A Dynamic Zoning Method for the Permeability of Water-Rich Tunnels Based on Multi-Source Data

Abstract

A dynamic permeability zoning method for water-rich tunnels based on multi-source data includes the following steps: acquiring permeability coefficient, rock compressive strength, integrity coefficient, RQD, and P-wave velocity data of the tunnel excavation area; constructing a multi-source feature matrix through spatiotemporal alignment and standardization; dynamically calculating the weights of each indicator based on the entropy weight method, and adaptively adjusting the parameter contribution based on geological type; generating standard cloud parameters using a cloud model, calculating the membership degree of real-time data to each permeability level through a forward cloud generator, and comprehensively determining the final grading result; mapping the grading result to a 3D BIM model to generate a dynamic permeability cloud map and triggering a real-time early warning mechanism. This invention overcomes the static nature and experience dependence of traditional grading methods through multi-source data fusion and dynamic weight allocation, significantly improving the accuracy of permeability assessment in complex strata; the cloud model quantifies the ambiguity of grade boundaries, avoiding the risk of misjudgment in critical intervals; the deep integration of advanced geological forecast data enables spatial prediction of permeability in unexcavated areas; combined with minute-level early warning and intelligent recommendation of grouting parameters, it effectively reduces the probability of sudden water inrush accidents, providing an intelligent solution for safe construction and resource optimization of water-rich tunnels.

Description

Technical Field

[0001] This invention belongs to the field of tunnel boring technology, and specifically relates to a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data. Background Technology

[0002] Current permeability grading studies largely rely on static indicators and empirical models, such as classifying grades based on permeability coefficient thresholds or constructing linear weighting systems by combining rock strength and integrity coefficients. However, existing methods are ill-suited to dynamic construction scenarios: on the one hand, traditional models depend on static data such as borehole water pressure tests, with update cycles lasting several hours, failing to capture real-time changes such as fracture propagation and seepage path migration during tunneling; on the other hand, advanced geological prediction data (such as TSP wave velocity and ground-penetrating radar images) and borehole parameters (such as RQD and fracture density) have not been deeply integrated, forming data silos and resulting in insufficient spatial resolution of grading results. Furthermore, the critical range of permeability coefficients (such as 10...) is often overlooked. -6 ~ 10 -7 The fuzzy boundary problem of the m / s level has remained unsolved for a long time, and traditional fuzzy mathematics methods have errors as high as 25% to 40%, which seriously affects the reliability of engineering decisions.

[0003] Therefore, it is urgent to break through the existing hierarchical framework and develop a dynamic permeability zoning method driven by multi-source data. This method constructs a dynamic permeability matrix by integrating advanced forecasting, real-time monitoring, and geological borehole data, and achieves adaptive weight adjustment and level boundary quantification based on an entropy weight-cloud model coupling algorithm. This method can realize hierarchical updates and disaster early warning, providing precise decision support for the safe construction of water-rich tunnels and promoting the industry's transformation from "passive emergency response" to "proactive prevention and control." Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data. By integrating advanced geological forecasting, real-time monitoring and borehole parameters to construct a dynamic classification model, the permeability zoning can be updated and accurately warned within minutes.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A dynamic zoning method for the permeability of water-rich tunnels based on multi-source data, comprising the following steps:

[0007] S1, acquire permeability coefficient, rock compressive strength, integrity coefficient, RQD, P-wave velocity and advanced geological prediction data of the tunnel excavation area, and standardize the data; the RQD is a rock quality index.

[0008] S2, based on the entropy weight method, dynamically calculates the weights of each indicator, constructs a decision matrix including permeability coefficient, rock compressive strength, integrity coefficient, RQD and P-wave velocity, calculates the information entropy of each indicator and dynamically assigns weights;

[0009] S3 divides the penetration levels into I to V and sets standard cloud parameters, including expected value, entropy, and hyperentropy; the membership degree of real-time data to each level is calculated through a positive cloud generator, and the final level is determined by combining dynamic weights.

[0010] S4 maps the grading results to the geological model to generate a three-dimensional tunnel permeability zoning cloud map, enabling the prediction of permeability trends in the tunneling direction; when the permeability level of a region is greater than or equal to level IV or the level difference between adjacent regions is greater than or equal to level 2, an early warning command is triggered to the terminal equipment at the tunnel face.

