A tunnel heat hazard detection method, system, medium and device

Abstract

This application provides a method, system, medium, and equipment for detecting thermal hazards in tunnels, relating to the field of tunnel safety detection. The method includes: acquiring geological feature data of the target tunnel; identifying a group of thermal hazard parameters with a correlation greater than a set threshold; setting a quantitative scoring range based on this; using the analytic hierarchy process (AHP) to assign weights and calculate hazard scores; then, combining relevant data to establish a dataset and training a lightweight gradient boosting model; finally, inputting data from an unexcavated tunnel section into the model to output the thermal hazard level. This application, by acquiring geological feature data, can accurately understand the environment, screen key parameters to eliminate redundancy, quantify parameters, and rationally allocate weights to ensure objective and accurate scores. It enables tunnel thermal hazard detection and early warning, providing guidance for prevention and control, reducing impact, and is of great significance to tunnel engineering.

Description

Technical Field

[0001] This application relates to the field of tunnel safety inspection, and in particular to a method, system, medium and equipment for detecting thermal hazards in tunnels. Background Technology

[0002] The heat hazards in tunnels severely impact tunnel construction. For example, high temperature and humidity can easily lead to heat stress reactions such as heatstroke and dehydration among construction workers. Sustained high temperatures accelerate the wear and tear and failure rate of construction machinery and equipment. Furthermore, high temperatures can alter the mechanical properties of the surrounding rock and adversely affect the setting and hardening process of the lining concrete.

[0003] Currently, the prediction of tunnel construction risks relies heavily on existing construction data and expert experience, which makes it difficult to guarantee the accuracy of tunnel thermal hazard prediction and poses serious construction safety risks. Summary of the Invention

[0004] The purpose of this application is to provide a method, system, computer-readable storage medium, and electronic device for detecting thermal hazards in tunnels, which can realize early warning of thermal hazards in tunnels.

[0005] To address the aforementioned technical problems, this application provides a method for detecting thermal hazards in tunnels, the specific technical solution of which is as follows:

[0006] Obtain geological feature data of the target tunnel;

[0007] Identify a group of thermal hazard parameters whose correlation with the geological feature data is greater than a set threshold.

[0008] The range of values ​​for the thermal hazard quantitative score is set based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group.

[0009] The weight values ​​corresponding to each of the heat hazard hazard parameters are set using the analytic hierarchy process (AHP). Based on the weight values ​​and the range of values ​​for the heat hazard quantitative scoring, the heat hazard hazard score is calculated, and the heat hazard hazard level corresponding to the heat hazard hazard score is determined.

[0010] A dataset is established based on the geological feature data, the thermal hazard risk level, and the extreme case data, and a lightweight gradient boosting model is trained based on the dataset.

[0011] The geological feature data of the current unexcavated tunnel segment is input into the lightweight gradient enhancement model, and the tunnel thermal hazard level of the current unexcavated tunnel segment is output.

[0012] Optionally, after the current tunnel segment is completed, the following steps are also included:

[0013] Obtain the actual thermal hazard parameters of the current tunnel section;

[0014] The Shapley values ​​corresponding to each of the actual thermal hazard parameters are calculated using a lightweight gradient boosting model, and the optimization weights are calculated based on the Shapley values.

[0015] The optimized weights are used to replace the weight values, and the updated heat hazard level is determined.

[0016] Based on the updated thermal hazard risk level, the training set in the dataset is reconstructed, and a real-time tunnel thermal hazard detection model is trained.

[0017] The real-time tunnel heat hazard detection model is used to predict the tunnel heat hazard level of the next tunnel section.

[0018] Optionally, before setting the range of values ​​for the thermal hazard quantification score based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group, the method further includes:

[0019] The primary thermal hazard parameters and their corresponding secondary thermal hazard parameters are determined. The primary thermal hazard parameters include hydrothermal activity characteristics, rock mass heat-enrichment characteristics, geological structural conditions, and tunnel engineering conditions. The secondary thermal hazard parameters corresponding to the hydrothermal activity characteristics include the maximum temperature and maximum flow rate of the hot spring. The secondary thermal hazard parameters for the rock mass heat-enrichment characteristics include the rock mass thermal conductivity, rock mass specific heat capacity, geothermal gradient, and predicted geothermal temperature. The secondary thermal hazard parameters for the geological structural conditions include the width of the fault zone and the fissure opening. The secondary thermal hazard parameters for the tunnel engineering conditions include the tunnel burial depth, the distance between the tunnel and the fault zone, the distance between the tunnel and the hot spring outlet, and the distance between the tunnel and surface water.

[0020] Optionally, the weight values ​​corresponding to each of the heat hazard parameters are set using the analytic hierarchy process (AHP), and a heat hazard score is calculated based on the weight values ​​and the range of values ​​for the heat hazard quantitative score. The heat hazard level corresponding to the heat hazard score is then determined, including:

[0021] The weight values ​​corresponding to each of the aforementioned heat hazard parameters were set using the analytic hierarchy process.

[0022] The weighted values ​​are input into the heat hazard score calculation formula to calculate the heat hazard risk score;

[0023] The formula for calculating the heat damage score is as follows: S T Score for heat hazard risk; w i S represents the weight value of the i-th indicator; i Let be the quantitative grading score of the i-th indicator.

