A targeted numerical analysis method for high-risk tunnel sections

Abstract

The present invention provides a targeted numerical analysis method for a high-risk tunnel, comprising: obtaining a plurality of tunnel numerical value groups of a high-risk tunnel; preprocessing the numerical values ​​in each tunnel numerical value group to obtain a plurality of preprocessed tunnel numerical value groups; obtaining a targeted numerical value group corresponding to each preprocessed tunnel numerical value group based on the preprocessed tunnel numerical values ​​in each preprocessed tunnel numerical value group and the corresponding numerical acquisition time; analyzing the targeted numerical values ​​in each targeted numerical value group to obtain a plurality of targeted numerical analysis results; generating a tunnel analysis report for the high-risk tunnel according to the plurality of targeted numerical analysis results. The method achieves the technical effect of obtaining the tunnel condition analysis results of the high-risk tunnel by accurately and effectively analyzing the targeted numerical values ​​of the high-risk tunnel, thereby timely discovering the problems existing in the high-risk tunnel and performing tunnel maintenance according to the tunnel condition analysis results, thereby ensuring the safety, stability and availability of the high-risk tunnel.

Description

Technical Field

[0001] The invention relates to the technical field of numerical analysis, and in particular to a high-risk section tunnel targeted numerical analysis method. Background Art

[0002] High-risk tunnel sections refer to tunnel sections that have higher engineering and construction risks due to specific geological conditions, engineering design and construction restrictions. These tunnel sections usually have geological and hydrogeological problems such as extrusion, strong weathering, high geostress, fault activity, and high water levels. They may also be affected by complex geological structures, stress states, and overloads. High-risk tunnel sections may experience rock bursts and collapses, water gushing and flooding accidents, fire and smoke accidents, and traffic accidents, causing huge damage to people's lives and property.

[0003] Therefore, in order to ensure the safety of people's lives and property and to ensure the safety and availability of high-risk tunnels, it is necessary to avoid the above problems that may occur in high-risk tunnels as much as possible. The above problems can be effectively avoided by analyzing the targeted numerical values ​​of high-risk tunnels. Analyzing the targeted numerical values ​​of high-risk tunnels is a crucial engineering practice, which involves many aspects such as tunnel engineering safety, efficiency and cost control. This analysis can help engineers and decision makers better understand and predict potential risks, so as to take appropriate preventive measures to ensure the smooth progress of the project and its long-term stability.

[0004] Analyzing the targeted numerical values ​​of high-risk tunnels has the following advantages: (1) By analyzing the targeted numerical values ​​of high-risk tunnels, it is possible to identify which parts may have major safety hazards or technical challenges, so as to solve them in a timely manner; (2) By analyzing the targeted numerical values ​​of high-risk tunnels, it is possible to predict and prevent factors that may cause serious safety accidents such as collapse, water damage, and fire, and protect the lives of workers and later users; (3) Construction efficiency: Through targeted numerical analysis, a more reasonable construction plan and schedule can be formulated, reducing delays caused by unknown factors that may be encountered during the construction process and improving construction efficiency. However, the existing technical solutions lack technical means for accurately and effectively analyzing the targeted numerical values ​​of high-risk tunnels.

[0005] Therefore, how to accurately and effectively analyze the targeted numerical values ​​of high-risk tunnels to obtain the tunnel condition analysis results of high-risk sections of tunnels, so as to timely discover problems in high-risk sections of tunnels and perform tunnel maintenance based on the tunnel condition analysis results, thereby ensuring the safety, stability and availability of high-risk sections of tunnels, is one of the urgent problems to be solved in the field of numerical analysis technology. Summary of the invention

[0006] The present invention aims to solve the technical problems in the above-mentioned technologies to at least a certain extent. To this end, the purpose of the present invention is to provide a method for targeted numerical analysis of high-risk tunnel sections, by preprocessing several tunnel numerical groups obtained for high-risk tunnel sections to obtain several preprocessed tunnel numerical groups, determining several targeted numerical groups based on the several preprocessed tunnel numerical groups, and analyzing the several targeted numerical groups to obtain several targeted numerical analysis results, thereby achieving the technical effect of obtaining tunnel condition analysis results of high-risk tunnel sections by accurately and effectively analyzing the targeted numerical values ​​of high-risk tunnels, thereby timely discovering problems existing in high-risk tunnel sections and performing tunnel maintenance according to the tunnel condition analysis results, thereby ensuring the safety, stability and availability of high-risk tunnel sections.

[0007] The present invention provides a high-risk tunnel targeted numerical analysis method, comprising:

[0008] Obtaining several tunnel value groups of high-risk tunnel sections;

[0009] Preprocessing the values ​​in each tunnel value group to obtain a plurality of preprocessed tunnel value groups;

[0010] Based on the preprocessing tunnel values ​​in each preprocessing tunnel value group and the corresponding value acquisition time, a targeting value group corresponding to each preprocessing tunnel value group is obtained;

[0011] Analyze the target values ​​in each target value group to obtain several target value analysis results;

[0012] Generate a tunnel analysis report for high-risk tunnel sections based on several targeted numerical analysis results.

[0013] Preferably, in a targeted numerical analysis method for high-risk sections of tunnels, several tunnel numerical groups include a tunnel groundwater level numerical group, a tunnel groundwater pressure numerical group, a tunnel temperature numerical group, a tunnel humidity numerical group, a tunnel gas content numerical group, a tunnel rock stratum inclination numerical group, a tunnel rock strength numerical group, a longitudinal and transverse wave velocity numerical group when propagating in the tunnel, a tunnel convergence numerical group, and a tunnel offset numerical group.

[0014] Preferably, in the targeted numerical analysis method for high-risk tunnel sections, the tunnel gas content numerical value group includes a tunnel carbon monoxide content numerical value group, a tunnel carbon dioxide content numerical value group, a tunnel combustible gas content numerical value group and a tunnel oxygen content numerical value group.