[0011] Preferably, in step S1, the advanced geological prediction data needs to be spatiotemporally aligned with the borehole parameters, and the discrete data is converted into a continuous permeability field using the Kriging interpolation method, with the spatial resolution controlled within 0.5m.

[0012] Preferably, in step S2, the weight calculation formula is:

[0013] ;

[0014] in, w j For the first j The weight of each indicator, E j For the first j Information entropy of each indicator m For the total number of indicators, p ij For the first i The sample at the th j Normalized values ​​for each indicator x ij For the first i The first sample j The actual measured value of each indicator n The total number of samples.

[0015] Preferably, in step S3, the standard cloud parameters are optimized using a particle swarm optimization algorithm based on historical engineering data to adaptively correct the level boundaries;

[0016] The membership degree calculation formula is as follows:

[0017] ;

[0018] in, For the first j The data point pair iMembership degree of each penetration level, x j For the first j The actual measured value of each data point. E xi For the first i Standard cloud expectations for each level, E ni For the first i The entropy of a standard cloud at each level, H ei For the first i The level of hyperentropy, This is a randomized implementation of entropy.

[0019] Preferably, in step S4, the response time for pushing the early warning command to the terminal equipment at the working face shall not exceed 3 minutes, and disaster prevention and mitigation measures shall be implemented in a timely manner after the command is pushed.

[0020] Preferably, in step S4, the disaster prevention and mitigation measures include grouting to reinforce the surrounding rock, precipitation, real-time monitoring, etc. In actual projects, appropriate measures should be selected based on engineering geological conditions and disaster levels.

[0021] Preferably, in S1, the permeability coefficient is obtained through borehole water pressure test or transient hydraulic test; the rock mechanical parameters are determined through borehole core laboratory test; and the longitudinal wave velocity is extracted by TSP system and ground-penetrating radar.

[0022] Preferably, the geological model is a three-dimensional BIM model.

[0023] Preferably, after receiving the early warning command, the terminal equipment at the working face responds through methods such as sound and light prompts, display of grading results, and recommendations for grouting parameters.

[0024] A dynamic zoning system for the permeability of water-rich tunnels based on multi-source data, employing the aforementioned dynamic zoning method for the permeability of water-rich tunnels based on multi-source data, includes:

[0025] Multi-source geological data acquisition and standardization processing module: used to acquire permeability coefficient, rock compressive strength, integrity coefficient, RQD, P-wave velocity and advanced geological prediction data of the tunnel excavation area, and to standardize the data; the RQD is a rock quality index.

[0026] Dynamic weight intelligent calculation module: used to dynamically calculate the weight of each indicator based on the entropy weight method, construct a decision matrix including permeability coefficient, rock compressive strength, integrity coefficient, RQD and P-wave velocity, calculate the information entropy of each indicator and dynamically assign weights;

[0027] The penetration level cloud model determination module is used to classify penetration levels from I to V and set standard cloud parameters, including expected value, entropy, and hyperentropy; it calculates the membership degree of real-time data to each level through a positive cloud generator and determines the final level by combining dynamic weights.

[0028] The 3D permeability zoning and early warning response module is used to map the grading results to the geological model, generate a 3D tunnel permeability zoning cloud map, and realize the prediction of permeability trend in the tunneling direction. When the permeability level of the area is greater than or equal to level IV or the level difference between adjacent areas is greater than or equal to level 2, an early warning command is triggered to the terminal equipment at the tunnel face.

[0029] A computer device, comprising:

[0030] One or more processors, wherein one or more executable programs are stored on the processors;

[0031] When one or more executable programs are executed by one or more processors, they are used to implement the aforementioned method for dynamic zoning of water-rich tunnel permeability based on multi-source data.

[0032] A storage medium storing an executable program, which, when executed, is used to implement the aforementioned dynamic zoning method for the permeability of water-rich tunnels based on multi-source data.