[0024] Optionally, establishing a dataset based on the geological feature data, the thermal hazard risk level, and extreme case data, and training a lightweight gradient boosting model based on the dataset includes:

[0025] The thermal hazard level of the unexcavated tunnel section was assessed based on the geological feature data.

[0026] Collect thermal hazard risk assessment indicators for the excavated tunnel sections and calculate the thermal hazard risk score;

[0027] Constructing extreme case data based on prior knowledge data:

[0028] The heat hazard risk level, the heat hazard risk score, and the extreme case data are integrated into a dataset, which is then divided into a training set and a test set. The model parameters of the lightweight gradient boosting model are adjusted through cross-validation and Bayesian optimization methods to train the lightweight gradient boosting model.

[0029] Optionally, calculating the optimization weights based on the Shapley value includes:

[0030] The mean of the absolute values ​​of the normalized Shapley values ​​is used to obtain the optimized weights.

[0031] This application also provides a tunnel thermal hazard detection system, including:

[0032] The data acquisition module is used to acquire geological feature data of the target tunnel;

[0033] The parameter determination module is used to determine a group of thermal hazard parameters that have a correlation with the geological feature data that is greater than a set threshold.

[0034] The range assessment module is used to set the range of thermal hazard quantitative score values ​​based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group.

[0035] The weight setting module is used to set the weight values ​​corresponding to each of the heat hazard parameters using the analytic hierarchy process, calculate the heat hazard score based on the weight values ​​and the range of the heat hazard quantitative score, and determine the heat hazard level corresponding to the heat hazard score.

[0036] The model training module is used to establish a dataset based on the geological feature data, the thermal hazard level and extreme case data, and to train a lightweight gradient boosting model based on the dataset.

[0037] The thermal hazard detection module is used to input the geological feature data of the current unexcavated tunnel section into the lightweight gradient enhancement model and output the tunnel thermal hazard level of the current unexcavated tunnel section.

[0038] Optional, also includes:

[0039] The model update detection module is used to obtain the actual thermal hazard parameters of the current tunnel segment after the current tunnel segment is excavated; calculate the Shapley value corresponding to each actual thermal hazard parameter through a lightweight gradient boosting model; calculate the optimized weight based on the Shapley value; replace the weight value with the optimized weight and determine the updated thermal hazard level; and predict the tunnel thermal hazard level of the next tunnel segment based on the real-time tunnel thermal hazard detection model.

[0040] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0041] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described above when it invokes the computer program in the memory.

[0042] This application provides a method for detecting thermal hazards in tunnels, comprising: acquiring geological feature data of a target tunnel; determining a group of thermal hazard parameters whose correlation with the geological feature data is greater than a set threshold; setting a range of thermal hazard quantitative scoring values ​​for each thermal hazard parameter in the thermal hazard parameter group; setting weight values ​​corresponding to each thermal hazard parameter using the analytic hierarchy process (AHP); calculating a thermal hazard score based on the weight values ​​and the range of thermal hazard quantitative scoring values; determining the thermal hazard level corresponding to the thermal hazard score; establishing a dataset based on the geological feature data, the thermal hazard level, and extreme case data; training a lightweight gradient boosting model based on the dataset; inputting the geological feature data of the unexcavated current tunnel segment into the lightweight gradient boosting model; and outputting the tunnel thermal hazard level of the unexcavated current tunnel segment.

[0043] This application, by acquiring geological feature data of the target tunnel, enables a comprehensive and accurate understanding of the tunnel's natural geographical environment. The geological feature data encompasses crucial information related to tunnel thermal hazards, laying a solid foundation for subsequent thermal hazard risk assessment and making the assessment more targeted and scientific. Secondly, by identifying a group of thermal hazard parameters whose correlation with the geological feature data exceeds a set threshold, key parameters significantly impacting tunnel thermal hazards are effectively screened, redundant information is eliminated, and the efficiency and accuracy of thermal hazard assessment are improved, ensuring the reliability of the assessment results. Based on the thermal hazard parameter group, a range of quantitative thermal hazard scoring values ​​is set, realizing the quantitative processing of thermal hazard parameters. This transforms abstract thermal hazard risks into concrete numerical values, facilitating comparison and analysis, and providing a unified quantitative standard for subsequent weight setting and hazard score calculation. By using the analytic hierarchy process (AHP) to set weight values ​​for each thermal hazard parameter and calculating the thermal hazard score based on these weight values, the interrelationships and differences in importance among the various thermal hazard parameters can be fully considered. This allows for a reasonable allocation of weights, making the calculated thermal hazard score more objective and accurate in reflecting the true situation of tunnel thermal hazards, providing a strong basis for early warning and prevention of tunnel thermal hazards. Furthermore, by establishing a dataset based on geological feature data, thermal hazard scores, and extreme case data, and training a lightweight gradient boosting model on this dataset, the model can quickly and accurately predict the thermal hazard level of the current unexcavated tunnel section. This provides timely and effective guidance for thermal hazard prevention during tunnel construction, helping to take preventative measures to reduce the impact of thermal hazards on tunnel construction and operation, ensuring the safety and quality of tunnel projects. It also facilitates the optimization of construction plans, improves construction efficiency, and reduces construction costs, which is of great significance for the smooth implementation and long-term stable operation of tunnel projects.

