Safety risk evaluation method and system for shield tunneling under existing tunnel

Through the trapezoidal cloud Bayesian model, a Bayesian network model is established, and the digital characteristics and membership of the evaluation index cloud are calculated, which solves the randomness and fuzzy uncertainty of multi-source risk information in the construction of shield underpass existing tunnels. The safety risk assessment of the shield structure underpasses the existing tunnel is realized, and the accuracy of risk perception and construction safety are improved.

CN119026897APending Publication Date: 2024-11-26HUAZHONG UNIV OF SCI & TECH
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410864303.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-30
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Traditional methods are difficult to effectively deal with the randomness and fuzzy uncertainty of multi-source risk information in the construction of shield underpass existing tunnels, resulting in inaccurate safety risk assessment and difficulty in identifying potential threats and taking effective measures.

Method used

Using the method based on the Bayesian model of trapezoidal cloud, the Bayesian network model is established, and the digital characteristics and membership of the evaluation index of trapezoidal cloud model are calculated, and the prior probability of Bayesian network is determined, and pre-inference and in-process sensitivity are carried out. Factor analysis and post-hoc reverse diagnosis realize the safety risk level assessment of the shield underpass to existing tunnels.

Benefits of technology

This method can accurately perceive the impact of multi-source risk factors, improve the credibility of risk assessment, provide full-time-range dynamic management decision support for security risks, reduce information loss, and enhance construction safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026897A_ABST
    Figure CN119026897A_ABST
Patent Text Reader

Abstract

The invention discloses a trapezoidal cloud Bayesian model-based safety risk evaluation method and equipment for a shield to pass through an existing tunnel, and a storage medium. The safety risk evaluation method comprises the following steps of: designing a Bayesian network model; determining a Bayesian network prior probability based on a trapezoidal cloud model; and reasoning and applying the Bayesian network model to determine the safety risk level of the shield tunneling machine passing through the existing tunnel. According to the method provided by the invention, the fuzzy and random uncertainty of the risk information is considered, so that the original data can be directly applied, normalization processing is not needed, information loss is avoided, an accurate security risk evaluation result is obtained, and the credibility of a risk perception result is improved; and meanwhile, pre-reasoning prediction, in-event sensitive factor identification and post-event risk diagnosis are further carried out, and full-time-range effective decision support is provided for dynamic management of safety risks of adjacent buildings in the pre-event stage, the middle stage and the post stage.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of safety risk assessment of a shield machine passing through an existing tunnel, and more specifically, relates to a safety risk assessment method, system, electronic equipment and storage medium for a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model. Background Art

[0002] When the shield tunnel is excavated and passed through, it disturbs the soil, causing soil deformation and stratum loss, which destroys the stability of the surrounding environment, making the underground tunnel construction process full of potential uncertain risks, such as geological conditions, construction methods, environment, management and other multi-source influencing factors, which will pose a potential threat to construction safety. For the construction of shield tunnels under existing tunnels, the construction difficulty and risk will increase due to the existence of existing tunnels. In addition to ensuring the safe penetration of new tunnels, when the shield tunnel is under the existing tunnel, it also involves the structural stability and bearing capacity of the existing tunnel. If the construction risk is too great and there is no timely intervention and control, it may lead to serious safety accidents and cause huge economic losses. Therefore, it is very important to evaluate the construction risks of shield tunnels under existing tunnels. It can help identify and predict potential risks, take measures to control and manage, and thus ensure the safety, efficiency and sustainability of the entire project.

[0003] The complex correlation between influencing factors, the randomness and fuzzy uncertainty of multi-source information put forward high requirements for the risk modeling of shield tunneling under existing tunnel construction system. Traditional methods such as fuzzy hierarchical analysis, cloud model, and grey correlation analysis focus more on subjective judgment, making it difficult to directly use actual construction data for construction risk prediction, and there are great limitations on the processing and utilization of multi-source information on uncertainty of safety risk factors. Therefore, it is extremely necessary to solve the randomness and fuzzy uncertainty of multi-source risk information and conduct a comprehensive evaluation of the safety risk of shield tunneling under existing tunnels. Summary of the invention

[0004] In response to the above defects or improvement needs of the prior art, the present invention provides a safety risk assessment method, equipment and storage medium for a shield machine passing through an existing tunnel based on a trapezoidal Yunbeis model, so as to achieve accurate perception of the safety risk status of a shield machine passing through an existing tunnel under the influence of multi-source risk factors.