[0015] Preferably, the high-risk tunnel targeted numerical analysis method preprocesses the values ​​in each tunnel numerical value group to obtain a plurality of preprocessed tunnel numerical value groups, including:

[0016] Outliers are eliminated from the values ​​in each tunnel value group to obtain several optimized tunnel value groups;

[0017] The missing values ​​of each optimized tunnel value group are filled to obtain several preprocessed tunnel value groups.

[0018] Preferably, the high-risk tunnel targeted numerical analysis method removes abnormal values ​​from each tunnel numerical value group to obtain several optimized tunnel numerical value groups, including:

[0019] Select any tunnel value group as the tunnel value group to be optimized, sort the values ​​in the tunnel value group to be optimized in the order of value acquisition time from earliest to latest, and obtain a sorted tunnel value group;

[0020] Determine the partitioning step size based on the number of values ​​in the sorted tunnel value group;

[0021] Based on the partition step, the sorted tunnel value group is evenly divided into a number of sub-tunnel value groups;

[0022] Determine any sub-tunnel value group as a target tunnel value group, and determine a subsequent sub-tunnel value group of the target tunnel value group as an associated tunnel value group of the target tunnel value group;

[0023] Calculate a number of Pearson correlation coefficients between each value in the target tunnel value group and the value at the corresponding position in the associated tunnel value group;

[0024] Sort the Pearson coefficients in descending order, and determine the values ​​in the target tunnel value group corresponding to the first N Pearson coefficients as the first abnormal value;

[0025] Determine the median of all values ​​of the target tunnel value group except the first abnormal value;

[0026] Calculate the ratio of each first outlier to the median, and when the ratio is less than a preset ratio threshold, determine the first outlier corresponding to the ratio as the second outlier;

[0027] Calculate the mean of all values ​​in the target tunnel value group except the first abnormal value;

[0028] Calculate the absolute difference between each second outlier and the mean, and use the ratio of the absolute difference to the mean as the deviation coefficient of the second outlier;

[0029] The second outlier value whose deviation coefficient is greater than the preset deviation threshold is used as the final outlier value of the target tunnel value group, and the final outlier value is removed from the target tunnel value group to obtain an optimized target tunnel value group;

[0030] The above operation is performed on all sub-tunnel value groups to obtain the optimized target tunnel value group corresponding to each sub-tunnel value group, and all optimized target tunnel value groups constitute the tunnel value group to be optimized corresponding to the tunnel value group to be optimized;

[0031] The above operation is performed on all classified tunnel value groups to obtain several optimized tunnel value groups.

[0032] Preferably, the high-risk tunnel targeted numerical analysis method performs missing value filling on each optimized tunnel numerical value group to obtain a number of pre-processed tunnel numerical value groups, including:

[0033] Select any optimized tunnel value group as the tunnel value group to be filled, and use the position where the abnormal value is removed in the tunnel value group to be filled as the missing value filling position;

[0034] When the digits before and after the missing value filling position have values, the type of the missing value filling position is determined to be the first type of missing value filling position; otherwise, the type of the missing value filling position is determined to be the second type of missing value filling position;

[0035] When the missing value filling position is the first type of missing value filling position, the mean of the two digits before and after the missing value filling position is used as the filling value of the missing value filling position;

[0036] When the missing value filling position is the second type of missing value filling position, the missing value filling position with a value in the previous adjacent position is selected as the target filling position, and any target filling position is selected as the filling position to be processed;

[0037] Determine the value before the to-be-processed filling position as an auxiliary value, and count all other values ​​in the to-be-filled tunnel value group that are the same as the auxiliary value and their next adjacent values;

[0038] Generate a numerical conversion table based on all other numerical values ​​that are the same as the auxiliary numerical value and their next adjacent numerical values;

[0039] According to the numerical conversion table, calculate the probability that the value that is the same as the auxiliary value is converted into the same adjacent value, and use the probability as the weight of the adjacent value;

[0040] Determine the padding value of the to-be-processed padding position based on all adjacent values ​​and corresponding weights;

[0041] Perform the above operation on all target filling positions to obtain the filling value corresponding to each target filling position;

[0042] Perform the above operation on all the second-category missing value filling positions to obtain the filling value corresponding to each second-category missing value filling position;

[0043] The tunnel value group to be filled that has completed the value filling is used as the pre-processed tunnel value group, and the above operation is performed on all the optimized tunnel value groups to obtain a plurality of pre-processed tunnel value groups.

[0044] Preferably, the high-risk tunnel targeted numerical analysis method obtains a targeted numerical value group corresponding to each preprocessing tunnel numerical value group based on the preprocessing tunnel numerical value in each preprocessing tunnel numerical value group and the corresponding numerical value acquisition time, including:

[0045] Select any pre-processed tunnel value group as the to-be-processed value group, and match each value with a number of preset standard values ​​based on the value acquisition time corresponding to the values ​​in the to-be-processed value group to obtain the standard value corresponding to each value;

[0046] Calculate the absolute difference between each value and the corresponding standard value, and use the ratio of the absolute difference to the standard value as the target value corresponding to the value;

[0047] The target values ​​and target value generation times corresponding to all values ​​in the to-be-processed value group constitute a target value group corresponding to the to-be-processed value group; wherein the target value generation time is the value acquisition time of the corresponding value;

[0048] The above operation is performed on all pre-processing tunnel value groups to obtain a number of target value groups corresponding to the pre-processing tunnel value groups.

[0049] Preferably, the high-risk tunnel targeted numerical analysis method analyzes the targeted numerical values ​​in each targeted numerical value group to obtain several targeted numerical analysis results, including:

[0050] Select any target value group as the target value group to be processed, and obtain a target value change curve graph based on the target values ​​and target value generation time in the target value group to be processed;

[0051] Generate a standard threshold line in the target value change curve graph, take the points in the target value change curve graph that are higher than the standard threshold line as mutation points, and count the number of mutation points as the number of mutation points;

[0052] Calculate the average of the target values ​​corresponding to all mutation points as the mutation average;

[0053] When the number of mutation points is greater than the point threshold and the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the first analysis result;

[0054] When the number of mutation points is greater than the point threshold but the average mutation value is less than the average threshold, determining the targeted numerical analysis result as the second analysis result;

[0055] When the number of mutation points is less than the point threshold but the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the third analysis result;

[0056] When the number of mutation points is less than the point threshold and the average mutation value is less than the average threshold, the targeted numerical analysis result is determined to be the fourth analysis result.