[0044] This application also provides a tunnel thermal hazard detection system, a computer-readable storage medium, and an electronic device, which have the above-mentioned beneficial effects, and will not be elaborated here. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0046] Figure 1 A flowchart of a tunnel thermal hazard detection method provided in an embodiment of this application;

[0047] Figure 2 This is a schematic diagram of the tunnel thermal hazard grading index provided in the embodiments of this application;

[0048] Figure 3 A schematic diagram illustrating the weight optimization results combining LightGBM and Shapley values, provided for embodiments of this application;

[0049] Figure 4 This is a schematic diagram illustrating the exemplary tunnel thermal hazard risk assessment results provided in an embodiment of this application.

[0050] Figure 5 This is a schematic diagram of a tunnel thermal hazard detection system provided in an embodiment of this application;

[0051] Figure 6 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] See Figure 1 , Figure 1 A flowchart of a tunnel thermal hazard detection method provided in this application embodiment, the method comprising:

[0054] S101: Obtain geological feature data of the target tunnel;

[0055] S102: Determine a group of thermal hazard parameters whose correlation with the geological feature data is greater than a set threshold;

[0056] S103: Set the range of values ​​for the thermal hazard quantitative score based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group;

[0057] S104: Use the analytic hierarchy process to set the weight values ​​corresponding to each of the heat hazard parameters, calculate the heat hazard score based on the weight values ​​and the range of the heat hazard quantitative score, and determine the heat hazard level corresponding to the heat hazard score.

[0058] S105: Establish a dataset based on the geological feature data, the thermal hazard risk level, and the extreme case data, and train a lightweight gradient boosting model based on the dataset;

[0059] S106: Input the geological feature data of the current unexcavated tunnel segment into the lightweight gradient boosting model, and output the tunnel thermal hazard level of the current unexcavated tunnel segment.

[0060] Geological feature data of the target tunnel can be obtained in various ways. For example, Geographic Information System (GIS) technology can be used to query geological feature data such as topography, geological structure, and hydrogeology of the area where the target tunnel is located through existing geographic information databases. Geographic information databases typically contain rich geological information, providing basic data support for the geological feature analysis of the tunnel. For example, topographic elevation data along the tunnel route can be obtained to understand the topographic relief of the area where the tunnel is located; geological structure data, including fault and fold information, can also be obtained. This information is of great significance for analyzing the stability of the tunnel and the geological background of heat hazards.

[0061] Furthermore, during tunnel construction, on-site surveys can be directly conducted to obtain geological characteristic data. By carrying out geological exploration at or around the tunnel construction site, using methods such as drilling and geophysical exploration, detailed geological information about the tunnel's location can be directly obtained. Drilling can obtain rock core samples; analysis of these cores can determine the rock type, structure, and fracture development, all factors closely related to the thermal hazards generated by the tunnel. Geophysical exploration, through various physical methods such as seismic wave exploration and electromagnetic exploration, can detect the distribution of underground geological structures and bodies, providing more comprehensive information for obtaining geological characteristic data.

[0062] In addition, satellite remote sensing technology can also serve as an auxiliary means of acquiring geological feature data. Satellite remote sensing can provide large-scale topographic and geomorphological images and related information. By interpreting and analyzing these images, the macro-geographical characteristics of the tunnel area can be obtained, such as topographic and geomorphological types and vegetation cover. This data can be combined with other acquired data to provide a more comprehensive geographical background for subsequent analysis of thermal hazard parameters.

[0063] In step S102, it is necessary to determine the thermal hazard parameter group. Since different thermal hazard parameters have different levels of hazard, only thermal hazard parameters whose correlation with the geological feature data is greater than a set threshold are processed and grouped into thermal hazard parameter groups.

[0064] See Figure 2 , Figure 2This diagram illustrates the tunnel thermal hazard grading indicators provided in this application embodiment. In one feasible implementation, the tunnel thermal hazard hazard evaluation indicators refer to four primary indicators: hydrothermal activity characteristics, rock mass heat-enrichment characteristics, geological structural conditions, and tunnel engineering conditions, as well as corresponding secondary indicators. The secondary indicators for hydrothermal activity characteristics are the highest temperature and maximum flow rate of the hot spring; the secondary indicators for rock mass heat-enrichment characteristics are the rock mass thermal conductivity, rock mass specific heat capacity, geothermal gradient, and predicted geothermal temperature; the secondary indicators for geological structural conditions are the fault zone width and fracture opening; and the secondary indicators for tunnel engineering conditions are the tunnel burial depth, the distance between the tunnel and the fault zone, the distance between the tunnel and the hot spring outlet, and the distance between the tunnel and surface water.