[0005] To achieve the above object, the present invention provides a method for evaluating the safety risk of a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model, the method comprising the following steps:

[0006] S1: Establish an indicator system, define risk nodes and topological nodes, and design a Bayesian network model;

[0007] S2: Calculate the digital characteristics and membership of the evaluation index cloud of the trapezoidal cloud model, and determine the Bayesian network prior probability based on the trapezoidal cloud model;

[0008] S3: The Bayesian network model is used for reasoning to determine the safety risk level of shield tunneling under existing tunnels.

[0009] Furthermore, in step S1, the Bayesian network model design includes the following steps:

[0010] S11: Establishment of safety risk assessment index system and division of assessment standards;

[0011] S12: Definition and discretization of risk nodes: The factors and result events involved in the safety risk evaluation index system are defined as nodes of the Bayesian network model, and the three-layer nodes are discretized into five states: no risk, low risk, medium risk, high risk and extremely high risk based on the classification of the safety risk evaluation index system; the nodes include root nodes, intermediate nodes and leaf nodes;

[0012] S13: Topological structure definition and parameter design, wherein the topological structure includes nodes and directed edges connecting the nodes, and the directed edges represent the causal relationship or interaction between the nodes; the parameter design is to design the conditional probability between the nodes to describe the connection relationship between the nodes.

[0013] Furthermore, in step S2, the evaluation index cloud digital features of the trapezoidal cloud model include cloud digital feature expected value, entropy value and super entropy value:

[0014]

[0015] He=k=0.001 (4)

[0016] Among them, Ex 1,ij ,Ex 2,ij are the upper and lower bounds of the expected interval of the j-th risk level cloud model of the ith evaluation index, En is the corresponding entropy value, He is the super entropy value, and R ij,max , R ij,min They are the upper and lower bounds of the interval value of the j-th risk level cloud model of the ith evaluation index;

[0017] Furthermore, in step S2, the calculation of the evaluation index membership of the trapezoidal cloud model includes discretizing the continuous node data through the forward cloud algorithm of the trapezoidal cloud model to obtain the measured data x of the index. i , to obtain x iDegree of certainty of the diagnostic cloud model relative to each risk level:

[0018]

[0019] Furthermore, in step S2, the determination of the Bayesian network prior probability based on the trapezoidal cloud model requires converting the membership of the trapezoidal cloud model of each risk level to the prior probability using the Bayesian network model:

[0020]

[0021] Among them, α represents the consistency parameter. In order to maximize the consistency of certainty and probability conversion, α=1.

[0022] Furthermore, in step S3, the reasoning application of the Bayesian network model includes the following steps:

[0023] Pre-emptive forward prediction reasoning: Divide the risk level into n states, and calculate the posterior probability of the leaf node under the joint action of the root node; when the root node is observed to be in a certain state, use the evidence to update the node state:

[0024] P(S=s)=P(S=s|X1=x1,X2=x2,…,X n =x n )×P(X1=x1,X2=x2,…,X n =x n ) (7)

[0025]

[0026] Among them, X i =x i Represents the root node X i The state, P(S=s|X1=x1,X2=x2,…,X n =x n ) represents the conditional probability distribution of leaf node S, P(X1=x1,X2=x2,…,X n =x n ) represents the joint probability distribution of the root node;

[0027] Analysis of sensitive factors during the incident:

[0028]

[0029] Where C is the value of COD defuzzification using risk probability distribution, A mn Indicates different building safety levels;

[0030] Post-accident reverse diagnostic reasoning: After an accident occurs, reverse diagnostic reasoning is used to determine the posterior probability of each node, and then the most likely cause of the accident is found:

[0031]

[0032] Among them, P(X i =x i The higher the value of |S=s), the closer it is to 1, indicating that the most likely cause of the accident is S=s.

[0033] According to a second aspect of the present invention, a safety risk assessment system for shield tunneling under an existing tunnel based on a trapezoidal cloud Bayesian model is provided, which is characterized by comprising:

[0034] The first main module is used to input the factors affecting the safety risk of shield tunneling under the existing tunnel and establish a safety risk evaluation index system;

[0035] The second main module is used for calculating and processing the cloud digital features of the trapezoidal cloud model and determining the Bayesian network prior probability based on the trapezoidal cloud model;

[0036] The third main module is used to determine and output the safety risk assessment level of shield tunneling under existing tunnels based on the Bayesian network model.

