages before and after the package containing the median value from all data, calculate the slope of the selected package using a weighted average, and denot it as K500; Multiply K500 by a preset factor to obtain the threshold K_threshold for calculating Vmax; Iterative calculations begin from the maximum value in the sorted results to the median value. The window size is step, and the sliding step size is 1. The slope between the maximum and minimum values ​​within the window is calculated. If the slope is greater than the threshold K_threshold, it is recorded as Vmax'. This process continues until the minimum value is reached, which is then recorded as Vmax.

Images