[0057] Preferably, the high-risk tunnel targeted numerical analysis method generates a tunnel analysis report for the high-risk tunnel according to a number of targeted numerical analysis results, including:

[0058] Fill in the basic information of the high-risk section tunnel in the tunnel report template;

[0059] Filling a plurality of targeted numerical analysis results and corresponding targeted numerical variation curves in a preset result display position of a tunnel report template;

[0060] Based on a number of targeted numerical analysis results, corresponding tunnel inspection suggestions are generated, and the tunnel inspection suggestions are filled into the preset suggestion filling positions of the tunnel report template to generate a tunnel analysis report for the high-risk section tunnel.

[0061] Preferably, the high-risk tunnel targeted numerical analysis method, after generating a tunnel analysis report of the high-risk tunnel according to a plurality of targeted numerical analysis results, further comprises:

[0062] Based on the tunnel inspection opinions in the tunnel analysis report, the tunnel is inspected to obtain the tunnel inspection results;

[0063] Based on the tunnel inspection results, the tunnel is maintained.

[0064] The present invention provides a targeted numerical analysis method for high-risk tunnels, including: obtaining several tunnel numerical value groups of high-risk tunnels; preprocessing the values ​​in each tunnel numerical value group to obtain several preprocessed tunnel numerical value groups; obtaining the targeted numerical value group corresponding to each preprocessed tunnel numerical value group based on the preprocessed tunnel numerical value in each preprocessed tunnel numerical value group and the corresponding numerical acquisition time; analyzing the targeted numerical value in each targeted numerical value group to obtain several targeted numerical analysis results; generating a tunnel analysis report for the high-risk tunnel according to the several targeted numerical analysis results. The method achieves the technical effect of obtaining the tunnel condition analysis results of the high-risk tunnel by accurately and effectively analyzing the targeted numerical values ​​of the high-risk tunnel, thereby timely discovering the problems existing in the high-risk tunnel and performing tunnel maintenance according to the tunnel condition analysis results, thereby ensuring the safety, stability and availability of the high-risk tunnel.

[0065] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0066] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0068] Figure 1 This is a flow chart of a high-risk tunnel targeted numerical analysis method in an embodiment of the present invention;

[0069] Figure 2 This is an optional flow chart of obtaining a target value group in an embodiment of the present invention;

[0070] Figure 3 This is an optional flow chart for determining a target numerical analysis result in an embodiment of the present invention;

[0071] Figure 4 A schematic diagram of an optional tunnel analysis report in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0073] Reference Figure 1 The embodiment of the present invention provides a high-risk tunnel targeted numerical analysis method, including:

[0074] Step S1, obtaining a plurality of tunnel value groups of high-risk tunnel sections;

[0075] Step S2, preprocessing the values ​​in each tunnel value group to obtain a plurality of preprocessed tunnel value groups;

[0076] Step S3, obtaining a targeting value group corresponding to each preprocessing tunnel value group based on the preprocessing tunnel value in each preprocessing tunnel value group and the corresponding value acquisition time;

[0077] Step S4, analyzing the target values ​​in each target value group to obtain a plurality of target value analysis results;

[0078] Step S5, generating a tunnel analysis report for the high-risk section tunnel according to a number of targeted numerical analysis results.

[0079] In this embodiment, the stable seepage mode is used for equilibrium calculation and tunnel excavation in the high-risk section tunnel. During the excavation process, the tunnel body is an impermeable model, the tunnel wall and the tunnel face are free boundaries, and the shape change of rock particles is ignored.

[0080] In this embodiment, the surrounding rock of the high-risk tunnel section is regarded as the ideal elastic-plastic material of Moore-Kulan, the supporting structure is regarded as the elastic material, and the steel arch is set in full ring. The supporting effect of the steel arch and the steel mesh is converted according to the bending stiffness equivalent principle, and the converted concrete elastic modulus is as follows: Among them, E c is the converted elastic modulus of concrete, E 0 is the elastic modulus of the original concrete, A s is the area of ​​the steel arch, E s is the cross-sectional area of ​​the steel arch, A c The concrete area.

[0081] In this embodiment, the high-risk section tunnel excavation was carried out in the excavation-support-excavation mode on site, using the two-step method and the three-step method respectively, with the excavation cycle advance of 2m and the step safety distance of 10m. When the tunnel was excavated by the two-step method and the three-step method, plastic zones were mainly generated at the face, spandrel and arch foot, with shear failure as the main failure, and tensile shear failure occurred locally at the arch top and the bottom of the excavation face. The plastic zone range generated by the two-step method excavation face is larger than that generated by the three-step method excavation, and the disturbance to the surrounding rock in front of the face is also greater, which is more concentrated at the spandrel and arch foot. However, the plastic zones of the two construction methods are not connected, that is, the stability of the tunnel surrounding rock can be ensured. The pore water pressure of the surrounding rock in the two excavation conditions showed a layer-by-layer increase. Affected by the excavation action, the surrounding rock behind the face, especially the upper area, was reduced the most, and a low-value area of ​​pore water pressure appeared near the face. At the same time, in front of the tunnel face, the pore water pressure gradually increases away from the tunnel face along the excavation direction. The pore water pressure of the arch waist begins to stabilize after 30 steps of the three-step method, while it gradually stabilizes after 15 steps of the two-step method. As the tunnel face continues to advance, the pore water pressure at each monitoring point eventually gradually approaches 0.

