Tunnel collapse risk prediction method and system based on multi-parameter fusion

Abstract

The invention provides a tunnel collapse risk prediction method and system based on multi-parameter fusion, and relates to the technical field of tunnel construction. The method comprises the steps of collecting various monitoring data of a to-be-predicted tunnel; constructing a tunnel collapse risk index system; calculating a collapse risk index of the to-be-predicted tunnel according to the tunnel collapse risk index system and various monitoring data; judging whether the collapse risk index is greater than or equal to a collapse risk index threshold; if yes, constructing a graph data structure according to various monitoring data; combining the one-dimensional convolutional neural network and the graph convolutional network to construct a tunnel collapse risk prediction model; and inputting various monitoring data and the graph data structure into the tunnel collapse risk prediction model, performing collapse risk prediction on the to-be-predicted tunnel, and outputting a tunnel collapse risk level. The method considers the superposition of geological conditions, construction methods and environmental factors, does not depend on single monitoring data any more, improves the reasonability of a sample set, and is suitable for tunnel collapse risk prediction in a complex environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel construction, in particular to a tunnel collapse risk prediction method and system based on multi-parameter fusion. BACKGROUND

[0002] More and more highways and railways in China extend to mountainous areas where geological disasters occur frequently. As an underground structure crossing mountains and hills, tunnels need to pass through rock masses where complex terrain, variable hydrogeological conditions, and groups of unfavorable geological problems such as faults, karst, and disaster-causing water bodies occur during construction. Tunnels face many safety hazards, among which tunnel collapse, as a serious and common safety accident, is related to many factors such as geological conditions, construction methods, and environmental changes. Collapse disasters such as water and mud inrush, rock burst, and lining cracking caused by the superposition of multiple factors are unpredictable, rapid, and destructive, causing serious safety hazards and possibly leading to personnel injuries and delays in project progress, resulting in significant economic losses and social impact. Therefore, the prevention and risk assessment of tunnel collapse play a key role in disaster avoidance during tunnel construction.

[0003] In the prior art, traditional tunnel collapse risk prediction methods mainly rely on expert experience and traditional engineering analogy methods, ignoring the superposition effects of multiple factors such as tunnel environment, geological conditions, construction methods, and support structures, making it difficult to comprehensively and accurately reflect the actual risks of tunnels under different conditions and construction stages.

[0004] In addition, traditional tunnel collapse risk prediction methods mainly rely on single types of monitoring data such as support structure stress, temperature, and humidity. Although these data can provide some risk warnings, tunnel collapse is the superposition of multiple factors, and multiple indicators are interrelated, making the prediction accuracy and real-time performance of a single factor often limited and unable to comprehensively describe the risks of tunnels in complex environments. SUMMARY

[0005] To solve the technical problem that traditional tunnel collapse risk prediction methods in the prior art mainly rely on expert experience and traditional engineering analogy methods, ignore the superposition effects of multiple factors such as tunnel environment, geological conditions, construction methods, and support structures, and are difficult to comprehensively and accurately reflect the actual risks of tunnels under different conditions and construction stages, and mainly rely on single types of monitoring data, making the prediction accuracy and real-time performance often limited and unable to comprehensively describe the risks of tunnels in complex environments, the present application provides a tunnel collapse risk prediction method and system based on multi-parameter fusion.

[0006] The technical solutions provided by the embodiments of the present application are as follows: First aspect: The tunnel collapse risk prediction method based on multi-parameter fusion provided by the embodiments of the present application comprises: S1: collecting various monitoring data of a tunnel to be predicted; S2: constructing a tunnel collapse risk index system; S3: calculating a collapse risk index of the tunnel to be predicted according to the tunnel collapse risk index system and the various monitoring data; S4: judging whether the collapse risk index is greater than or equal to a collapse risk index threshold value; if yes, proceeding to S5; otherwise, returning to S2; S5: constructing a graph data structure according to the various monitoring data; S6: constructing a tunnel collapse risk prediction model in combination with a one-dimensional convolutional neural network and a graph convolutional network; S7: inputting the various monitoring data and the graph data structure into the tunnel collapse risk prediction model to predict the collapse risk of the tunnel to be predicted and output a tunnel collapse risk level.

[0007] Second aspect: The embodiment of the present application provides a tunnel collapse risk prediction system based on multi-parameter fusion, which comprises: a processor; a memory, wherein the memory stores computer readable instructions, and the computer readable instructions are executed by the processor to realize the tunnel collapse risk prediction method based on multi-parameter fusion as described in the first aspect.

[0008] Third aspect: The embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the tunnel collapse risk prediction method based on multi-parameter fusion as described in the first aspect.

[0009] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: In the embodiment of the present application, the tunnel collapse risk index of the tunnel to be predicted is calculated according to the tunnel collapse risk index system and the various monitoring data by constructing the tunnel collapse risk index system, and the prediction of the tunnel collapse risk is no longer dependent on the expert experience and the traditional engineering analogy method, and the superposition effects of various factors such as tunnel environment, geological conditions, construction method and supporting structure are considered, so that the actual risk of the tunnel in different conditions and construction stages can be comprehensively and accurately reflected, the graph data structure is constructed, the various monitoring data and the graph data structure are input into the tunnel collapse risk prediction model, the collapse risk of the tunnel to be predicted is predicted, and the tunnel collapse risk level is output, so that the prediction of the tunnel collapse risk is no longer dependent on a single type of monitoring data, the accuracy and real-time performance of the prediction of the tunnel collapse risk are no longer limited, and the risk of the tunnel can be comprehensively described in a complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiments description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0011] Figure 1 A flowchart of a tunnel collapse risk prediction method based on multi-parameter fusion provided by an embodiment of the present application is shown in FIG. 1. Figure 2 A structural diagram of a tunnel collapse risk prediction system based on multi-parameter fusion provided by an embodiment of the present application is shown in FIG. 2. DETAILED DESCRIPTION

[0012] The technical solutions in the present application will be described below with reference to the drawings.