[0037] According to a third aspect of the present invention, there is provided an electronic device comprising a processor, a memory and a communication interface, wherein the processor, the memory and the communication interface communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the safety risk assessment method for a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model according to the first aspect or any corresponding embodiment thereof.

[0038] In addition, to achieve the above-mentioned purpose, according to the third aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the safety risk assessment method for shield tunneling under an existing tunnel based on the trapezoidal cloud Bayesian model of the above-mentioned first aspect or any corresponding embodiment.

[0039] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0040] 1. The safety risk assessment method for shield tunneling under an existing tunnel based on the trapezoidal cloud Bayesian model provided by the present invention takes into account the fuzziness and random uncertainty of risk information, so that the original data can be used directly without normalization processing, avoiding information loss, obtaining accurate safety risk assessment results, and improving the credibility of risk perception results.

[0041] 2. The safety risk assessment method for shield tunneling under an existing tunnel based on the trapezoidal cloud Bayesian model provided by the present invention further carries out pre-inference prediction, in-process sensitive factor identification and post-event risk diagnosis, and provides effective decision-making support over the entire time range for the dynamic management of safety risks of adjacent buildings in the pre-event, mid-event and post-event stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the safety risk assessment method of shield tunneling under an existing tunnel based on trapezoidal cloud Bayes of the present invention;

[0043] Figure 2 A schematic diagram of a shield tunnel optimization design method and equipment structure based on CatBoost-NSGAⅢ provided in an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention;

[0045] Figure 4 A Bayesian network diagram of the safety risk of a shield tunnel under an existing tunnel provided by a trapezoidal cloud Bayesian embodiment of the present invention;

[0046] Figure 5 Provides a priori probability distribution of each risk factor at the Q1 monitoring point for the embodiment of the present invention;

[0047] Figure 6 A local sensitivity map of the Q3 section tunnel provided by an embodiment of the present invention;

[0048] Figure 7 The risk diagnosis analysis of the monitoring point in the Q1 section provided in the embodiment of the present invention: (a) Level IV risk (b) Level V risk. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0050] The present invention provides a method for evaluating the safety risk of a shield tunnel under an existing tunnel based on trapezoidal cloud Bayesian, see Figure 1 The method includes: designing a Bayesian network model; determining the prior probability of the Bayesian network based on the trapezoidal cloud model; and applying the reasoning of the Bayesian network model to determine the safety risk assessment level of the shield tunneling under the existing tunnel.

[0051] Based on the content of the above method embodiment, as an optional embodiment, the embodiment of the present invention provides a safety risk assessment method for shield tunneling under an existing tunnel based on trapezoidal cloud Bayesian. In step S1, the Bayesian network model is designed, and the steps include:

[0052] S11: Establishment of safety risk assessment index system and division of assessment standards;

[0053] Specifically, the safety risk assessment index system for shield tunneling under existing tunnels includes three dimensions: tunnel geometry parameters, geological parameters, and construction factors, covering a total of 12 influencing factors. The safety assessment of shield tunneling under existing tunnels is divided into five levels [I, II, III, IV, V], corresponding to five states: I (no risk), II (low risk), III (medium risk), IV (high risk), and V (extremely high risk). The higher the level, the higher the risk level of the evaluation index and the worse the safety situation, as shown in Table 1.

[0054] In one embodiment of the present invention, eight construction sections between Julong Avenue Station and Tang Yunhai Station of Wuhan Metro Line 7 are taken as the research objects of underpass construction, and an investigation and analysis are conducted. The actual data of the evaluation indicators of the eight construction sections are obtained according to Table 1, as shown in Table 2.

[0055] Table 1 Safety risk assessment index system and value range of shield tunneling under existing tunnels

[0056]

[0057]

[0058] Table 2 Evaluation index data acquisition of adjacent buildings along the tunnel

[0059]

[0060]

[0061] S12: Definition and discretization of risk nodes: The factors and result events involved in the safety risk evaluation index system are defined as nodes of the Bayesian network model, and the three-layer nodes are discretized into five states: no risk, low risk, medium risk, high risk and extremely high risk based on the level of the safety risk evaluation index system; the nodes include root nodes, intermediate nodes and leaf nodes;

[0062] Specifically, the safety risk evaluation index system of shield tunneling under existing tunnels defines tunnel safety as the leaf node S, tunnel geometric parameters R1, geological parameters R2 and construction factors R3 as intermediate nodes, and each index factor X1~X12 under the three types of factors as the root node. The topological structure of the Bayesian network of the safety risk of shield tunneling under existing tunnels is determined, see Figure 4Furthermore, according to the classification results of the safety risk evaluation index system of shield tunneling under existing tunnels, the states of each node in the Bayesian network are divided into five corresponding states: I (no risk), II (low risk), III (medium risk), IV (high risk), and V (extremely high risk). Based on the engineering safety knowledge of domain experts, a conditional probability table CPT associating parent nodes and child nodes is established. Taking the leaf node S as an example, its partial CPT is shown in Table 3.