[0065] Subsequently, in step S103, when setting the range of values ​​for the thermal hazard quantification score, it can be determined based on the physical meaning of the thermal hazard parameters and actual engineering experience. For each thermal hazard parameter, the degree of thermal hazard impact under different values ​​is analyzed. In specific applications, the range of values ​​for the thermal hazard quantification score can be set based on each thermal hazard parameter in the thermal hazard parameter group. The quantification and grading of qualitative indicators can adopt a grading method, specifically a numerical interval grading method, with all quantification gradings being five levels. An exemplary grading of tunnel thermal hazard risk and indicators can be found in the grading standards below:

[0066] The quantitative classification standard for the highest temperature of hot springs is as follows: Level I (less than or equal to 28℃), Level II (28~37℃), Level III (37~50℃), Level IV (50~80℃), and Level V (>80℃).

[0067] The maximum flow rate of hot springs is quantified and classified as follows: Level I (0L / s), Level II (0~2L / s), Level III (2~10L / s), Level IV (10~20L / s), and Level V (>20L / s).

[0068] The quantitative classification standard for thermal conductivity of rock mass is as follows: Grade I (less than or equal to 1.0 W / (m·K)), Grade II (1.0~1.5 W / (m·K)), Grade III (1.5~2.4 W / (m·K)), Grade IV (2.4~3.4 W / (m·K)), and Grade V (>3.4 W / (m·K)).

[0069] The rock mass specific heat capacity classification standard is as follows: Grade I (less than or equal to 600 J / (kg·K)), Grade II (600~750 J / (kg·K)), Grade III (750~800 J / (kg·K)), Grade IV (800~1000 J / (kg·K)), and Grade V (>1000 J / (kg·K)).

[0070] The quantitative classification standard for geothermal gradient is as follows: Level I (less than or equal to 1.8℃ / 100m), Level II (1.8~3.0℃ / 100m), Level III (3.0~6.0℃ / 100m), Level IV (6.0~20.0℃ / 100m), and Level V (>20.0℃ / 100m).

[0071] The quantitative classification standard for predicted ground temperature values ​​is as follows: Level I (less than or equal to 28℃ / 100m), Level II (28~37℃ / 100m), Level III (37~50℃ / 100m), Level IV (50~60℃ / 100m), and Level V (>60℃ / 100m).

[0072] The quantitative classification standard for fault zone width is as follows: Class I (0m), Class II (0~5m), Class III (5~50m), Class IV (50~200m), and Class V (>200m).

[0073] The quantitative classification standard for fracture opening is as follows: Grade I (<0.1mm), Grade II (0.1~0.5mm), Grade III (0.5~1mm), Grade IV (20~50mm), and Grade V (>50mm).

[0074] The quantitative classification standard for tunnel burial depth is as follows: Level I (less than or equal to 50m), Level II (50~100m), Level III (100~500m), Level IV (500~2000m), and Level V (>2000m).

[0075] The quantitative classification standard for the distance between tunnels and fault zones is as follows: Level I (>1.3km), Level II (0.7~1.3km), Level III (0.2~0.7km), Level IV (0.1~0.2km), and Level V (less than or equal to 0.1km).

[0076] The quantitative classification standard for the distance between the tunnel and the hot spring is as follows: Level I (>20km), Level II (10~20km), Level III (5~10km), Level IV (0.5~5km), and Level V (less than or equal to 0.5km).

[0077] The quantitative classification standard for the distance between tunnels and surface water is as follows: Level I (less than or equal to 0.5km), Level II (0.5~1km), Level III (1~3km), Level IV (3~10km), and Level V (>10km).

[0078] The quantitative classification standard for tunnel thermal hazard risk is as follows: Level I (less than or equal to 28℃, low risk), Level II (28~37℃, relatively low risk), Level III (37~50℃, medium risk), Level IV (50~60℃, high risk), and Level V (>60℃, extremely high risk).

[0079] The rating range for the indicator classification is as follows: Level I (0 ~ 4), Level II (4 ~ 8), Level III (8 ~ 12), Level IV (12 ~ 16), and Level V (16 ~ 20). The rating range for the heat hazard risk is as follows: Level I (0 ~ 30), Level II (30 ~ 45), Level III (45 ~ 60), Level IV (60 ~ 75), and Level V (75 ~ 100).

[0080] In step S104, the weight values ​​corresponding to each heat hazard parameter can be set using the analytic hierarchy process (AHP), and the weight values ​​can be input into the heat hazard score calculation formula to calculate the heat hazard score.

[0081] The formula for calculating the heat damage score is as follows:

[0082] , (1).

[0083] in, Score the risk of heat damage; Let be the weight value of the i-th indicator; Let be the quantitative grading score of the i-th indicator.

[0084] In another feasible implementation, a hierarchical model can be constructed first. The tunnel thermal hazard assessment is used as the target layer, and each thermal hazard parameter in the thermal hazard parameter group is used as the criterion layer. A hierarchical diagram is then constructed, and scores based on expert prior knowledge are obtained.

[0085] Next, using the calculation steps of the Analytic Hierarchy Process (AHP), the weight values ​​of each heat hazard parameter are calculated. The AHP quantifies expert prior knowledge and ensures the rationality and reliability of the weight values ​​through consistency checks and other steps. Based on the calculated weight values ​​and the quantified scores of each heat hazard parameter, the heat hazard score can be calculated. Specifically, the quantified score of each parameter is multiplied by its corresponding weight value, and then all the multiplications are summed to obtain the final heat hazard score.