[0082] The technical principle and technical effect of the above technical solution are: obtaining several tunnel value groups of high-risk tunnels; preprocessing the values ​​in each tunnel value group to obtain several preprocessed tunnel value groups; obtaining the targeted value group corresponding to each preprocessed tunnel value group based on the preprocessed tunnel values ​​in each preprocessed tunnel value group and the corresponding value acquisition time; analyzing the targeted values ​​in each targeted value group to obtain several targeted value analysis results; generating a tunnel analysis report for high-risk tunnels based on several targeted value analysis results. The technical effect of obtaining the tunnel condition analysis results of high-risk tunnels by accurately and effectively analyzing the targeted values ​​of high-risk tunnels is achieved, so that problems in high-risk tunnels can be discovered in time based on the tunnel condition analysis results and tunnel maintenance can be performed, thereby ensuring the safety, stability and availability of high-risk tunnels.

[0083] An embodiment of the present invention provides a targeted numerical analysis method for high-risk tunnel sections, wherein several tunnel numerical groups include a tunnel groundwater level numerical group, a tunnel groundwater pressure numerical group, a tunnel temperature numerical group, a tunnel humidity numerical group, a tunnel gas content numerical group, a tunnel rock stratum inclination numerical group, a tunnel rock strength numerical group, a longitudinal and transverse wave velocity numerical group when propagating in the tunnel, a tunnel convergence numerical group, and a tunnel offset numerical group.

[0084] An embodiment of the present invention provides a targeted numerical analysis method for high-risk tunnel sections, wherein a tunnel gas content numerical value group includes a tunnel carbon monoxide content numerical value group, a tunnel carbon dioxide content numerical value group, a tunnel combustible gas content numerical value group, and a tunnel oxygen content numerical value group.

[0085] The embodiment of the present invention provides a high-risk tunnel targeted numerical analysis method, which pre-processes the values ​​in each tunnel numerical value group to obtain a plurality of pre-processed tunnel numerical value groups, including:

[0086] Outliers are eliminated from the values ​​in each tunnel value group to obtain several optimized tunnel value groups;

[0087] The missing values ​​of each optimized tunnel value group are filled to obtain several preprocessed tunnel value groups.

[0088] The technical principle and technical effect of the above technical solution are: outliers are eliminated from the values ​​in each tunnel value group to obtain several optimized tunnel value groups; missing values ​​are filled in each optimized tunnel value group to obtain several pre-processed tunnel value groups. The technical effect of improving the quality of the values ​​in the tunnel value group is achieved, thereby ensuring the quality of the obtained target values, and then ensuring the accuracy of the target value analysis results.

[0089] The embodiment of the present invention provides a targeted numerical analysis method for high-risk tunnel sections, which removes abnormal values ​​from each tunnel numerical value group to obtain a number of optimized tunnel numerical value groups, including:

[0090] Select any tunnel value group as the tunnel value group to be optimized, sort the values ​​in the tunnel value group to be optimized in the order of value acquisition time from earliest to latest, and obtain a sorted tunnel value group;

[0091] Determine the partitioning step size based on the number of values ​​in the sorted tunnel value group;

[0092] Based on the partition step, the sorted tunnel value group is evenly divided into a number of sub-tunnel value groups;

[0093] Determine any sub-tunnel value group as a target tunnel value group, and determine a subsequent sub-tunnel value group of the target tunnel value group as an associated tunnel value group of the target tunnel value group;

[0094] Calculate a number of Pearson correlation coefficients between each value in the target tunnel value group and the value at the corresponding position in the associated tunnel value group;

[0095] Sort the Pearson coefficients in descending order, and determine the values ​​in the target tunnel value group corresponding to the first N Pearson coefficients as the first abnormal value;

[0096] Determine the median of all values ​​of the target tunnel value group except the first abnormal value;

[0097] Calculate the ratio of each first outlier to the median, and when the ratio is less than a preset ratio threshold, determine the first outlier corresponding to the ratio as the second outlier;

[0098] Calculate the mean of all values ​​in the target tunnel value group except the first abnormal value;

[0099] Calculate the absolute difference between each second outlier and the mean, and use the ratio of the absolute difference to the mean as the deviation coefficient of the second outlier;

[0100] The second outlier value whose deviation coefficient is greater than the preset deviation threshold is used as the final outlier value of the target tunnel value group, and the final outlier value is removed from the target tunnel value group to obtain an optimized target tunnel value group;

[0101] The above operation is performed on all sub-tunnel value groups to obtain the optimized target tunnel value group corresponding to each sub-tunnel value group, and all optimized target tunnel value groups constitute the tunnel value group to be optimized corresponding to the tunnel value group to be optimized;

[0102] The above operation is performed on all classified tunnel value groups to obtain several optimized tunnel value groups.

[0103] In this embodiment, the ratio of the number of values ​​in the sorting tunnel value group to the division step is an integer. For example, when the number of values ​​in the sorting tunnel value group is 30, the division step can be 5, 6, 10 or 15. It should be noted that the division step is greater than or equal to 5.

[0104] In this embodiment, the average value of the target tunnel value group is determined as The average value of the associated tunnel value group is The standard deviation of the target tunnel value group is S 1 , the standard deviation of the numerical value group of the associated tunnel is S 2 , determine the value in the target tunnel value group is x 1 , the value in the associated tunnel value group is y 1 , then x 1 and 1 The Pearson correlation coefficient between them is:

[0105] In this embodiment, N is determined according to the number of values ​​in the target tunnel value group, and is not specifically limited here.

[0106] In this embodiment, several ratios of each value except the first abnormal value in the target tunnel value group to the median are calculated, and the average value of the several ratios is used as the preset ratio threshold.

[0107] In this embodiment, several absolute differences between each value in the target tunnel value group except the first abnormal value and the mean are calculated, several deviation coefficients between each absolute difference and the mean are calculated, and the mean of the several deviation coefficients is used as the preset deviation threshold.