[0013] In the embodiments of the present application, the words such as "example", "for example" and the like are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0014] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0015] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0016] In order to make the technical problems, technical solutions and advantages to be solved by the present application more clear, the following will be described in detail with reference to the drawings and specific embodiments.

[0017] Reference is made to the drawings accompanying the specification Figure 1 , a flowchart of a tunnel collapse risk prediction method based on multi-parameter fusion provided by an embodiment of the present application is shown in FIG. 1.

[0018] The embodiment of the present application provides a tunnel collapse risk prediction method based on multi-parameter fusion, which can be realized by a tunnel collapse risk prediction device based on multi-parameter fusion, and the device can be a terminal or a server. The processing flow of the tunnel collapse risk prediction method based on multi-parameter fusion can include the following steps: S1: Collecting multiple monitoring data of a tunnel to be predicted.

[0019] Optionally, the monitoring data includes tunnel rock mass integrity coefficients, tunnel rock mass quality grades, tunnel longitudinal wave velocities, tunnel lighting brightness, tunnel rock mass stresses, tunnel support structure stresses, tunnel spans and heights, tunnel rock mass deformation amounts, tunnel groundwater pressures, tunnel rock mass microseismic characteristics and tunnel temperatures collected by different monitoring devices at multiple time points.

[0020] It should be noted that the tunnel rock mass integrity coefficient refers to an index for measuring the structure type and integrity of the tunnel rock mass, which reflects the development degree of fault fracture zones, weak interlayers, joint fissures and weathering degree in the rock mass. The tunnel rock mass quality grade is a parameter for evaluating the engineering characteristics of the rock mass, which not only reflects the strength of the rock mass, but also reflects the judgment criteria related to the engineering characteristics. In the national standard, five types are classified, i.e., type I, type II, type III, type IV and type V (which will not be described herein again). The tunnel longitudinal wave velocity refers to the propagation velocity of sound waves in the tunnel rock mass, which is closely related to the density and elastic properties of the rock mass. The tunnel lighting brightness refers to the light intensity inside the tunnel, which has an important influence on the field of view, safety and driver's perception in the tunnel. The tunnel rock mass stress refers to the stress value generated in the tunnel rock mass under the joint influence of gravity and tectonic stress, which affects the stability of the tunnel and the design of excavation and support. The tunnel support structure stress refers to the pressure or tension borne by the tunnel support structure during use, which reflects the damage critical value and bearing capacity of the support structure. The tunnel span and height refer to the horizontal distance between the two side walls of the tunnel and the vertical distance between the arch top and the bottom plate, which determines the excavation face size, excavation amount and disturbance degree to the rock mass. The tunnel rock mass deformation amount refers to the relative change amount of the horizontal distance between the two side walls of the tunnel and the vertical distance between the arch top and the bottom plate, which reflects the stress release amount of the tunnel rock mass and the precursor information of rock mass instability. The tunnel groundwater pressure refers to the fissure water pressure in the tunnel rock mass, which affects the waterproof design of the tunnel and the water level control in the construction process and the weakening effect on the rock mass structure surface. The tunnel rock mass microseismic characteristic refers to the vibration signal emitted when the rock mass in the tunnel cracks and expands during stress release, which reflects the distance between the current state of the rock mass and the critical value of rupture and instability. The tunnel humidity refers to the water vapor content in the air in the tunnel, which has an important influence on the corrosion of the tunnel, equipment maintenance and air quality.

[0021] In the embodiment of the present application, by comprehensively collecting various monitoring data closely related to the tunnel stability, combining with the detailed index definition, the scientificity, comprehensiveness and operability of the data basis of the risk prediction model are ensured, so that the accuracy and practicability of the tunnel collapse risk prediction are effectively improved.

[0022] S2: Construct a tunnel collapse risk index system.

[0023] It should be noted that the tunnel collapse risk index system is a set of indexes for system evaluation of tunnel stability and safety, according to the tunnel structure characteristics, environmental conditions and operation state, the key monitoring parameters reflecting the tunnel collapse risk are selected, and the scoring interval and weight of each index are set according to the scientific method. By comprehensively monitoring the data of each index, the collapse risk level of the tunnel can be quantified, and a quantitative basis is provided for risk prediction and safety decision. This system not only considers a single factor, but also reflects the overall stress, environmental change and structure health state of the tunnel through multi-dimensional and multi-level information fusion, and has comprehensiveness, scientificity and dynamic adaptability.

[0024] Optionally, the tunnel collapse risk index system includes a plurality of tunnel collapse risk indexes and a scoring interval corresponding to each tunnel collapse risk index.

[0025] It should be noted that the size of the scoring interval corresponding to each tunnel collapse risk index can be set by the person skilled in the art according to actual needs, which is not limited in the present application.

[0026] The tunnel collapse risk index includes: positive index and negative index.

[0027] It should be noted that the positive index refers to the factors that have a positive influence on the safety and stability of the tunnel, the larger the value of these indexes, the lower the risk of the tunnel. The negative index refers to the factors that have a negative influence on the safety and stability of the tunnel, the larger the value of these indexes, the higher the risk of the tunnel.

[0028] The positive index includes: tunnel rock mass integrity coefficient, tunnel rock mass quality grade, tunnel longitudinal wave velocity and tunnel lighting brightness.

[0029] The negative index includes: tunnel rock mass stress, tunnel support structure stress, tunnel span and height, tunnel rock mass deformation, tunnel groundwater pressure, tunnel rock mass microseismic characteristics, tunnel rock mass microseismic characteristics and tunnel temperature.

[0030] In the embodiment of the present application, by constructing a scientific and reasonable tunnel collapse risk index system, comprehensively integrating structural characteristics, environmental factors and operation state data, and clearly distinguishing positive and reverse indicators, the scientificity and accuracy of risk assessment are improved, and standardized and normalized data support is provided for subsequent risk prediction models based on deep learning, significantly improving the reliability and practicality of tunnel collapse risk identification and decision support.

[0031] S3: Calculate the collapse risk index of the tunnel to be predicted according to the tunnel collapse risk index system and various monitoring data.

[0032] It should be noted that the collapse risk index refers to a numerical value calculated by combining the tunnel collapse risk index system and the monitoring data, which reflects the potential risk level of tunnel collapse.