[0063] Table 3 Part of the conditional probability table CPT

[0064]

[0065] S13: Topological structure definition and parameter design, wherein the topological structure includes nodes and directed edges connecting the nodes, and the directed edges represent the causal relationship or interaction between the nodes; the parameter design is to design the conditional probability between the nodes to describe the connection relationship between the nodes.

[0066] Furthermore, in step S2, the Bayesian network prior probability determination based on the trapezoidal cloud model includes the following steps:

[0067] S21: Calculate the evaluation index cloud digital features of the trapezoidal cloud model, wherein the evaluation index cloud digital features include the cloud digital feature expected value, entropy value and super entropy value:

[0068]

[0069] He=k=0.001 (4)

[0070] Among them, Ex 1,ij ,Ex 2,ij are the upper and lower bounds of the expected interval of the j-th risk level cloud model of the ith evaluation index, En is the corresponding entropy value, He is the super entropy value, and R ij,max , R ij,min They are the upper and lower bounds of the interval value of the j-th risk level cloud model of the ith evaluation index;

[0071] S22: Apply the forward cloud algorithm of the trapezoidal cloud model to discretize the continuous node data and obtain the measured data x of the indicator i , find x i Certainty of the diagnostic cloud model relative to each risk level:

[0072]

[0073] S23: Convert the obtained certainty of the diagnostic cloud model for each risk level into a priori probability using the Bayesian network model:

[0074]

[0075] Among them, α represents the consistency parameter. In order to maximize the consistency of certainty and probability conversion, α=1.

[0076] Substitute the indicator observation data of the 8 construction sections in Table 2 into the trapezoidal cloud model, and obtain the cloud digital features of the trapezoidal cloud model according to formulas (1)-(4). Take construction section Q1 as an example for calculation, and the results are shown in Table 4. According to formula (5), the certainty of the evaluation indicators in different construction sections for each risk level is calculated, and then according to formula (6), the certainty is converted into a priori probability in the Bayesian network. The prior probability is shown in Figure 5 . So that it can be input into the constructed Bayesian model for probabilistic reasoning and prediction.

[0077] Table 4 Trapezoidal cloud model eigenvalues ​​of risk factors

[0078]

[0079] Further, in step S3, the Bayesian network model is used for reasoning to determine the safety risk assessment level of the shield tunneling under the existing tunnel, and the steps include:

[0080] 1) Ex ante risk reasoning. The prior probability is input into the Bayesian network of the safety risk of shield tunneling under the existing tunnel, and the probability distribution of each indicator is used as the prior probability corresponding to each root node.

[0081] In one embodiment of the present invention, the actual data of the tunnel is iterated 200 times in the trapezoidal cloud model, and the 200 groups of membership are converted into distribution probabilities and the safety risk status is obtained and input into the Bayesian model. The reasoning process is repeated to obtain the probability distribution of each risk level state of the intermediate nodes B1~B3 and the tunnel safety S. The probability of the tunnel safety risk state of each construction section is obtained by probabilistic aggregation of the 200 iteration results using equations (7) and (8) as shown in Table 5. The final risk level is determined by the maximum probability.

[0082] Table 5 Probability distribution of tunnel safety risk status in eight construction sections

[0083]

[0084] 2) In-process sensitivity analysis: defuzzify the probability distribution of the event structure to obtain accurate decision-making values.