[0108] The technical principle and technical effect of the above technical scheme are: select any tunnel value group as the tunnel value group to be optimized, sort the values ​​in the tunnel value group to be optimized in the order of value acquisition time from early to late, and obtain the sorted tunnel value group; determine the division step size based on the number of values ​​in the sorted tunnel value group; divide the sorted tunnel value group into several sub-tunnel value groups on an even basis based on the division step size; determine the target tunnel value group and the corresponding associated tunnel value group; calculate several Pearson correlation coefficients between each value in the target tunnel value group and the value at the corresponding position in the associated tunnel value group; sort several Pearson coefficients in order from large to small, and determine the values ​​in the target tunnel value group corresponding to the first N Pearson coefficients as the first abnormal value; determine the target tunnel value group except the first abnormal value. The median of all remaining values; determine the second outlier according to the ratio of each first outlier to the median; calculate the mean of all values ​​in the target tunnel value group except the first outlier; calculate the absolute difference between each second outlier and the mean, and use the ratio of the absolute difference to the mean as the deviation coefficient of the second outlier; use the second outlier with a deviation coefficient greater than the preset deviation threshold as the final outlier of the target tunnel value group, remove the final outlier from the target tunnel value group, and obtain the optimized target tunnel value group; perform the above operations on all sub-tunnel value groups to obtain the optimized target tunnel value group corresponding to each sub-tunnel value group, and all optimized target tunnel value groups constitute the tunnel value group to be optimized corresponding to the tunnel value group to be optimized; perform the above operations on all classified tunnel value groups to obtain several optimized tunnel value groups. By removing the outliers in the tunnel value group, the quality of the values ​​in the tunnel values ​​is improved, thereby ensuring the accuracy of the targeted numerical analysis results.

[0109] The embodiment of the present invention provides a targeted numerical analysis method for high-risk tunnel sections, which fills missing values ​​for each optimized tunnel numerical group to obtain a plurality of pre-processed tunnel numerical groups, including:

[0110] Select any optimized tunnel value group as the tunnel value group to be filled, and use the position where the abnormal value is removed in the tunnel value group to be filled as the missing value filling position;

[0111] When the digits before and after the missing value filling position have values, the type of the missing value filling position is determined to be the first type of missing value filling position; otherwise, the type of the missing value filling position is determined to be the second type of missing value filling position;

[0112] When the missing value filling position is the first type of missing value filling position, the mean of the two digits before and after the missing value filling position is used as the filling value of the missing value filling position;

[0113] When the missing value filling position is the second type of missing value filling position, the missing value filling position with a value in the previous adjacent position is selected as the target filling position, and any target filling position is selected as the filling position to be processed;

[0114] Determine the value before the to-be-processed filling position as an auxiliary value, and count all other values ​​in the to-be-filled tunnel value group that are the same as the auxiliary value and their next adjacent values;

[0115] Generate a numerical conversion table based on all other numerical values ​​that are the same as the auxiliary numerical value and their next adjacent numerical values;

[0116] According to the numerical conversion table, calculate the probability that the value that is the same as the auxiliary value is converted into the same adjacent value, and use the probability as the weight of the adjacent value;

[0117] Determine the padding value of the to-be-processed padding position based on all adjacent values ​​and corresponding weights;

[0118] Perform the above operation on all target filling positions to obtain the filling value corresponding to each target filling position;

[0119] Perform the above operation on all the second-category missing value filling positions to obtain the filling value corresponding to each second-category missing value filling position;

[0120] The tunnel value group to be filled that has completed the value filling is used as the pre-processed tunnel value group, and the above operation is performed on all the optimized tunnel value groups to obtain a plurality of pre-processed tunnel value groups.

[0121] In this embodiment, an example is given to illustrate the process of performing missing value filling when the missing value filling bit is the second type of missing value filling bit: Assume that the optimized tunnel value group at this time is a 1 ,a 2 ,_,_,_,a 3 ,...a n , where the position of _ is the missing value filling position. Since the three positions do not have values ​​in the preceding and following digits, the three missing value filling positions are all the second type of missing value filling positions. The missing value filling position with a value in the previous adjacent position is selected as the target filling position, that is, the first missing value filling position is selected as the target filling position, and the auxiliary value corresponding to the target filling position is a 2 , determine the a generated in the optimized tunnel value group 2 The corresponding numerical conversion table is as follows:

[0122] <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 2 ]]> <![CDATA[a 5 ]]> <![CDATA[a 8 ]]> <![CDATA[a 5 ]]> <![CDATA[a 5 ]]> <![CDATA[a 8 ]]> <![CDATA[a 6 ]]> <![CDATA[a 5 ]]> <![CDATA[a 5 ]]> <![CDATA[a 8 ]]> <![CDATA[a 6 ]]>

[0123] According to the above numerical conversion table, a 2 Convert to a 5 The probability of a 2 Convert to a6 The probability of a 2 Convert to a 8 The probability of So a 5 The weight is a 6 The weight is a 8 The weight is Then the padding value corresponding to the target padding position is: Fill the first missing value filling position with the filling value, and then use the second missing value filling position as the target filling position, and follow the above steps to fill until the filling is completed.

[0124] The technical principle and technical effect of the above technical scheme are: select any optimized tunnel value group as the tunnel value group to be filled, and use the position in the tunnel value group to be filled where abnormal values ​​have been eliminated as the missing value filling position; when the digits before and after the missing value filling position have values, determine the type of the missing value filling position as the first type of missing value filling position, otherwise, determine the type of the missing value filling position as the second type of missing value filling position; when the missing value filling position is the first type of missing value filling position, use the numerical mean of the two digits before and after the missing value filling position as the filling value of the missing value filling position; when the missing value filling position is the second type of missing value filling position, select the missing value filling position with a value in the previous adjacent position as the target filling position, and select any target filling position as the filling position to be processed; determine the value of the previous position of the to-be-processed filling position as the auxiliary value, and statistically calculate the value of the previous position of the to-be-processed filling position. All other values ​​in the tunnel value group to be filled that are the same as the auxiliary value and their next adjacent values; generate a value conversion table based on all other values ​​that are the same as the auxiliary value and their next adjacent values; according to the value conversion table, calculate the probability that the value that is the same as the auxiliary value is converted into the same adjacent value, and use the probability as the weight of the adjacent value; determine the filling value of the pending filling position based on all adjacent values ​​and the corresponding weights; perform the above operation on all target filling positions to obtain the filling value corresponding to each target filling position; perform the above operation on all second-category missing value filling positions to obtain the filling value corresponding to each second-category missing value filling position; use the tunnel value group to be filled that has completed the value filling as the preprocessing tunnel value group, perform the above operation on all optimized tunnel value groups, and obtain several preprocessing tunnel value groups. By filling the missing values, the technical effect of ensuring the numerical integrity of the tunnel value group, improving the numerical quality, and thus ensuring the accuracy of the targeted numerical analysis results is achieved.