[0033] In the embodiment of the present application, by constructing the collapse risk index based on the tunnel collapse risk index system and the monitoring data, the quantitative integration and unified expression of tunnel multi-source monitoring data can be realized, the scientificity, accuracy and dynamic adaptability of tunnel risk assessment are improved, and a solid data foundation and decision support are provided for subsequent risk prediction modeling and intelligent operation management.

[0034] In one possible implementation, S3 specifically includes sub-steps S301 to S305: S301: According to each scoring interval, calculate the first weight coefficient of each monitoring data by Delphi method.

[0035] It should be noted that the Delphi method is a system method for prediction and decision-making based on expert group opinions. It usually collects experts' judgments and suggestions on specific problems through multiple rounds of anonymous questionnaire surveys, and the results are statistically analyzed and fed back after each round of investigation to promote experts to gradually reach a consensus through repeated thinking and mutual reference. This method can effectively integrate expert experience and reduce the influence of individual bias on decision-making, and is widely used in complex system analysis, index weight determination and future trend prediction, etc. It has scientificity, systematicness and high reliability.

[0036] In one possible implementation, S301 specifically includes sub-steps S3011 to S3017: S3011. Set the initial number of experts.

[0037] S3012: According to each scoring interval, score each monitoring data by each expert.

[0038] S3013: Calculate the average score of each monitoring data according to the scoring results of each monitoring data.

[0039] Optionally, the average score of various monitoring data can be calculated according to the following formula:

[0040] in, Indicates the first j The average score of the monitoring data. I Indicates the number of experts. A ij Indicates the first i The expert on the first j Scoring of various monitoring data.

[0041] S3014: Calculate the standard deviation of various monitoring data based on the mean score.

[0042] Optionally, the standard deviation of the scores for various monitoring data can be calculated using the following formula:

[0043] in, B j Indicates the first j The standard deviation of the scoring data.

[0044] S3015: Calculate the average standard deviation of the scores based on the standard deviation of the scores of various monitoring data.

[0045] Optionally, the standard deviation of the average rating can be calculated using the following formula:

[0046] in, Indicates the standard deviation of the average score. J This indicates the amount of monitoring data.

[0047] S3016: Determine if the average standard deviation of the ratings is greater than the rating standard deviation threshold. If yes, increase the number of experts and return to S3012. Otherwise, proceed to S3017.

[0048] It should be noted that those skilled in the art can set the value of the standard deviation threshold for scoring according to actual needs, and this invention does not limit this.

[0049] Optionally, the number of experts can be increased according to the following formula:

[0050] in, I new This indicates the number of newly added experts. Represents the ceiling function. B target This represents the standard deviation threshold for the scoring.

[0051] S3017: Calculate the first weight coefficient of each monitoring data according to the average score.

[0052] Optionally, the first weight coefficient of each monitoring data is calculated according to the following formula:

[0053] wherein, w represents the first weight coefficient of the i-th monitoring data. j

[0054] In the embodiment of the present application, by means of multiple rounds of expert scoring and dynamic supplement mechanism, the number of experts is dynamically increased when the standard deviation of the average score exceeds the preset threshold, which can not only quantify the difference of opinions among experts and correct the deviation in time, but also make the final average score truly reflect the consensus of multiple experts. The first weight coefficient calculated based on the average score not only takes into account the guiding significance of expert experience, but also improves the stability and credibility of the weighting through iterative feedback mechanism, thereby laying a scientific and reliable foundation for subsequent subjective and objective fusion weighting combined with entropy weight method, and significantly improving the accuracy, robustness and generalizability of the tunnel collapse risk assessment result.

[0055] S302: Normalize each monitoring data.

[0056] S303: Calculate the second weight coefficient of each monitoring data by entropy weight method according to the normalized each monitoring data.

[0057] It should be noted that the entropy weight method is an objective weighting method for determining index weight based on information entropy theory, which measures the contribution of each index data to the overall uncertainty of the system by analyzing the dispersion degree of the index data. The greater the change of the index data, the more information it provides, the smaller the entropy value, and the greater the weight. On the contrary, the smaller the change and the less the information, the smaller the weight. The entropy weight method can avoid the interference of artificial subjective factors, objectively reflect the relative importance of each index in multi-index decision-making problems, and is widely used in comprehensive evaluation, risk analysis and multi-attribute decision-making fields, and has the characteristics of scientificity, fairness and simple operation.

[0058] In one possible implementation, S303 specifically includes sub-steps S3031 to S3034: S3031. Calculate a plurality of entropy values of the normalized each monitoring data.

[0059] Optionally, the entropy values include information entropy values and fuzzy entropy values.

[0060] Further, the calculation formula of the information entropy value is specifically:

[0061] wherein,​​ Represents the normalized th j The information entropy value of the monitoring data, ln Represents the natural logarithm function. p ( Z j ) represents the normalized i-th j The probability distribution of various monitoring data.

[0062] Furthermore, the specific formula for calculating the fuzzy entropy value is as follows:

[0063] in, Represents the normalized th j Fuzzy entropy values ​​of monitoring data m Indicates the embedding dimension. H ( ) represents the Heaviside step function. r Indicates the tolerance threshold. Represents the normalized th j The monitoring data and the first The distance between various monitoring data Represents the normalized th Various monitoring data.

[0064] It should be noted that those skilled in the art can set the size of the embedding dimension and tolerance threshold according to actual needs, and this invention does not limit them.

[0065] S3032: Calculate various coefficients of variation for normalized monitoring data based on various entropy values.

[0066] Optionally, the coefficient of variation includes: a first coefficient of variation and a second coefficient of variation.

[0067] Furthermore, the formula for calculating the coefficient of variation is as follows:

[0068] in, Represents the normalized th j The first coefficient of variation of the monitoring data, Represents the normalized th j The second coefficient of variation of the monitoring data.

[0069] S3033: Calculate various importance indices for normalized monitoring data based on various coefficients of variation.

[0070] Optionally, the coefficient of variation includes: the first importance index and the second importance index.