[0085] In one embodiment of the present invention, one of the buildings is selected for further analysis. The probability distribution of the 200 iteration results is defuzzified according to formula (9), and the probability distribution is converted into a numerical value. Then, the local sensitivity of the indicators X1-X12 to the safety status of the Q3 section tunnel is calculated by data analysis, and the sensitivity results of each factor are obtained as follows: Figure 6 As shown. Figure 6 As shown in the figure, the three most sensitive factors affecting the high risk (IV) state of the Q3 monitoring point are R1=IV|X1=V,X2=I,X3=IV,X4=IV; S=IV|R1=IV,R2=IV,R3=II; R3=II|X10=IV,X9=II,X12=II,X11=III. That is to say, when the child node under the given parent node state is at the corresponding level, it has a greater impact on the target node S, and should be the focus of safety management attention, and should be monitored in real time, so as to take timely measures to control and prevent the tunnel safety risk from being too high.

[0086] 3) Post-event risk diagnosis. Bayesian networks can not only judge the safety status of buildings through forward reasoning, but also use the reverse reasoning technology of Bayesian networks to calculate the posterior probability of each risk factor when the building's safety status is relatively dangerous, thereby diagnosing the key factors that are most likely to cause the building's dangerous state to be too high. According to the risk diagnosis results, the risk causes can be quickly and targeted to prevent the further deterioration of the tunnel's safety status. Assuming that the tunnel safety status is level IV (relatively dangerous) and level V (dangerous), taking the Q1 section monitoring point as an example, the posterior probability distribution of each risk factor is calculated according to formula (10) when P(T=IV)=1 and P(T=V)=1, as follows: Figure 7 shown.

[0087] from Figure 7 It can be found that when the tunnel safety status is level IV (high risk) and level V (extremely high risk), the most important factors leading to safety damage in shield tunneling are X1 (clearance between the new and old tunnels) and X3 (thickness of the existing tunnel cover). Figure 7 The results show that the posterior probability distribution of each factor in the two cases is basically the same, and the relative fluctuation of the probability distribution of the factors is not much different. In both cases, the probability of X1 and X3 being at level IV and level V is the highest, close to 1. It can be seen that X1 and X3 are the most critical factors that lead to excessive risk at the Q1 monitoring point and need more attention. Secondly, the posterior probability of X8 (Poisson's ratio) at level V is between 0.33 and 0.41, which may also be one of the factors that damage construction safety and induce dangerous accidents. In the process of risk diagnosis, these three factors should be focused on and continuously tracked and controlled until the tunnel safety risk is better controlled.

[0088] Figure 2 A schematic diagram of a shield tunnel optimization design method and equipment structure based on CatBoost-NSGAⅢ provided in an embodiment of the present invention; Figure 2As shown, another embodiment of the present invention also provides a safety risk assessment system for shield tunneling underpasses of existing tunnels based on the trapezoidal cloud Bayesian model, comprising: a first main module, used to input influencing factors of the safety risk of shield tunneling underpasses of existing tunnels, and establish a safety risk assessment index system; a second main module, used to calculate and process the cloud digital features of the trapezoidal cloud model, and determine the Bayesian network prior probability based on the trapezoidal cloud model; and a third main module, used to determine and output the safety risk assessment level of the shield tunneling underpasses of existing tunnels based on the Bayesian network model.

[0089] Figure 3 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Figure 3 As shown, an embodiment of the present invention also provides an electronic device, characterized in that it includes a processor, a memory and a communication interface, wherein the processor, the memory and the communication interface communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the safety risk assessment method for shield tunneling under an existing tunnel based on the trapezoidal cloud Bayesian model described in the above embodiment.

[0090] In an embodiment of the present invention, a non-transitory computer-readable storage medium is also provided for implementing all the steps in the method for risk assessment of shield tunneling based on the trapezoidal cloud Bayesian model in the above embodiment. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute all the steps in the method for safety risk assessment of shield tunneling based on the trapezoidal cloud Bayesian model in the above embodiment.

[0091] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented with the help of computing software and equipment. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be reflected in the form of calculation results, and the trapezoidal cloud Bayesian model is used to calculate and evaluate the safety risk status of the shield tunneling under the existing tunnel.

[0092] The flowchart in the accompanying drawings shows the architecture, functions and operations that may be implemented by the device system and method according to multiple embodiments of the present invention. Based on this understanding, each box in the flowchart can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and sometimes in the opposite order, depending on the functions involved.

[0093] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A safety risk assessment method for shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model, characterized in that: The following steps are involved: S1: Establish an indicator system, define risk nodes and topological nodes, and design a Bayesian network model; S2: Calculate the digital characteristics and membership of the evaluation index cloud of the trapezoidal cloud model, and determine the Bayesian network prior probability based on the trapezoidal cloud model; S3: The Bayesian network model is used for reasoning to determine the safety risk level of shield tunneling under existing tunnels.