[0125] Reference Figure 2The embodiment of the present invention provides a high-risk tunnel targeted numerical analysis method, which obtains a targeted numerical value group corresponding to each preprocessing tunnel numerical value group based on the preprocessing tunnel numerical value in each preprocessing tunnel numerical value group and the corresponding numerical value acquisition time, including:

[0126] Step S31, selecting any pre-processed tunnel value group as the to-be-processed value group, and matching each value with a plurality of preset standard values ​​based on the value acquisition time corresponding to the values ​​in the to-be-processed value group to obtain a standard value corresponding to each value;

[0127] Step S32, calculating the absolute difference between each value and the corresponding standard value, and taking the ratio of the absolute difference to the standard value as the target value corresponding to the value;

[0128] Step S33, the target values ​​and target value generation times corresponding to all the values ​​in the to-be-processed value group form a target value group corresponding to the to-be-processed value group; wherein the target value generation time is the value acquisition time of the corresponding value;

[0129] Step S34, performing the above operation on all pre-processing tunnel value groups to obtain a plurality of targeting value groups corresponding to the plurality of pre-processing tunnel value groups.

[0130] In this embodiment, the preset standard value is a predicted value of a certain parameter of the tunnel at a certain moment in time, which can be predicted by a numerical prediction model pre-trained based on sample data.

[0131] The technical principle and technical effect of the above technical solution are as follows: select any pre-processed tunnel value group as the value group to be processed, match each value with several preset standard values ​​based on the value acquisition time corresponding to the value in the value group to be processed, and obtain the standard value corresponding to each value; calculate the absolute difference between each value and the corresponding standard value, and use the ratio of the absolute difference to the standard value as the target value corresponding to the value; the target values ​​and the target value generation time corresponding to all the values ​​in the value group to be processed constitute the target value group corresponding to the value group to be processed; wherein the target value generation time is the value acquisition time of the corresponding value; perform the above operation on all pre-processed tunnel value groups to obtain several target value groups corresponding to several pre-processed tunnel value groups. The technical effect of accurately obtaining the target value and generation time, thereby ensuring the accuracy of the subsequent determined target value analysis results, is achieved.

[0132] Reference Figure 3 The embodiment of the present invention provides a targeted numerical analysis method for a high-risk tunnel section, which analyzes the targeted values ​​in each targeted numerical value group to obtain several targeted numerical analysis results, including:

[0133] Step S41, selecting any target value group as the target value group to be processed, and obtaining a target value change curve graph based on the target values ​​and target value generation time in the target value group to be processed;

[0134] Step S42, generating a standard threshold line in the target value change curve graph, taking points in the target value change curve graph that are higher than the standard threshold line as mutation points, and counting the number of mutation points as the number of mutation points;

[0135] Step S43, calculating the average of the target values ​​corresponding to all mutation points as the mutation average value;

[0136] Step S44, when the number of mutation points is greater than the point threshold and the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the first analysis result;

[0137] Step S45, when the number of mutation points is greater than the point threshold but the average mutation value is less than the average threshold, determining the targeted numerical analysis result as the second analysis result;

[0138] Step S46, when the number of mutation points is less than the point threshold but the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the third analysis result;

[0139] Step S47, when the number of mutation points is less than the point threshold and the average mutation value is less than the average threshold, determining the targeted numerical analysis result as the fourth analysis result.

[0140] In this embodiment, the median value and the average value in the target value group are obtained, and the mean between the median value and the average value is used as the standard threshold.

[0141] In this embodiment, the number of mutation points in each targeted value group is obtained, and the average of all mutation point numbers is used as the point threshold; the average mutation value of each targeted value group is obtained, and the average of all mutation average values ​​is used as the average threshold.

[0142] In this embodiment, the first analysis result is that the number of mutation points is greater than the point threshold and the mutation average value is greater than the average threshold, and the corresponding parameter change state of the targeted numerical value group is in an ultra-high risk change state; the second analysis result is that the number of mutation points is greater than the point threshold but the mutation average value is less than the average threshold, and the corresponding parameter change state of the targeted numerical value group is in a high risk change state; the third analysis result is that the number of mutation points is less than the point threshold but the mutation average value is greater than the average threshold, and the corresponding parameter change state of the targeted numerical value group is in a medium risk change state; the fourth analysis result is that the number of mutation points is less than the point threshold and the mutation average value is less than the average threshold, and the corresponding parameter change state of the targeted numerical value group is in a low risk change state.

[0143] The technical principle and technical effect of the above technical scheme are as follows: select any target value group as the target value group to be processed, and obtain the target value change curve diagram based on the target values ​​in the target value group to be processed and the target value generation time; generate a standard threshold line in the target value change curve diagram, take the points in the target value change curve diagram that are higher than the standard threshold line as mutation points, and count the number of mutation points as the number of mutation points; calculate the average value of the target values ​​corresponding to all mutation points as the mutation average value; when the number of mutation points is greater than the point threshold and the mutation average value is greater than the average threshold, determine the target value analysis result as the first analysis result; when the number of mutation points is greater than the point threshold but the mutation average value is less than the average threshold, determine the target value analysis result as the second analysis result; when the number of mutation points is less than the point threshold but the mutation average value is greater than the average threshold, determine the target value analysis result as the third analysis result; when the number of mutation points is less than the point threshold and the mutation average value is less than the average threshold, determine the target value analysis result as the fourth analysis result. The technical effect of determining the target value analysis result based on the number of mutation points and the mutation average value, thereby ensuring the accuracy of the target value analysis result, is achieved.