[0086] By setting weight values ​​and calculating thermal hazard scores using the analytic hierarchy process (AHP), the importance differences of various thermal hazard parameters in the thermal hazard assessment can be fully considered, making the assessment more scientific, reasonable, and accurate. Furthermore, the AHP calculation process is relatively simple, easy to understand and operate, and suitable for multi-factor comprehensive assessment problems such as tunnel thermal hazard assessment.

[0087] Subsequently, a lightweight gradient boosting model was trained. When building the dataset, it was necessary to integrate the acquired geological feature data, the calculated thermal hazard score, and extreme case data. The geological feature data consisted of the tunnel's geographical background information; the thermal hazard score was a quantitative assessment result considering various thermal hazard parameters; and the extreme case data represented situations where all indicators were low-risk (low values) or high-risk (high values). These data collectively constituted the dataset used to train the model.

[0088] Then, a lightweight gradient boosting model was selected as the training model. During training, methods such as cross-validation can be used to optimize and validate the model. By dividing the dataset into training and validation sets, the model is trained on the training set, and then its performance is evaluated on the validation set. Based on the evaluation results, the model parameters are adjusted, the model structure is optimized, and the model's prediction accuracy and generalization ability are improved. By establishing a dataset and training a lightweight gradient boosting model, factors such as the geological characteristics of tunnels, thermal hazard assessment results, and extreme conditions can be comprehensively considered to construct a model that can accurately predict the thermal hazard level of tunnels. This model can provide a scientific basis for the prevention and control of tunnel thermal hazards, improving the safety and reliability of tunnel engineering.

[0089] In one feasible implementation, this step may include the following detailed steps:

[0090] The first step is to assess the thermal hazard risk level of the unexcavated tunnel section based on the geological feature data.

[0091] The second step is to collect thermal hazard risk assessment indicators for the excavated tunnel sections and calculate the thermal hazard risk score.

[0092] The third step is to construct extreme case data based on prior knowledge data:

[0093] The fourth step involves integrating the heat hazard risk level, the heat hazard risk score, and the extreme case data into a dataset, dividing the dataset into a training set and a test set, and adjusting the model parameters of the lightweight gradient boosting model through cross-validation and Bayesian optimization methods to train the lightweight gradient boosting model.

[0094] In practical applications, the dataset can be divided into training and test sets. This can be done randomly, dividing the data into training and test sets according to a certain ratio (e.g., 70% - 30%, 80% - 20%, etc.). Alternatively, stratified sampling can be used to ensure that the distribution of data in the training and test sets is consistent with the overall dataset. A lightweight gradient boosting model (such as LightGBM) is used, and cross-validation is employed to evaluate the model's performance under different parameter combinations. K-fold cross-validation can be used, dividing the training set into k subsets. Each time, k-1 subsets are used as training data, and the remaining subset is used as validation data, repeated k times, and the average value is used as the model performance evaluation metric. Bayesian optimization is then used to optimize the parameters of the lightweight gradient boosting model. Bayesian optimization automatically adjusts the parameter range and search strategy based on the model's performance during training to find the optimal combination of model parameters. During training, the model parameters are continuously adjusted based on the results of cross-validation and Bayesian optimization, ultimately resulting in a high-performance lightweight gradient boosting model.

[0095] Subsequently, the geological feature data of the current unexcavated tunnel segment can be input into the lightweight gradient enhancement model to output the tunnel thermal hazard level of the current unexcavated tunnel segment.

[0096] This application embodiment, by acquiring geological feature data of the target tunnel, can comprehensively and accurately understand the natural geographical environment in which the tunnel is located. The geological feature data covers important information related to tunnel thermal hazards, laying a solid foundation for subsequent thermal hazard risk assessment and making the assessment work more targeted and scientific. Secondly, by identifying a group of thermal hazard parameters whose correlation with the geological feature data exceeds a set threshold, key parameters with a significant impact on tunnel thermal hazards are effectively screened, redundant information is eliminated, and the efficiency and accuracy of thermal hazard assessment are improved, ensuring the reliability of the assessment results. Based on the thermal hazard parameter group, a range of values ​​for quantitative thermal hazard scoring is set, realizing the quantitative processing of thermal hazard parameters. This transforms abstract thermal hazard risks into concrete numerical values, facilitating comparison and analysis, and providing a unified quantitative standard for subsequent weight setting and hazard score calculation. By using the analytic hierarchy process (AHP) to set weight values ​​for each thermal hazard parameter and calculating the thermal hazard score based on these weight values, the interrelationships and differences in importance among the various thermal hazard parameters can be fully considered. This allows for a reasonable allocation of weights, making the calculated thermal hazard score more objective and accurate in reflecting the true situation of tunnel thermal hazards, providing a strong basis for early warning and prevention of tunnel thermal hazards. Furthermore, by establishing a dataset based on geological feature data, thermal hazard scores, and extreme case data, and training a lightweight gradient boosting model on this dataset, the model can quickly and accurately predict the thermal hazard level of the current unexcavated tunnel section. This provides timely and effective guidance for thermal hazard prevention during tunnel construction, helping to take preventative measures to reduce the impact of thermal hazards on tunnel construction and operation, ensuring the safety and quality of tunnel projects. It also facilitates the optimization of construction plans, improves construction efficiency, and reduces construction costs, which is of great significance for the smooth implementation and long-term stable operation of tunnel projects.