[0033] The present invention can achieve the following beneficial effects:

[0034] This invention overcomes the static and experience-dependent nature of traditional permeability grading methods by integrating multi-source data and dynamic weight allocation, achieving accurate grading that adapts to geological conditions. The dynamic weighting mechanism based on entropy weighting can autonomously identify dominant parameters in different sections, such as enhancing the influence of fracture development in fault zones and emphasizing rock strength characteristics in intact rock masses, significantly improving the geological adaptability of the grading model. The cloud model, through fuzzy boundary quantization technology, effectively solves the problem of fuzzy grading in critical intervals, avoiding engineering risks caused by misjudgments. Deep integration with advanced geological prediction systems constructs a three-dimensional dynamic permeability cloud map, enabling spatial prediction and risk visualization of permeability characteristics in unexcavated areas, providing forward-looking decision support for grouting reinforcement and support design. The real-time early warning mechanism and intelligent linkage with construction parameters significantly shorten the risk response cycle, reduce the probability of sudden water inrush accidents, and optimize resource allocation efficiency, promoting the upgrading of water-rich tunnel engineering towards intelligent and refined management. Attached Figure Description

[0035] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0036] Figure 1 This is a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data;

[0037] Figure 2This is a schematic diagram of data acquisition established by the present invention. Detailed Implementation

[0038] Preferred solutions include Figures 1 to 2 As shown, a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data is proposed.

[0039] S1: Obtain permeability coefficient, rock compressive strength, integrity coefficient, RQD (rock quality index), longitudinal wave velocity, and advanced geological prediction data of the tunnel excavation area, and perform standardized processing on the data;

[0040] In this embodiment, the permeability coefficient was obtained through borehole water pressure testing, rock mechanics parameters were determined through in-situ tests using core samples obtained from an XY-2 drilling rig, geophysical data was collected using a TSP303 system to capture wave velocity anomaly data, and fracture density within a 30m range ahead of the drilling face was extracted using a MALA ProEx ground-penetrating radar. The TSP and ground-penetrating radar data were integrated with the borehole coordinate system, and a spatially continuous permeability characteristic field was generated using Kriging interpolation. A schematic diagram of the data acquisition is shown below. Figure 2 As shown.

[0041] S2, based on the entropy weight method, dynamically calculates the weights of each indicator, constructs a decision matrix including permeability coefficient, rock compressive strength, integrity coefficient, RQD and P-wave velocity, calculates the information entropy of each indicator and dynamically assigns weights;

[0042] In this embodiment, a standardized decision matrix is ​​constructed. ,in n For the number of measurement points, x ij For the first i The measurement point corresponds to the first... j Each indicator is used to calculate its entropy value and assign dynamic weights.

[0043] S3 divides the penetration levels into I to V and sets standard cloud parameters, including expected value, entropy, and hyperentropy; the membership degree of real-time data to each level is calculated through a positive cloud generator, and the final level is determined by combining dynamic weights.

[0044] In this embodiment, the standard cloud parameters are shown in Table 1. The membership calculation process is as follows: generate random entropy, calculate membership, select the level corresponding to the maximum weighted membership value, and use the entropy weight method as the output result for the weight.

[0045] Table 1 Standard Cloud Parameters

[0046]

[0047] S4 maps the grading results to the geological model to generate a three-dimensional tunnel permeability zoning cloud map, enabling the prediction of permeability trends in the tunneling direction; when the permeability level of a region is greater than or equal to level IV or the level difference between adjacent regions is greater than or equal to level 2, an early warning command is triggered to the terminal equipment at the tunnel face.

[0048] In this embodiment, a permeability zoning module is developed based on the Revit BIM platform. The grading results are imported to generate a heat map. The visualization rules are as follows: red indicates grades IV to V, requiring immediate disaster treatment; yellow indicates grade III, with local reinforcement recommended; green indicates grades I to II, allowing normal tunneling. When a region's grade is ≥ IV or the grade difference between adjacent regions is ≥ 2 grades, the system automatically pushes an early warning to the tunnel face terminal. The tunnel face terminal pre-sets grouting treatment plans: for grade III, grouting pressure 1.2~1.8 MPa, water-cement ratio 1:1; for grade IV: pressure 2.0~2.5 MPa, water-cement ratio 0.8:1; for grade V: pressure ≥ 3.0 MPa, using two-component grout.