[0144] Reference Figure 4 The embodiment of the present invention provides a targeted numerical analysis method for a high-risk tunnel section, and generates a tunnel analysis report for the high-risk tunnel section according to a number of targeted numerical analysis results, including:

[0145] Fill in the basic information of the high-risk section tunnel in the tunnel report template;

[0146] Filling a plurality of targeted numerical analysis results and corresponding targeted numerical variation curves in a preset result display position of a tunnel report template;

[0147] Based on a number of targeted numerical analysis results, corresponding tunnel inspection suggestions are generated, and the tunnel inspection suggestions are filled into the preset suggestion filling positions of the tunnel report template to generate a tunnel analysis report for the high-risk section tunnel.

[0148] In this embodiment, the tunnel report template can be as follows: Figure 4 As shown, the basic information is filled into the basic information filling position in the figure; several targeted numerical analysis results and the corresponding targeted numerical change curve graphs are filled into the preset result display positions in the figure, wherein the number of tunnel numerical groups is determined to be M, and the targeted numerical analysis results and the corresponding targeted numerical change curve graphs of each tunnel numerical group have corresponding positions, such as the result 1 display position, the result 2 display position, until the result M display position in the figure.

[0149] In this embodiment, tunnel inspection suggestions can be prioritized according to the targeted analysis results corresponding to the tunnel value groups, for example: the first analysis result priority > the second analysis result priority > the third analysis result priority > the fourth analysis result priority. Assuming that the targeted numerical analysis results of the tunnel groundwater level value group, the tunnel groundwater pressure value group, the tunnel temperature value group and the tunnel humidity value group are the first analysis results, and the targeted numerical analysis results of the tunnel gas content value group, the tunnel rock layer inclination value group and the tunnel rock strength value group are the second analysis results; the targeted numerical analysis results of the wave velocity value group and the tunnel convergence value group when the longitudinal and transverse waves propagate in the tunnel are the third analysis results; and the targeted numerical analysis results of the tunnel offset value group are the fourth analysis results, then the current tunnel inspection suggestions are: first check the tunnel groundwater level, tunnel groundwater pressure, tunnel temperature and tunnel humidity, then check the tunnel gas content, tunnel rock layer inclination and tunnel rock strength, then check the state of the longitudinal and transverse waves propagating in the tunnel and the tunnel convergence state, and finally check the tunnel offset state.

[0150] In this embodiment, the tunnel troubleshooting suggestion can be filled in as follows: Figure 4 The presets shown suggest fill positions.

[0151] The technical principle and technical effect of the above technical solution are: fill in the basic information of the high-risk tunnel section in the tunnel report template; fill in a number of targeted numerical analysis results and the corresponding targeted numerical change curves in the preset result display position of the tunnel report template; generate corresponding tunnel inspection suggestions based on a number of targeted numerical analysis results, fill the tunnel inspection suggestions in the preset suggestion filling position of the tunnel report template, and generate a tunnel analysis report for the high-risk tunnel section. By generating a tunnel analysis report, the tunnel status is intuitively displayed, thereby reducing the difficulty of tunnel inspection and saving resources.

[0152] The embodiment of the present invention provides a targeted numerical analysis method for a high-risk tunnel section. After generating a tunnel analysis report for the high-risk tunnel section according to a plurality of targeted numerical analysis results, the method further includes:

[0153] Based on the tunnel inspection opinions in the tunnel analysis report, the tunnel is inspected to obtain the tunnel inspection results;

[0154] Based on the tunnel inspection results, the tunnel is maintained.

[0155] In this embodiment, the tunnel inspection result may be that the tunnel groundwater level is too high, the tunnel temperature is too high, etc.

[0156] In this embodiment, tunnel maintenance may include draining water and pumping water, taking wall protection and anti-seepage measures, building pumping wells or caissons, adding ventilation equipment and ventilation holes inside the tunnel, etc.

[0157] The technical principle and technical effect of the above technical solution are: based on the tunnel inspection opinions in the tunnel analysis report, the tunnel is inspected to obtain the tunnel inspection results; based on the tunnel inspection results, the tunnel is maintained. The technical effect of ensuring the safety and stability of high-risk tunnels is achieved.