[0097] Based on the above embodiments, as a preferred embodiment, in practical applications, the lightweight gradient boosting model can be applied each time to predict the thermal hazard level of the unexcavated tunnel section. After the current tunnel section is excavated, the lightweight gradient boosting model can be updated based on the data of the current tunnel section to more accurately detect the thermal hazard level of the next tunnel section. The specific process is as follows:

[0098] The first step is to obtain the actual thermal hazard parameters of the current tunnel section;

[0099] The second step is to calculate the Shapley value corresponding to each of the actual heat hazard parameters using a lightweight gradient boosting model, and then calculate the optimization weights based on the Shapley values.

[0100] The third step is to replace the weight value with the optimized weight and determine the updated heat hazard level.

[0101] Step 4: Reconstruct the training set in the dataset based on the updated thermal hazard risk level, and train the real-time tunnel thermal hazard detection model.

[0102] Step 5: Based on the real-time tunnel heat hazard detection model, predict the tunnel heat hazard level of the next tunnel section.

[0103] The Shapley value is calculated as shown in formula (2).

[0104] (2);

[0105] In the above formula, N is the set of all features, which mainly include the secondary heat hazard parameters described above; i is the i-th feature; S is a subset of features that does not include feature i. Let be the value function, representing the total value (or revenue, cost, etc.) that can be obtained when the representation includes features. Let be the Shapley value of the i-th feature.

[0106] The objective weights of each indicator after optimization are calculated according to formula (3).

[0107] (3);

[0108] In the formula, Let be the Shapley value at the i-th point; Let be the optimized weight for the j-th feature. The optimized weight can be obtained by normalizing the mean of the absolute values ​​of the Shapley values. Subsequently, the optimized objective weight is used to replace the original weight value (i.e., formula (1)) to re-evaluate the thermal hazard level of each tunnel section.

[0109] As the tunnel is excavated, thermal hazard index data of new tunnel sections are continuously collected, added to the training set, and the model is updated to achieve dynamic updates of weights and dynamic optimization of hazard assessment.

[0110] In practical applications, sensors can be deployed at the tunnel site to collect real-time data on heat hazard-related indicators, while LightGBM and SHAP models are deployed in the cloud. The weights of each indicator are dynamically calculated based on the Shapley value, enabling real-time dynamic assessment of heat hazard risk.

[0111] See Figure 3 , Figure 3 This diagram illustrates the weight optimization results combining LightGBM and Shapley values, as provided in an embodiment of this application. Figure 3 (a) shows the mean absolute SHAP value results for the LightGBM model. Figure 3 (b) shows the weighting optimization results, illustrating the weight and percentage of each secondary thermal hazard parameter when applying the Shapley value and prior weights, respectively. It is easy to understand that... Figure 3This is merely an example of how to set weights and the result of Shapley value calculation.

[0112] See Figure 4 , Figure 4 The exemplary tunnel thermal hazard assessment results provided in this application embodiment are illustrated in the diagram, showing the distribution of different geological layers, altitude, and the relative location of the tunnel. The following is... Figure 4 The meaning of each element:

[0113] Altitude (m): Figure 4 The elevation is marked on the left and right sides, in meters (m), showing the undulating terrain.

[0114] Geological layers: Granite: Represented by dotted patterns. Granitic gneiss: Represented by oblique lines. Paleo-Mesoproterozoic: Marked with text. Himalayan granite: Represented by a net-like pattern.

[0115] Geological structures: Faults: Represented by lines, showing the location and orientation of fractures. Unconformities: Represented by wavy lines, indicating discontinuous sedimentary deposits between strata.

[0116] tunnel: Figure 4 The areas filled with diagonal lines indicate the location and direction of the tunnel. The tunnels are numbered from f5-7 to f5-11, showing the different sections of the tunnel.

[0117] Topographic lines: Represented by thin solid lines, showing the undulations of the terrain.

[0118] Tunnel thermal hazard classification: Figure 4 The bar chart at the bottom represents the thermal hazard risk classification of different sections of the tunnel, from I to V, indicating a risk level from low to high.

[0119] Mileage (m): The scale at the bottom of the map indicates the mileage of the tunnel, from K0+000 to K10+295, in meters.

[0120] angle: Figure 4 The 53° angle in the upper left corner represents the azimuth angle, indicating the direction of the cross-section.

[0121] Figure 4 The diagram shows the tunnel thermal hazard level of the current unexcavated tunnel section, which is output by a lightweight gradient boosting model and can be used to analyze and plan geological conditions and potential risks in tunnel construction.

[0122] This embodiment improves prediction accuracy by integrating data from excavated tunnels, evaluation results from unexcavated sections, and prior knowledge from experts, and trains the model using a lightweight gradient boosting model. Furthermore, by combining subjective weights (i.e., the weight values ​​in formula (1)) with objective weights (i.e., the Shapley value), subjective bias and data errors are reduced, thus improving evaluation accuracy.