[0071] Furthermore, the specific formula for calculating the importance index is as follows:

[0072] in, Represents the normalized th j The most important index of this monitoring data, Represents the normalized th j The second most important index of the monitoring data.

[0073] S3034: Calculate the second weighting coefficient for various monitoring data based on various importance indices.

[0074] Optionally, the second weighting coefficients for various monitoring data are calculated according to the following formula:

[0075] in, Indicates the first j The second weighting index for monitoring data.

[0076] In this embodiment of the invention, by calculating the information entropy value and fuzzy entropy value of the normalized monitoring data respectively, the two entropy values ​​are converted into corresponding coefficients of variation, and further normalized to obtain two importance indices. These indices are then averaged to generate a second weighting coefficient. This approach can simultaneously capture the differences in the overall probability distribution and local similarity of the monitoring data, thereby comprehensively reflecting the information contribution and fuzzy differences of each indicator. It reduces the bias of the single entropy method when dealing with noise or outliers, achieving a more objective, fair, and robust weighting. This provides a solid foundation for subsequent fusion with subjective weights and input to deep learning models, significantly improving the accuracy and stability of tunnel collapse risk assessment.

[0077] S304: Calculate the final weighting coefficients for various monitoring data based on the first and second weighting coefficients.

[0078] Optionally, the final weighting coefficients for various monitoring data are calculated according to the following formula:

[0079] in, W j Indicates the first j The final weighting coefficients for various monitoring data.

[0080] S305: Based on the final weighting coefficients of various monitoring data, the collapse risk index of the tunnel to be predicted is calculated using the ideal point method.

[0081] It should be noted that the ideal point method is a multi-attribute decision-making method. It ranks solutions by constructing ideal and negative ideal solutions and measuring their proximity to the ideal state. The ideal solution consists of the optimal values ​​of each evaluation index, while the negative ideal solution consists of the worst values. Actual solutions are then ranked based on their distance from both the ideal and negative ideal solutions. Solutions closer to the ideal solution and further away from the negative ideal solution are considered superior. The ideal point method comprehensively considers the influence of multiple indicators, intuitively reflecting the overall merits of solutions. It is widely used in engineering evaluation, risk assessment, and decision support, and is characterized by its clear logic and simple calculation.

[0082] In this embodiment of the invention, by combining the Delphi method and the entropy weight method to assign initial and objective weights to the monitoring data, and using normalization to eliminate dimensional differences, the collapse risk index of the tunnel is quantified and calculated using the ideal point method. This allows multi-source heterogeneous monitoring information to be integrated into an intuitive and comparable risk value, taking into account both expert experience and data characteristics, improving the scientific and rational nature of weight allocation, and enhancing the accuracy and stability of the risk index through standardization and comprehensive evaluation. This provides a high-quality data foundation and labels for subsequent deep learning-based tunnel risk prediction models, significantly improving the model's predictive performance and dynamic early warning capabilities, and providing reliable decision support for tunnel safety management.

[0083] In one possible implementation, S305 specifically includes sub-steps S3051 to S3055: S3051: Calculate the positive and negative ideal points of various monitoring data.

[0084] Optionally, the positive and negative ideal points for various monitoring data can be calculated using the following formulas:

[0085]

[0086] in, Indicates the first j The ideal point of the monitoring data max This indicates taking the maximum value. f kj Indicates in k The first time the data was collected j Such monitoring data, β 1 indicates a positive indicator. min This indicates taking the minimum value. β 2 indicates a contrarian indicator. Indicates the first j The negative ideal point of the monitoring data.

[0087] S3052: Calculate the dimensional standardization length of various monitoring data.

[0088] Optionally, the dimensionless length of each monitoring data is calculated according to the following formula:

[0089] wherein, L j represents the dimensionless length of the i-th monitoring data, | | represents taking absolute value, j represents the maximum value of the i-th monitoring data, represents the minimum value of the i-th monitoring data. j j

[0090] S3053: According to the dimensionless length of each monitoring data and the maximum weight coefficient, the distance between each monitoring data and the positive ideal point is calculated as the first distance.

[0091] Optionally, the first distance is specifically:

[0092] wherein, D 1 represents the distance between the moment and the positive ideal point, that is, the first distance.

[0093] S3054: According to the dimensionless length of each monitoring data and the maximum weight coefficient, the distance between each monitoring data and the negative ideal point is calculated as the second distance.

[0094] Optionally, the second distance is specifically:

[0095] wherein, D 2 represents the distance between the moment and the negative ideal point, that is, the second distance.

[0096] S3055: According to the first distance and the second distance, the collapse risk index of the tunnel to be predicted is calculated.

[0097] Optionally, the collapse risk index of the tunnel to be predicted is calculated according to the following formula:

[0098] wherein, D represents the collapse risk index of the tunnel to be predicted.

[0099] ​​​In the embodiment of the present application, by calculating the positive ideal point and the negative ideal point for each monitoring data respectively, and using the dimensionless length to unify the different indicators, the deviation degree of the current state from the positive ideal point / negative ideal point is obtained in the form of weighted Euclidean distance, and then the collapse risk index of the tunnel is calculated through the relative closeness formula, which not only considers the optimal / worst state of the positive indicators and the negative indicators, but also eliminates the dimensional differences of the monitoring data, so that the risk index is more sensitive to the key risk factors and less affected by the noise or abnormal values, and the unified quantification and intuitive expression of the multi-dimensional indicators are realized. At the same time, the method can dynamically update the risk index according to the latest monitoring data at any time, provide reliable quantitative labels for the subsequent deep learning model, and significantly improve the scientificity, stability and generalizability of the tunnel risk assessment.

[0100] S4: determining whether the collapse risk index is greater than or equal to the collapse risk index threshold value. If yes, go to S5. Otherwise, return to S2.

[0101] Specifically, when the collapse risk index is greater than or equal to the collapse risk index threshold value, it is determined that the tunnel to be predicted has an abnormality and needs to be further predicted, and when the collapse risk index is less than the collapse risk index threshold value, it is determined that the tunnel to be predicted does not have an abnormality and returns to S1 to reacquire the monitoring data.