[0049] Furthermore, in step S1, the permeability coefficient is obtained and dimensionlessly standardized through borehole water pressure test or transient hydraulic test, the rock mechanics parameters are determined through borehole core laboratory test, and the geophysical data is analyzed by TSP system to analyze wave velocity profile and extracted by ground-penetrating radar to extract fracture density.

[0050] The advanced geological prediction data needs to be spatiotemporally aligned with the borehole parameters. The discrete data is converted into a continuous permeability field using the Kriging interpolation method, and the spatial resolution is controlled within 0.5m.

[0051] Furthermore, in step S2, the weights should be adjusted according to the geological type. For example, the weight of the integrity coefficient in fault zone areas should be increased to above 0.4, while the weight of the rock compressive strength should be emphasized in intact rock mass areas. The weight calculation formula is as follows:

[0052] ;

[0053] in, w j For the first j The weight of each indicator, E j For the first j Information entropy of each indicator m For the total number of indicators, p ij For the first i The sample at the th j Normalized values ​​for each indicator x ij For the first i The first sample j The actual measured value of each indicator n The total number of samples.

[0054] Furthermore, in step S3, the standard cloud parameters should be optimized using a particle swarm optimization algorithm based on historical engineering data to adaptively correct the level boundaries.

[0055] The membership degree calculation formula is as follows:

[0056] ;

[0057] in, For the first j The data point pair i Membership degree of each penetration level, x j For the first j The actual measured value of each data point. E xi For the first i Standard cloud expectations for each level, E ni For the first i The entropy of a standard cloud at each level, H ei For the first i The level of hyperentropy, This is a randomized implementation of entropy.

[0058] Furthermore, in step S4, the response time for pushing the early warning command to the terminal equipment at the working face shall not exceed 3 minutes, and disaster prevention and mitigation measures shall be implemented in a timely manner after the command is pushed.

[0059] The disaster prevention and mitigation measures include grouting to reinforce the surrounding rock, precipitation, and real-time monitoring. In actual projects, appropriate measures should be selected based on engineering geological conditions and disaster levels.

[0060] By integrating multi-source data and dynamically assigning weights, this method overcomes the static and experience-dependent limitations of traditional permeability grading methods, achieving accurate grading that adapts to geological conditions. The dynamic weighting mechanism based on entropy weighting can autonomously identify dominant parameters in different sections, such as enhancing the influence of fracture development in fault zones and emphasizing rock strength characteristics in intact rock masses, significantly improving the geological adaptability of the grading model. The cloud model, through fuzzy boundary quantization technology, effectively solves the problem of fuzzy grading in critical intervals, avoiding engineering risks caused by misjudgments. Deep integration with advanced geological prediction systems constructs a three-dimensional dynamic permeability cloud map, enabling spatial prediction and risk visualization of permeability characteristics in unexcavated areas, providing forward-looking decision support for grouting reinforcement and support design. The real-time early warning mechanism and intelligent linkage with construction parameters significantly shorten the risk response cycle, reduce the probability of sudden water inrush accidents, and optimize resource allocation efficiency, promoting the upgrading of water-rich tunnel engineering towards intelligent and refined management.

[0061] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be defined as the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A dynamic zoning method for the permeability of water-rich tunnels based on multi-source data, characterized in that... Includes the following steps: S1, acquire permeability coefficient, rock compressive strength, integrity coefficient, RQD, P-wave velocity and advanced geological prediction data of the tunnel excavation area, and standardize the data; the RQD is a rock quality index. S2, based on the entropy weight method, dynamically calculates the weights of each indicator, constructs a decision matrix including permeability coefficient, rock compressive strength, integrity coefficient, RQD and P-wave velocity, calculates the information entropy of each indicator and dynamically assigns weights; S3 classifies permeability levels from I to V and sets standard cloud parameters, including expected value, entropy, and hyperentropy; The membership degree of real-time data to each level is calculated by a positive cloud generator, and the final level is determined by combining dynamic weights. S4 maps the grading results to the geological model to generate a three-dimensional tunnel permeability zoning cloud map, enabling the prediction of permeability trends in the tunneling direction; When the area's permeability level is greater than or equal to Level IV or the difference in level between adjacent areas is greater than or equal to Level 2, an early warning command is triggered to the working face terminal equipment. In step S2, the weight calculation formula is: ； in, w j For the first j The weight of each indicator, E j For the first j Information entropy of each indicator p ij For the first i The sample at the th j Normalized values ​​for each indicator n The total number of samples; In step S3, the standard cloud parameters are optimized using a particle swarm optimization algorithm based on historical engineering data to adaptively correct the level boundaries; The membership degree calculation formula is as follows: ； in, For the first j The data point pair i Membership degree of each penetration level, x j For the first j The actual measured value of each data point. E xi For the first i Standard cloud expectations for each level, E ni For the first i The entropy of a standard cloud at each level, H ei For the first i The level of hyperentropy, This is a randomized implementation of entropy.