[0123] See Figure 5 , Figure 5 This application provides a schematic diagram of a tunnel thermal hazard detection system, which includes:

[0124] The data acquisition module is used to acquire geological feature data of the target tunnel;

[0125] The parameter determination module is used to determine a group of thermal hazard parameters that have a correlation with the geological feature data that is greater than a set threshold.

[0126] The range assessment module is used to set the range of thermal hazard quantitative score values ​​based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group.

[0127] The weight setting module is used to set the weight values ​​corresponding to each of the heat hazard parameters using the analytic hierarchy process, calculate the heat hazard score based on the weight values ​​and the range of the heat hazard quantitative score, and determine the heat hazard level corresponding to the heat hazard score.

[0128] The model training module is used to establish a dataset based on the geological feature data, the thermal hazard level and extreme case data, and to train a lightweight gradient boosting model based on the dataset.

[0129] The thermal hazard detection module is used to input the geological feature data of the current unexcavated tunnel section into the lightweight gradient enhancement model and output the tunnel thermal hazard level of the current unexcavated tunnel section.

[0130] Based on the above embodiments, as a preferred embodiment, it further includes:

[0131] The model update detection module is used to obtain the actual thermal hazard parameters of the current tunnel segment after the current tunnel segment is excavated; calculate the Shapley value corresponding to each actual thermal hazard parameter through a lightweight gradient boosting model; calculate the optimized weight based on the Shapley value; replace the weight value with the optimized weight and determine the updated thermal hazard level; and predict the tunnel thermal hazard level of the next tunnel segment based on the real-time tunnel thermal hazard detection model.

[0132] Based on the above embodiments, as a preferred embodiment, it further includes:

[0133] The thermal hazard parameter determination module is used to determine primary thermal hazard parameters and corresponding secondary thermal hazard parameters. The primary thermal hazard parameters include hydrothermal activity characteristics, rock mass heat-enrichment characteristics, geological structural conditions, and tunnel engineering conditions. The secondary thermal hazard parameters corresponding to the hydrothermal activity characteristics include the highest temperature and maximum flow rate of the hot spring. The secondary thermal hazard parameters for the rock mass heat-enrichment characteristics include the rock mass thermal conductivity, rock mass specific heat capacity, geothermal gradient, and predicted geothermal temperature. The secondary thermal hazard parameters for the geological structural conditions include the width of the fault zone and the fracture opening. The secondary thermal hazard parameters for the tunnel engineering conditions include the tunnel depth, the distance between the tunnel and the fault zone, the distance between the tunnel and the hot spring outlet, and the distance between the tunnel and surface water.

[0134] This application also provides an embodiment of a computer-readable storage medium and a computer program product. Both the computer-readable storage medium and the computer program product may store a computer program that, when executed by a processor, implements the steps of the method described in the above method embodiments.

[0135] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] The computer-readable storage medium provided in this embodiment includes the method mentioned above, and has the same effect.

[0137] This application also provides an electronic device, see [link to document]. Figure 6 The present application provides a structural diagram of an electronic device, such as... Figure 6 As shown, it may include a processor 1410 and a memory 1420.

[0138] The processor 1410 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 1410 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1410 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 1410 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 1410 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0139] The memory 1420 may include one or more computer-readable storage media, which may be non-transitory. The memory 1420 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 1420 is used to store at least the following computer program 1421, which, after being loaded and executed by the processor 1410, is capable of implementing the relevant steps in the methods executed by the electronic device side as disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 1420 may also include an operating system 1422 and data 1423, etc., and the storage method may be temporary storage or permanent storage. The operating system 1422 may include Windows, Linux, Android, etc.

[0140] In some embodiments, the electronic device may further include a display screen 1430, an input / output interface 1440, a communication interface 1450, a sensor 1460, a power supply 1470, and a communication bus 1480.

[0141] certainly, Figure 6 The structure of the electronic device shown does not constitute a limitation on the electronic device in the embodiments of this application. In practical applications, the electronic device may include more than [other components]. Figure 6 More or fewer components as shown, or combinations of certain components.

[0142] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. As the system provided in the embodiments corresponds to the method provided in the embodiments, the description is relatively simple; relevant parts can be found in the method section.

[0143] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

[0144] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for detecting thermal hazards in tunnels, characterized in that, include: Obtain geological feature data of the target tunnel; Identify a group of thermal hazard parameters whose correlation with the geological feature data is greater than a set threshold. The range of values ​​for the thermal hazard quantitative score is set based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group. The weight values ​​corresponding to each of the heat hazard hazard parameters are set using the analytic hierarchy process (AHP). Based on the weight values ​​and the range of values ​​for the heat hazard quantitative scoring, the heat hazard hazard score is calculated, and the heat hazard hazard level corresponding to the heat hazard hazard score is determined. A dataset is established based on the geological feature data, the thermal hazard risk level, and the extreme case data, and a lightweight gradient boosting model is trained based on the dataset. The geological feature data of the current unexcavated tunnel segment is input into the lightweight gradient enhancement model, and the tunnel thermal hazard level of the current unexcavated tunnel segment is output.