[0102] It should be noted that the size of the collapse risk index threshold value can be set by the person skilled in the art according to actual needs, which is not limited in the present application.

[0103] In the embodiment of the present application, by introducing the judgment between the collapse risk index and the collapse risk index threshold value, the conventional monitoring and the deep prediction are connected in series into a two-stage process of "light screening deep analysis", which not only significantly saves the calculation and manpower cost, but also realizes the rapid early warning, dynamic threshold adjustment and system scalability, and comprehensively improves the real-time performance, robustness and safety guarantee level of the tunnel monitoring system.

[0104] S5: constructing a graph data structure according to various monitoring data.

[0105] In the embodiment of the present application, by constructing the graph data structure, the time sequence evolution relationship of the monitoring data and the cross-indicator physical / topological association are organically fused, so that the graph convolution network can fully utilize the multi-layer and multi-dimensional structured information, and the expression ability, robustness and accuracy of the tunnel collapse risk prediction model are significantly enhanced.

[0106] In one possible implementation, S5 specifically includes sub-steps S501 to S503: S501: constructing a plurality of single-layer graph data structures based on single monitoring data respectively.

[0107] Optionally, multiple single-layer graph data structures based on single monitoring data can be constructed according to the following formula:

[0108]

[0109] in, Indicates in k Time and The first time the data was collected j The cosine similarity between monitoring data is the weight of the edges in a single-layer graph data structure. cos ( ) represents the cosine function. F ( ) represents the Fourier transform. Indicates in The first time the data was collected j Such monitoring data, epsilon j Indicates based on the first j The threshold of edges in a single-layer graph data structure constructed from monitoring data. num ( ) indicates the number of edges.

[0110] It should be noted that those skilled in the art can set the threshold value of the edges in a single-layer graph data structure according to actual needs, and this invention does not limit this.

[0111] S502: Construct a multi-layer graph data structure based on the combination of various monitoring data.

[0112] Optionally, a multi-layer graph data structure based on the combination of various monitoring data can be constructed according to the following formula:

[0113] in, Indicates in k The first time the data was collected j The monitoring data and the first The cosine similarity between monitoring data is equivalent to the edge weights in a multi-layer graph data structure. Indicates in k The first time the data was collected Such monitoring data, T This indicates the topology of the monitoring device location.

[0114] S503: Construct a graph data structure based on each single-level graph data structure and multi-level graph data structure.

[0115] Optionally, the graph data structure is as follows:

[0116] in, aThe super-adjacency matrix, that is, the graph data structure is represented.

[0117] In the embodiment of the present application, by constructing a single-layer graph and a multi-layer graph respectively, and splicing the two into a graph data structure, the graph convolutional network can not only aggregate the time series evolution information of each indicator, but also fuse the linkage relationship across indicators when passing messages, thereby suppressing time series noise interference and reducing computational complexity, and providing a multi-dimensional, rich and robust structured feature for the deep learning model, thereby significantly enhancing the accuracy and reliability of the tunnel collapse risk prediction.

[0118] S6: A tunnel collapse risk prediction model is constructed by combining a one-dimensional convolutional neural network and a graph convolutional network.

[0119] It should be noted that the one-dimensional convolutional neural network (1D-CNN) is a deep learning model specially used for processing one-dimensional sequence data, which can effectively capture the time series pattern and local dependency relationship in the sequence by sliding the convolution kernel on the input data. It is usually composed of convolutional layers, pooling layers and fully connected layers, and has the characteristics of parameter sharing and local connection, which reduces the computational complexity and improves the feature extraction capability. 1D-CNN is widely used in time series analysis, speech recognition and signal processing, etc., which can automatically learn the key features in the sequence data, and improve the modeling accuracy and generalization ability.

[0120] It should be noted that the graph convolutional network (Graph Convolutional Network, GCN) is a deep learning model specially used for processing graph structure data, which aggregates the features of each node and its neighbors by "convolution" operation between the nodes and their neighbors, in order to capture the relationship and topological structure information between the nodes in the graph. Unlike traditional convolutional neural networks that can only act on regular grids (such as images), GCN can directly operate on any sparse, non-Euclidean graph, and through multi-layer stacking, the nodes can gradually integrate the features of more distant neighbors, so as to learn the embedding representation of the nodes in the global or local structure, and is widely used in node classification, graph classification, link prediction, etc.

[0121] In the embodiment of the present application, by combining 1D-CNN and GCN, the time series pattern of tunnel monitoring data can be deeply mined, and the structured relationship between multiple indicators and multiple time points can be captured, which are complementary to each other, not only enhancing the adaptability and robustness of the model to multi-source heterogeneous data, but also laying a solid foundation for subsequent attention fusion and deep prediction, thereby greatly improving the accuracy and reliability of the tunnel collapse risk prediction.

[0122] S7: Various monitoring data and graph data structure are input into the tunnel collapse risk prediction model, the tunnel to be predicted is predicted for collapse risk, and the tunnel collapse risk level is output.

[0123] Optionally, the tunnel collapse risk level comprises no tunnel collapse risk, first-level tunnel collapse risk, second-level tunnel collapse risk, and third-level tunnel collapse risk.

[0124] It should be noted that no tunnel collapse risk means that the tunnel is in a completely safe state. First-level tunnel collapse risk means that the tunnel has a slight collapse risk. Second-level tunnel collapse risk means that the tunnel has a moderate collapse risk. Third-level tunnel collapse risk means that the tunnel has a high collapse risk.

[0125] In the embodiments of the present application, by combining the monitoring data and the graph data structure for tunnel collapse risk prediction and dividing the risk level, the present application can provide an accurate, scientific and dynamic risk assessment tool to help managers take appropriate measures according to different risk levels, realize fine management of the tunnel health condition, ensure the safety of tunnel operation, and improve the efficiency of resource allocation and emergency response.

[0126] In one possible implementation, S7 specifically comprises sub-steps S701-S710: S701: input various monitoring data into a one-dimensional convolutional neural network to output a first feature vector.

[0127] S702: input the graph data structure into a graph convolutional network to output a second feature vector.