[0158] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A high-risk tunnel targeted numerical analysis method, characterized in that: include: Acquire multiple tunnel value groups of high-risk tunnel sections; Preprocessing the values ​​in each tunnel value group to obtain a plurality of preprocessed tunnel value groups; Based on the preprocessing tunnel values ​​in each preprocessing tunnel value group and the corresponding value acquisition time, a targeting value group corresponding to each preprocessing tunnel value group is obtained; Analyze the target values ​​in each target value group to obtain multiple target value analysis results; Generating a tunnel analysis report of the high-risk section tunnel according to the multiple targeted numerical analysis results; The multiple tunnel value groups include a tunnel groundwater level value group, a tunnel groundwater pressure value group, a tunnel temperature value group, a tunnel humidity value group, a tunnel gas content value group, a tunnel rock layer inclination value group, a tunnel rock strength value group, a longitudinal and transverse wave velocity value group when propagating in the tunnel, a tunnel convergence value group and a tunnel offset value group; The tunnel gas content value group includes a tunnel carbon monoxide content value group, a tunnel carbon dioxide content value group, a tunnel combustible gas content value group and a tunnel oxygen content value group; Preprocessing the values ​​in each tunnel value group to obtain a plurality of preprocessed tunnel value groups, including: removing abnormal values ​​from the values ​​in each tunnel value group to obtain a plurality of optimized tunnel value groups; Filling missing values ​​for each optimized tunnel value group to obtain multiple preprocessed tunnel value groups; Outliers are eliminated from the values ​​in each tunnel value group to obtain multiple optimized tunnel value groups, including: Select any tunnel value group as the tunnel value group to be optimized, sort the values ​​in the tunnel value group to be optimized in the order of value acquisition time from earliest to latest, and obtain a sorted tunnel value group; Determine the partitioning step size based on the number of values ​​in the sorted tunnel value group; Based on the partition step, the sorted tunnel value group is evenly divided into a plurality of sub-tunnel value groups; Determine any sub-tunnel value group as a target tunnel value group, and determine a subsequent sub-tunnel value group of the target tunnel value group as an associated tunnel value group of the target tunnel value group; Calculating multiple Pearson correlation coefficients between each value in the target tunnel value group and the value at the corresponding position in the associated tunnel value group; Sort multiple Pearson coefficients in descending order to determine the top The value in the target tunnel value group corresponding to the Pearson coefficient is the first abnormal value; Determine the median of all values ​​of the target tunnel value group except the first abnormal value; Calculating a ratio of each first outlier value to the median, and when the ratio is less than a preset ratio threshold, determining the first outlier value corresponding to the ratio as a second outlier value; Calculate the mean of all values ​​in the target tunnel value group except the first abnormal value; Calculate the absolute difference between each second outlier and the mean, and use the ratio of the absolute difference to the mean as the deviation coefficient of the second outlier; Taking the second abnormal value whose deviation coefficient is greater than the preset deviation threshold as the final abnormal value of the target tunnel value group, and removing the final abnormal value from the target tunnel value group to obtain an optimized target tunnel value group; The above operation is performed on all sub-tunnel value groups to obtain the optimized target tunnel value group corresponding to each sub-tunnel value group, and all optimized target tunnel value groups constitute the tunnel value group to be optimized corresponding to the tunnel value group to be optimized; The above operation is performed on all classified tunnel value groups to obtain multiple optimized tunnel value groups; Fill missing values ​​for each optimized tunnel value group to obtain multiple pre-processed tunnel value groups, including: Select any optimized tunnel value group as the tunnel value group to be filled, and use the position where the abnormal value is removed in the tunnel value group to be filled as the missing value filling position; When the digits before and after the missing value filling position have values, the type of the missing value filling position is determined to be the first type of missing value filling position; otherwise, the type of the missing value filling position is determined to be the second type of missing value filling position; When the missing value filling position is a first type of missing value filling position, the numerical mean of the two digits before and after the missing value filling position is used as the filling value of the missing value filling position; When the missing value filling position is the second type of missing value filling position, the missing value filling position with a value in the previous adjacent position is selected as the target filling position, and any target filling position is selected as the filling position to be processed; Determine the value before the to-be-processed filling position as an auxiliary value, and count all other values ​​in the to-be-filled tunnel value group that are the same as the auxiliary value and their next adjacent values; Generate a numerical conversion table based on all other numerical values ​​that are the same as the auxiliary numerical value and their next adjacent numerical values; According to the numerical conversion table, the probability of a numerical value identical to the auxiliary numerical value being converted into the same adjacent numerical value is calculated, and the probability is used as the weight of the adjacent numerical value; Determine the padding value of the to-be-processed padding position based on all adjacent values ​​and corresponding weights; Perform the above operation on all target filling positions to obtain the filling value corresponding to each target filling position; Perform the above operation on all the second-category missing value filling positions to obtain the filling value corresponding to each second-category missing value filling position; The tunnel value group to be filled that has completed the value filling is used as the pre-processed tunnel value group, and the above operation is performed on all the optimized tunnel value groups to obtain multiple pre-processed tunnel value groups; Based on the preprocessing tunnel values ​​in each preprocessing tunnel value group and the corresponding value acquisition time, a target value group corresponding to each preprocessing tunnel value group is obtained, including: Select any pre-processed tunnel value group as the to-be-processed value group, and match each value with a plurality of preset standard values ​​based on the value acquisition time corresponding to the values ​​in the to-be-processed value group to obtain the standard value corresponding to each value; Calculate the absolute difference between each value and the corresponding standard value, and use the ratio of the absolute difference to the standard value as the target value corresponding to the value; The target values ​​and target value generation times corresponding to all values ​​in the to-be-processed value group constitute a target value group corresponding to the to-be-processed value group; wherein the target value generation time is the value acquisition time of the corresponding value; Perform the above operation on all pre-processing tunnel value groups to obtain multiple target value groups corresponding to multiple pre-processing tunnel value groups. Select any target value group as the target value group to be processed, and obtain a target value change curve graph based on the target values ​​and target value generation time in the target value group to be processed; Generating a standard threshold line in the target value change curve graph, taking points in the target value change curve graph that are higher than the standard threshold line as mutation points, and counting the number of mutation points as the number of mutation points; The average of the targeted values ​​corresponding to all mutation points was calculated as the mutation average.

2. The high-risk tunnel targeted numerical analysis method according to claim 1, characterized in that: The target values ​​in each target value group are analyzed to obtain multiple target value analysis results, including: When the number of mutation points is greater than the point threshold and the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the first analysis result; When the number of mutation points is greater than the point threshold but the average mutation value is less than the average threshold, determining the targeted numerical analysis result as the second analysis result; When the number of mutation points is less than the point threshold but the average mutation value is greater than the average threshold, determining the targeted numerical analysis result as the third analysis result; When the number of mutation points is less than the point threshold and the average mutation value is less than the average threshold, the targeted numerical analysis result is determined to be the fourth analysis result.

3. The high-risk tunnel targeted numerical analysis method according to claim 1, characterized in that: Generating a tunnel analysis report of the high-risk section tunnel according to the multiple targeted numerical analysis results, including: Fill in the information of high-risk tunnel sections in the tunnel report template; Filling a plurality of targeted numerical analysis results and corresponding targeted numerical change curves in a preset result display position of a tunnel report template; Based on multiple targeted numerical analysis results, corresponding tunnel inspection suggestions are generated, and the tunnel inspection suggestions are filled into the preset suggestion filling positions of the tunnel report template to generate a tunnel analysis report for the high-risk section tunnel.

4. The high-risk tunnel targeted numerical analysis method according to claim 1, characterized in that: After generating a tunnel analysis report of the high-risk section tunnel according to the multiple targeted numerical analysis results, the method further includes: Based on the tunnel inspection opinions in the tunnel analysis report, the tunnel is inspected to obtain tunnel inspection results; Based on the tunnel inspection results, the tunnel is maintained.

Images