2. The tunnel thermal hazard detection method according to claim 1, characterized in that, After the current tunnel section is excavated, the following will also be included: Obtain the actual thermal hazard parameters of the current tunnel section; The Shapley values ​​corresponding to each of the actual thermal hazard parameters are calculated using a lightweight gradient boosting model, and the optimization weights are calculated based on the Shapley values. The optimized weights are used to replace the weight values, and the updated heat hazard level is determined. Based on the updated thermal hazard risk level, the training set in the dataset is reconstructed, and a real-time tunnel thermal hazard detection model is trained. The real-time tunnel heat hazard detection model is used to predict the tunnel heat hazard level of the next tunnel section.

3. The tunnel thermal hazard detection method according to claim 1, characterized in that, Before setting the range of values ​​for the quantitative score of heat hazard based on each heat hazard hazard parameter in the heat hazard hazard parameter group, the following steps are also included: The primary thermal hazard parameters and their corresponding secondary thermal hazard parameters are determined. The primary thermal hazard parameters include hydrothermal activity characteristics, rock mass heat-enrichment characteristics, geological structural conditions, and tunnel engineering conditions. The secondary thermal hazard parameters corresponding to the hydrothermal activity characteristics include the maximum temperature and maximum flow rate of the hot spring. The secondary thermal hazard parameters for the rock mass heat-enrichment characteristics include the rock mass thermal conductivity, rock mass specific heat capacity, geothermal gradient, and predicted geothermal temperature. The secondary thermal hazard parameters for the geological structural conditions include the width of the fault zone and the fissure opening. The secondary thermal hazard parameters for the tunnel engineering conditions include the tunnel burial depth, the distance between the tunnel and the fault zone, the distance between the tunnel and the hot spring outlet, and the distance between the tunnel and surface water.

4. The tunnel thermal hazard detection method according to claim 1, characterized in that, The analytic hierarchy process (AHP) is used to set weight values ​​for each of the aforementioned heat hazard parameters. Based on these weight values ​​and the range of values ​​for the heat hazard quantitative scoring, a heat hazard score is calculated, and the heat hazard level corresponding to each score is determined, including: The weight values ​​corresponding to each of the aforementioned heat hazard parameters were set using the analytic hierarchy process. The weighted values ​​are input into the heat hazard score calculation formula to calculate the heat hazard risk score; The formula for calculating the heat damage score is as follows: S T Score for heat hazard risk; w i S represents the weight value of the i-th indicator; i Let be the quantitative grading score of the i-th indicator.

5. The tunnel thermal hazard detection method according to claim 1, characterized in that, A dataset is established based on the geological feature data, the thermal hazard risk level, and extreme case data. A lightweight gradient boosting model is trained based on the dataset, including: The thermal hazard level of the unexcavated tunnel section was assessed based on the geological feature data. Collect thermal hazard risk assessment indicators for the excavated tunnel sections and calculate the thermal hazard risk score; Constructing extreme case data based on prior knowledge data: The heat hazard risk level, the heat hazard risk score, and the extreme case data are integrated into a dataset, which is then divided into a training set and a test set. The model parameters of the lightweight gradient boosting model are adjusted through cross-validation and Bayesian optimization methods to train the lightweight gradient boosting model.

6. The method according to claim 2, characterized in that, The optimization weights are calculated based on the Shapley value, including: The mean of the absolute values ​​of the normalized Shapley values ​​is used to obtain the optimized weights.

7. A tunnel thermal hazard detection system, characterized in that, include: The data acquisition module is used to acquire geological feature data of the target tunnel; The parameter determination module is used to determine a group of thermal hazard parameters that have a correlation with the geological feature data that is greater than a set threshold. The range assessment module is used to set the range of thermal hazard quantitative score values ​​based on each thermal hazard hazard parameter in the thermal hazard hazard parameter group. The weight setting module is used to set the weight values ​​corresponding to each of the heat hazard parameters using the analytic hierarchy process, calculate the heat hazard score based on the weight values ​​and the range of the heat hazard quantitative score, and determine the heat hazard level corresponding to the heat hazard score. The model training module is used to establish a dataset based on the geological feature data, the thermal hazard level and extreme case data, and to train a lightweight gradient boosting model based on the dataset. The thermal hazard detection module is used to input the geological feature data of the current unexcavated tunnel section into the lightweight gradient enhancement model and output the tunnel thermal hazard level of the current unexcavated tunnel section.

8. The tunnel thermal hazard detection system according to claim 7, characterized in that, Also includes: The model update detection module is used to obtain the actual thermal hazard parameters of the current tunnel segment after the current tunnel segment is excavated; The Shapley values ​​corresponding to each of the actual thermal hazard parameters are calculated using a lightweight gradient boosting model, and the optimization weights are calculated based on the Shapley values. The optimized weights are used to replace the weight values, and the updated heat hazard level is determined. The real-time tunnel heat hazard detection model is used to predict the tunnel heat hazard level of the next tunnel section.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the method as claimed in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, implements the steps of the method as described in any one of claims 1 to 6.

Images