[0128] S703: splice the first feature vector and the second feature vector to obtain a comprehensive feature vector.

[0129] Optionally, the comprehensive feature vector is specifically:

[0130] wherein, comprehensive feature vector, real set, B total number of monitoring data, OC channel number of the one-dimensional convolutional neural network, M feature dimension, first feature vector, second feature vector.

[0131] S704: calculate the channel attention value through the feedforward neural network according to the comprehensive feature vector.

[0132] It's important to note that a feedforward neural network (FNN) is a fundamental and widely used artificial neural network architecture. It consists of an input layer, one or more hidden layers, and an output layer connected sequentially. Data flows in a single direction within the network, with no feedback or loops. Neurons in each layer process the input signal through weighted summation and activation functions, passing the results to the next layer, extracting higher-order features layer by layer. Feedforward neural networks are characterized by their simple structure, high computational efficiency, and ease of training. They are widely used in tasks such as classification, regression, and feature extraction, and are one of the fundamental frameworks of deep learning models.

[0133] Optionally, the channel attention value is calculated using a feedforward neural network according to the following formula:

[0134] in, e c In a neural network, the first... c Channel attention values ​​for each channel. LeakyReLU ( ) represents the LeakyReLU activation function. This represents the weight matrix of the feedforward neural network. In a neural network, the first... c The combined feature vector of each channel.

[0135] S705: Normalize the attention score to obtain the channel attention coefficient.

[0136] Optionally, the attention score can be normalized according to the following formula to obtain the channel attention coefficient:

[0137] in, α c In a neural network, the first... c Channel attention coefficients for each channel. softmax ( ) represents the softmax activation function. exp ( ) represents an exponential function.

[0138] S706: Based on the channel attention coefficient, the comprehensive feature vector is weighted and fused to obtain the first final feature vector.

[0139] Optionally, the comprehensive feature vectors are weighted and fused according to the following formula to obtain the first final feature vector:

[0140] in, This represents the first final eigenvector.

[0141] S707: determining a second final feature vector according to the first final feature vector through a channel attention mechanism.

[0142] It should be noted that the channel attention mechanism is a method for improving the feature expression ability of a neural network, which aims to adaptively assign weights according to the importance of different channel features. By globally modeling each channel feature, the attention degree of different channels is learned, so that the network can highlight key features and suppress irrelevant or redundant information. The channel attention mechanism is usually realized through three steps of feature aggregation, weight calculation and weighted fusion, which enhances the perception ability of the model to key information and is widely used in image processing, natural language processing and time series data analysis, effectively improving the expression and performance of the model.

[0143] Optionally, the integrated feature vector is weighted and fused according to the following formula to obtain the first final feature vector:

[0144] wherein, indicates the second final feature vector, CAM the channel attention mechanism.

[0145] S708: inputting the first final feature vector into a graph convolution network to output a first initial prediction probability of each tunnel collapse risk level.

[0146] S709: inputting the second final feature vector into a one-dimensional convolutional neural network to output a second initial prediction probability of each tunnel collapse risk level.

[0147] S710: calculating a final prediction probability of each tunnel collapse risk level according to the first initial prediction probability and the second initial prediction probability of each tunnel collapse risk level, and selecting the tunnel collapse risk level with the highest final prediction probability as the output.

[0148] In this embodiment of the invention, by combining 1D-CNN and Graph Convolutional Network (GCN) and introducing a channel attention mechanism, different types of monitoring data can be effectively fused, improving the accuracy and robustness of the tunnel collapse risk prediction model. 1D-CNN is used to process time-series data and capture temporal patterns, while GCN is used to capture the spatial relationships between tunnel monitoring points, learning topological information between nodes through graph convolution operations. In the feature fusion stage, the channel attention mechanism adaptively assigns weights based on the importance of each feature channel, highlighting key features and suppressing redundant information, thereby further enhancing the model's expressive power. Finally, by weighted fusion of feature vectors and combining them with a feedforward neural network, the model can accurately predict the tunnel collapse risk level, providing reliable safety warnings and decision support, and offering a scientific basis for tunnel management and maintenance.

[0149] In one possible implementation, S710 specifically includes sub-steps S7101 to S7106: S7101: The first and second initial prediction probabilities of each tunnel collapse risk level are fused to obtain the fused initial prediction probability.

[0150] Optionally, the initial predicted probabilities are fused as follows:

[0151] in, P k Indicates in k The first time the data was collected j The initial prediction probability of the fusion of various monitoring data This represents the output of a one-dimensional convolutional neural network. k The first time the data was collected j The second initial prediction probability of the monitoring data, This represents the output of the graph convolutional neural network. k The first time the data was collected j The first initial prediction probability of the monitoring data.

[0152] S7102: Normalize the initial prediction probability of the fusion.

[0153] Optionally, the initial prediction probability of the fusion can be normalized according to the following formula:

[0154] in, Indicates the normalized result. k The first time the data was collected j This type of monitoring data belongs to the first n The fusion of initial predicted probabilities for each tunnel collapse risk level Pkjn indicates the first kind of monitoring data collected at the time k indicates the initial fusion prediction probability of the first kind of monitoring data collected at the time j indicates the initial fusion prediction probability of the first kind of monitoring data collected at the time n indicates the initial fusion prediction probability of the first kind of monitoring data collected at the time indicates the total number of tunnel collapse risk levels.

[0155] S7103: Calculate the information entropy according to the normalized initial fusion prediction probability.

[0156] Optionally, the information entropy is calculated according to the following formula:

[0157] wherein, kj indicates the information entropy of the first kind of monitoring data collected at the time k indicates the information entropy of the first kind of monitoring data collected at the time j indicates the information entropy of the first kind of monitoring data collected at the time

[0158] S7104: Normalize the information entropy to obtain the risk weight.

[0159] Optionally, the initial fusion prediction probability is normalized according to the following formula:

[0160] wherein, omega kj indicates the risk weight of the first kind of monitoring data collected at the time k indicates the risk weight of the first kind of monitoring data collected at the time j indicates the risk weight of the first kind of monitoring data collected at the time

[0161] S7105: Calculate the final prediction probability of each tunnel collapse risk level according to the risk weight and the normalized initial fusion prediction probability.