2. The method for dynamic zoning of permeability of water-rich tunnels based on multi-source data according to claim 1, characterized in that: In step S1, the advanced geological prediction data needs to be spatiotemporally aligned with the borehole parameters. The discrete data is converted into a continuous permeability field using the Kriging interpolation method, and the spatial resolution is controlled within 0.5m.

3. The dynamic zoning method for permeability of water-rich tunnels based on multi-source data according to claim 1, characterized in that: In step S4, the response time for pushing the early warning command to the terminal equipment at the working face shall not exceed 3 minutes, and disaster prevention and mitigation measures shall be implemented in a timely manner after the command is pushed.

4. The dynamic zoning method for permeability of water-rich tunnels based on multi-source data according to claim 3, characterized in that: In step S4, the disaster prevention and mitigation measures include grouting to reinforce the surrounding rock, precipitation, and real-time monitoring. In actual engineering projects, appropriate measures should be selected based on engineering geological conditions and disaster levels.

5. The dynamic zoning method for permeability of water-rich tunnels based on multi-source data according to claim 1, characterized in that: In S1, the permeability coefficient is obtained through borehole pressure water test or transient hydraulic test; rock mechanical parameters are determined through borehole core laboratory test, and P-wave velocity is extracted through TSP system and ground-penetrating radar analysis.

6. The method for dynamic zoning of permeability of water-rich tunnels based on multi-source data according to claim 1, characterized in that: The geological model is a 3D BIM model.

7. The method for dynamic zoning of permeability of water-rich tunnels based on multi-source data according to claim 1, characterized in that: After receiving the early warning command, the terminal equipment at the working face responds through audible and visual prompts, display of grading results, and recommendations for grouting parameters.

8. A dynamic zoning system for the permeability of water-rich tunnels based on multi-source data, characterized in that: The system employs a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data, as described in any one of claims 1-7. The system comprises: Multi-source geological data acquisition and standardization processing module: used to acquire permeability coefficient, rock compressive strength, integrity coefficient, RQD, P-wave velocity and advanced geological prediction data of the tunnel excavation area, and to standardize the data; the RQD is a rock quality index. Dynamic weight intelligent calculation module: used to dynamically calculate the weight of each indicator based on the entropy weight method, construct a decision matrix including permeability coefficient, rock compressive strength, integrity coefficient, RQD and P-wave velocity, calculate the information entropy of each indicator and dynamically assign weights; The penetration level cloud model determination module is used to classify penetration levels from I to V and set standard cloud parameters, including expected value, entropy, and hyperentropy; it calculates the membership degree of real-time data to each level through a positive cloud generator and determines the final level by combining dynamic weights. The 3D permeability zoning and early warning response module is used to map the grading results to the geological model, generate a 3D tunnel permeability zoning cloud map, and realize the prediction of permeability trend in the tunneling direction. When the permeability level of the area is greater than or equal to level IV or the level difference between adjacent areas is greater than or equal to level 2, an early warning command is triggered to the terminal equipment at the tunnel face.

9. A computer device, characterized in that: include: One or more processors, wherein one or more executable programs are stored on the processors; When one or more executable programs are executed by one or more processors, they are used to implement a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data as described in any one of claims 1-7.

10. A storage medium, characterized in that: It stores an executable program, which, when executed, is used to implement a dynamic zoning method for the permeability of water-rich tunnels based on multi-source data, as described in any one of claims 1-7.

Images