2. The method for evaluating the safety risk of a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model according to claim 1 is characterized in that: In step S1, the Bayesian network model design includes the following steps: S11: Establishment of safety risk assessment index system and division of assessment standards; S12: Definition and discretization of risk nodes: The factors and result events involved in the safety risk evaluation index system are defined as nodes of the Bayesian network model, and the three-layer nodes are discretized into five states: no risk, low risk, medium risk, high risk and extremely high risk based on the classification of the safety risk evaluation index system; the nodes include root nodes, intermediate nodes and leaf nodes; S13: Topological structure definition and parameter design, wherein the topological structure includes nodes and directed edges connecting the nodes, and the directed edges represent the causal relationship or interaction between the nodes; the parameter design is to design the conditional probability between the nodes to describe the connection relationship between the nodes.

3. The method for safety risk assessment of shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model according to claim 1 is characterized in that: In step S2, the evaluation index cloud digital features of the trapezoidal cloud model include cloud digital feature expected value, entropy value and super entropy value: He=k=0.001 (4) Among them, Ex 1,ij ,Ex 2,ij are the upper and lower bounds of the expected interval of the j-th risk level cloud model of the ith evaluation index, En is the corresponding entropy value, He is the super entropy value, and R ij,max , R ij,min They are respectively the upper and lower bounds of the interval value of the j-th risk level cloud model of the ith evaluation indicator.

4. The method for safety risk assessment of shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model according to claim 1 is characterized in that: In step S2, the calculation of the evaluation index membership of the trapezoidal cloud model includes discretizing the continuous node data through the forward cloud algorithm of the trapezoidal cloud model to obtain the measured data x of the index. i , find x i Certainty of the diagnostic cloud model relative to each risk level:

5. The method for safety risk assessment of shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model according to claim 1 is characterized in that: In step S2, the determination of the Bayesian network prior probability based on the trapezoidal cloud model requires converting the membership of the trapezoidal cloud model of each risk level to the prior probability using the Bayesian network model: Among them, α represents the consistency parameter. In order to maximize the consistency of certainty and probability conversion, α=1.

6. The method for safety risk assessment of shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model according to claim 1 is characterized in that: In step S3, the reasoning application of the Bayesian network model includes the following steps: Pre-emptive forward prediction reasoning: Divide the risk level into n states, and calculate the posterior probability of the leaf node under the joint action of the root node; when the root node is observed to be in a certain state, use the evidence to update the node state: P(S=s)=P(S=s|X1=x1,X2=x2,...,X n =x n )×P(X1=x1,X2=x2,…,X n =x n ) (7) Among them, X i =x i Represents the root node X i The state, P(S=s|X1=x1,X2=x2,…,X n =x n ) represents the conditional probability distribution of leaf node S, P(X1=x1,X2=x2,…,X n =x n ) represents the joint probability distribution of the root node; Analysis of sensitive factors during the incident: Where C is the value of COD defuzzification using risk probability distribution, A mn Indicates different building safety levels; Post-accident reverse diagnostic reasoning: After an accident occurs, reverse diagnostic reasoning is used to determine the posterior probability of each node, and then the most likely cause of the accident is found: Among them, P(X i =x i The higher the value of |S=s), the closer it is to 1, indicating that the most likely cause of the accident is S=s.

7. A safety risk assessment system for shield tunneling under existing tunnels based on trapezoidal cloud Bayesian model, characterized in that: include: The first main module is used to input the factors affecting the safety risk of shield tunneling under the existing tunnel and establish a safety risk evaluation index system; The second main module is used for calculating and processing the cloud digital features of the trapezoidal cloud model and determining the Bayesian network prior probability based on the trapezoidal cloud model; The third main module is used to determine and output the safety risk assessment level of shield tunneling under existing tunnels based on the Bayesian network model.

8. An electronic device, characterized in that: It includes a processor, a memory and a communication interface, wherein the processor, the memory and the communication interface communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the safety risk assessment method for a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model as described in any one of claims 1 to 4.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the safety risk assessment method for a shield machine passing through an existing tunnel based on a trapezoidal cloud Bayesian model as described in any one of claims 1 to 4.

Citation Information

Cited By

  • Building structure safety assessment method and system based on artificial intelligence

    CN121881116A

  • Artificial intelligence-based building structure safety evaluation method and system

    CN121881116B