[0162] Optionally, the final prediction probability of each tunnel collapse risk level is calculated according to the following formula:

[0163] wherein, indicates the final prediction probability of the first kind of monitoring data collected at the time k indicates the final prediction probability of the first kind of monitoring data collected at the time n indicates the final prediction probability of the first kind of monitoring data collected at the time

[0164] S7106: Select the tunnel collapse risk level with the highest final prediction probability to output.

[0165] Specifically, each tunnel collapse risk level in the current time is sorted in order of final prediction probability from low to high, and the tunnel collapse risk level with the highest sorting is selected to output.

[0166] In the embodiment of the present application, by combining the prediction results of 1D-CNN and graph convolution network GCN, and using information entropy, normalization processing and risk weight mechanism, the accurate prediction of tunnel collapse risk can be realized. This method optimizes the accuracy and reliability of the prediction results by fusing the prediction results of different models and comprehensively considering the time sequence and spatial characteristics of the tunnel monitoring data. At the same time, the information entropy calculation and risk weight assignment can adaptively adjust the importance of each risk level, improving the model's ability to handle uncertain data. Finally, through the channel attention mechanism and normalization processing, the model can accurately evaluate the risk level of tunnel collapse, providing scientific and quantitative decision support for the safety management of tunnels.

[0167] With reference to the accompanying drawings described in the specification Figure 2 , a structure schematic diagram of a tunnel collapse risk prediction system based on multi-parameter fusion provided by the present application is shown.

[0168] The present application also provides a tunnel collapse risk prediction system 20 based on multi-parameter fusion, applied to the tunnel collapse risk prediction method based on multi-parameter fusion described above, comprising: a processor 201; a memory 202, the memory 202 has computer readable instructions stored thereon, when the computer readable instructions are executed by the processor 201, the tunnel collapse risk prediction method based on multi-parameter fusion as described in the method embodiment is realized.

[0169] The tunnel collapse risk prediction system 20 based on multi-parameter fusion provided by the present application can execute the tunnel collapse risk prediction method based on multi-parameter fusion described above, and achieve the same or similar technical effects. To avoid repetition, the present application will not be described again.

[0170] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: In the embodiment of the present application, by constructing a tunnel collapse risk index system and calculating the collapse risk index of the tunnel to be predicted according to the tunnel collapse risk index system and various monitoring data, the prediction of tunnel collapse risk is no longer dependent on expert experience and traditional engineering analogy method, considering the superposition effect of various factors such as tunnel environment, geological conditions, construction method and supporting structure, it is difficult to comprehensively and accurately reflect the actual risk of the tunnel under different conditions and construction stages. By constructing a graph data structure and inputting various monitoring data and graph data structure into the tunnel collapse risk prediction model, the collapse risk of the tunnel to be predicted is predicted, and the tunnel collapse risk level is output, which no longer depends on a single type of monitoring data, so that the prediction accuracy and real-time performance of the tunnel collapse risk are no longer limited, and the risk of the tunnel can be described comprehensively in complex environment.

[0171] It should be appreciated that a processor in the embodiments of the present application can be a central processing unit (CPU). The processor can also be other general purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general purpose processor can be a microprocessor or the processor can be any conventional processor.

[0172] It should also be appreciated that the memory in the embodiments of the present application can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. The nonvolatile memory can be a read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0173] The above-described embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through a wired (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0174] It should be understood that the term "and / or" herein merely describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents that the associated objects before and after it are in an "or" relationship, but it can also represent an "and / or" relationship, which can be understood according to the context before and after it.

[0175] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0176] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined according to their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0177] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0178] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0179] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0180] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0181] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0182] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0183] The embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is characterized in that when the program is executed by a processor, the method for predicting tunnel collapse risk based on multi-parameter fusion is realized.

[0184] The computer readable storage medium provided by the present application can realize the steps and effects of the method for predicting tunnel collapse risk based on multi-parameter fusion described above. To avoid repetition, the present application will not be described again.

[0185] The technical solutions provided by the embodiment of the present application have at least the following beneficial effects: In the embodiment of the present application, by constructing a tunnel collapse risk index system and according to the tunnel collapse risk index system and various monitoring data, the collapse risk index of the tunnel to be predicted is calculated, the prediction of the tunnel collapse risk is no longer dependent on the expert experience and the traditional engineering analogy method, the superposition effects of various factors such as tunnel environment, geological conditions, construction method and supporting structure are considered, and the actual risk of the tunnel in different conditions and construction stages is difficult to comprehensively and accurately reflect. By constructing a graph data structure and inputting various monitoring data and the graph data structure into a tunnel collapse risk prediction model, the collapse risk of the tunnel to be predicted is predicted, and the tunnel collapse risk level is output. The prediction of the tunnel collapse risk is no longer dependent on a single type of monitoring data, the accuracy and real-time performance of the prediction of the tunnel collapse risk are no longer limited, and the risk of the tunnel can be comprehensively described in a complex environment.

[0186] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0187] The following points need to be explained: (1) The drawings of the embodiments of the present application only involve the structures involved in the embodiments of the present application, and other structures can be referred to the general design.

[0188] (2) For the sake of clarity, the thickness of a layer or region is exaggerated or reduced in the drawings used to describe the embodiments of the present application, that is, the drawings are not drawn according to the actual proportion. It can be understood that when an element such as a layer, a film, a region or a substrate is referred to as being located "on" or "under" another element, the element can be "directly" located "on" or "under" another element or there can be an intermediate element.

[0189] (3) In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other to obtain new embodiments.

[0190] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1.A tunnel collapse risk prediction method based on multi-parameter fusion, characterized in that, The method comprises the following steps: S1: collecting various monitoring data of a tunnel to be predicted; S2: constructing a tunnel collapse risk index system; S3: calculating a collapse risk index of the tunnel to be predicted according to the tunnel collapse risk index system and the various monitoring data; S4: judging whether the collapse risk index is greater than or equal to a collapse risk index threshold value; if yes, entering S5; otherwise, returning to S1; S5: constructing a graph data structure according to the various monitoring data; S6: constructing a tunnel collapse risk prediction model by combining a one-dimensional convolutional neural network and a graph convolutional network; S7: inputting the various monitoring data and the graph data structure into the tunnel collapse risk prediction model to predict the collapse risk of the tunnel to be predicted and output a tunnel collapse risk level. 2.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 1, wherein, The monitoring data comprises tunnel rock mass integrity coefficients, tunnel rock mass quality grades, tunnel longitudinal wave velocities, tunnel lighting brightnesses, tunnel rock mass stresses, tunnel support structure stresses, tunnel spans and heights, tunnel rock mass deformation amounts, tunnel groundwater pressures, tunnel rock mass microseismic characteristics and tunnel temperatures collected by different monitoring devices at multiple time points. The tunnel collapse risk index system comprises multiple tunnel collapse risk indexes and corresponding score intervals of each tunnel collapse risk index. The tunnel collapse risk indexes comprise positive indexes and reverse indexes. The positive indexes comprise the tunnel rock mass integrity coefficients, the tunnel rock mass quality grades, the tunnel longitudinal wave velocities and the tunnel lighting brightnesses. The reverse indexes comprise the tunnel rock mass stresses, the tunnel support structure stresses, the tunnel spans and heights, the tunnel rock mass deformation amounts, the tunnel groundwater pressures, the tunnel rock mass microseismic characteristics and the tunnel temperatures. 3.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 1, characterized in that, S3 specifically comprises the following steps: S301: calculating first weight coefficients of the various monitoring data by a Delphi method according to the score intervals; S302: performing normalization processing on the various monitoring data; S303: calculating second weight coefficients of the various monitoring data by an entropy weight method according to the normalized monitoring data; S304: calculating final weight coefficients of the various monitoring data according to the first weight coefficients and the second weight coefficients; S305: calculating the collapse risk index of the tunnel to be predicted by an ideal point method according to the final weight coefficients of the various monitoring data. 4.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 3, characterized in that, S301 specifically comprises the following steps: S3011: setting an initial number of experts; S3012: scoring the various monitoring data by each expert according to the score intervals; S3013: calculating score mean values of the various monitoring data according to the scoring results of the various monitoring data; S3014: calculating score standard deviations of the various monitoring data according to the score mean values; S3015: calculating an average score standard deviation according to the score standard deviations of the various monitoring data; S3016: judging whether the average score standard deviation is greater than a score standard deviation threshold value; if yes, increasing the number of experts and returning to S3012; otherwise, entering S3017; S3017: calculating the first weight coefficient of each monitoring data according to the score mean. 5.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 3, wherein, The S303 specifically comprises: S3031: calculating the entropy value of each normalized monitoring data; S3032: calculating the coefficient of variation of each normalized monitoring data according to the entropy value; S3033: calculating the importance index of each normalized monitoring data according to the coefficient of variation; S3034: calculating the second weight coefficient of each monitoring data according to the importance index. 6.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 3, wherein, The S305 specifically comprises: S3051: calculating the positive ideal point and the negative ideal point of each monitoring data; S3052: calculating the dimensionless length of each monitoring data; S3053: calculating the distance between each monitoring data and the positive ideal point as a first distance according to the dimensionless length of each monitoring data and the maximum weight coefficient; S3054: calculating the distance between each monitoring data and the negative ideal point as a second distance according to the dimensionless length of each monitoring data and the maximum weight coefficient; S3055: calculating the collapse risk index of the tunnel to be predicted according to the first distance and the second distance. 7.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 1, wherein, The S5 specifically comprises: S501: constructing a plurality of single-layer graph data structures based on single monitoring data respectively; S502: constructing a multi-layer graph data structure based on the combination of each monitoring data; S503: constructing the graph data structure according to each single-layer graph data structure and the multi-layer graph data structure. 8.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 1, wherein, The tunnel collapse risk level includes no tunnel collapse risk, first-level tunnel collapse risk, second-level tunnel collapse risk, and third-level tunnel collapse risk; The S7 specifically comprises: S701: inputting each monitoring data into the one-dimensional convolutional neural network to output a first feature vector; S702: inputting the graph data structure into the graph convolutional network to output a second feature vector; S703: splicing the first feature vector and the second feature vector to obtain a comprehensive feature vector; S704: calculating the channel attention value through the feedforward neural network according to the comprehensive feature vector; S705: normalizing the attention score to obtain a channel attention coefficient; S706: weighting and fusing the comprehensive feature vector according to the channel attention coefficient to obtain a first final feature vector; S707: determining a second final feature vector through the channel attention mechanism according to the first final feature vector; S708: inputting the first final feature vector into the graph convolutional network to output a first initial prediction probability of each tunnel collapse risk level; S709: inputting the second final feature vector into the one-dimensional convolutional neural network to output a second initial prediction probability of each tunnel collapse risk level; S710: According to the first initial prediction probability and the second initial prediction probability of each tunnel collapse risk level, a final prediction probability of each tunnel collapse risk level is calculated, and a tunnel collapse risk level with the highest final prediction probability is output. 9.The multi-parameter fusion based tunnel collapse risk prediction method according to claim 1, wherein, The S710 specifically comprises: S7101: The first initial prediction probability and the second initial prediction probability of each tunnel collapse risk level are fused to obtain a fused initial prediction probability; S7102: The fused initial prediction probability is normalized; S7103: According to the normalized fused initial prediction probability, information entropy is calculated; S7104: The information entropy is normalized to obtain a risk weight; S7105: According to the risk weight and the normalized fused initial prediction probability, a final prediction probability of each tunnel collapse risk level is calculated; S7106: A tunnel collapse risk level with the highest final prediction probability is output. 10.A tunnel collapse risk prediction system based on multi-parameter fusion, characterized in that, It comprises: A processor; A memory, wherein the memory has computer readable instructions stored thereon, and the computer readable instructions are executed by the processor to implement the multi-parameter fusion based tunnel collapse risk prediction method according to any one of claims 1 to 9.

Images