A probability model construction method and system for tunnel engineering geological risk discrimination

Abstract

The application discloses a kind of probability model construction method and system for tunnel engineering geology risk discrimination, it is related to data analysis processing technical field.The following steps are included:S1, according to the preset paragraph length along tunnel axis, tunnel is divided into several sections, in each section, at least one method is used in drilling while drilling measurement, machine vision scanning, advanced geological prediction, the surrounding rock characteristic parameter of the section is obtained.The present application simultaneously establishes the prior probability model of three main geological risks of collapse, gushing water and surrounding rock large deformation, and the overall probability of at least one disaster is calculated through the correlation coefficient correction joint probability formula, compared with the prior art only for single gushing water risk discrimination, the coupling effect of multiple risks coexistence can be quantified, the comprehensive danger degree of paragraph is truly reflected, and the problem that single risk assessment easily leads to insufficient measures is solved.

Description

Technical Field

[0001] This invention relates to the field of data analysis and processing technology, specifically to a method and system for constructing a probabilistic model for judging geological risks in tunnel engineering. Background Technology

[0002] Tunnel engineering traverses complex geological conditions, facing various geological risks during construction, such as surrounding rock instability and groundwater inrush. Accurately identifying the risk level and optimizing the support scheme accordingly is crucial to ensuring construction safety and controlling project costs.

[0003] For example, Chinese invention patent application publication number CN119005705A discloses a multi-source data fusion method and device for comprehensive prediction of tunnel inrush water inrush. This method acquires multiple tunnel inrush water inrush risk assessment indicators from the tunnel construction site and calculates the weight of each indicator. It then fuses advanced geological forecast information and monitoring measurement data from the tunnel construction site to establish a prediction model for inrush water inrush risk level and probability. Finally, it outputs the actual inrush water inrush risk level and probability at the tunnel construction site and makes a prediction when either the actual inrush water inrush risk level or probability meets certain prediction conditions. This solves the problems in related technologies, such as overly singular evaluation data sources, inability to accurately identify significant influencing factors leading to inrush water accidents, and low prediction accuracy.

[0004] Existing forecasting methods still have certain shortcomings. For example, they only assess the single risk of sudden water inrush and do not consider the simultaneous occurrence or mutual induction of multiple risks such as collapse and large deformation of surrounding rock. They cannot output the combined probability of multiple risks, resulting in an overall low risk assessment. In addition, in actual engineering, the choice of support level directly affects the probability of risk occurrence, but existing technologies only output the actual risk probability and do not establish a quantitative relationship of how the risk probability will change if different support levels are used, thus failing to provide optimal support schemes for construction decisions. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for constructing a probabilistic model for judging geological risks in tunnel engineering, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for constructing a probabilistic model for geological risk assessment in tunnel engineering, comprising the following steps:

[0007] S1. Divide the tunnel into several sections along the tunnel axis according to the preset section length. In each section, use at least one of the following methods to obtain the surrounding rock characteristic parameters of the section: drilling while drilling measurement, machine vision scanning, and advanced geological prediction.

[0008] Among them, the characteristic parameters of the surrounding rock include: rock quality index, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress and special geological body indication;

[0009] S2. Establish a prior risk probability function under unsupported conditions, and calculate the prior probability of various geological risk types respectively. The geological risk types include at least landslide, sudden water inrush and large deformation of surrounding rock.

[0010] S3. Based on the surrounding rock characteristic parameters collected in the section, determine the support level recommended in the construction specification document, and collect the support level to be applied in the section, and calculate the support matching degree.

[0011] S4. For each type of geological risk, calculate the risk reduction coefficient of the support scheme and the posterior risk probability of the support scheme based on the proposed support level and support matching degree.

[0012] S5. Based on the posterior risk probabilities of the three risks, calculate the joint probability of the three risks.

[0013] S6. Based on the posterior risk probability and joint probability of each individual risk, determine the risk level according to the preset threshold range, and output the risk determination result and support optimization suggestions.

[0014] Furthermore, in step S2, the expression for the prior risk probability function is:

[0015] Where i represents three types of geological risk, i=1 represents landslide, i=2 represents sudden water inrush, i=3 represents large deformation of surrounding rock; X1-X6 are rock quality indicators, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress and special geological body indication, respectively. When a special geological body exists, X6 takes the value of 1, and when there is no special geological body, X6 takes the value of 0. and - These are fixed constants corresponding to the geological risk type.

[0016] Furthermore, in step S3, the formula for calculating the support matching degree is:

[0017] ,in, The proposed support level is... The recommended support levels are specified in the standard documents.

[0018] Furthermore, in step S4, the formula for calculating the reduction coefficient is:

[0019] ,in, Let be the sensitivity coefficient of the i-th type of risk to the support structure, obtained through numerical simulation and engineering inversion. The value is 1.8. The value is 0.9. The value is 1.2. The highest support level is represented by a value of 5.

[0020] Furthermore, in the formula for calculating the reduction coefficient, The function is defined as:

[0021] .

[0022] Furthermore, the formula for calculating the posterior risk probability is as follows:

[0023] ;

[0024] in, , When the geological body is special, the weakening coefficient of the i-th type of risk on the support effect is... The value is 0.35. The value is 0.3. The value is 0.25.

[0025] Furthermore, in step S5, the formula for calculating the joint probability is:

[0026] ,in, The correlation coefficient between risks ranges from 0 to 0.5.

[0027] Furthermore, the risk level is determined according to the following rules:

[0028] If 0 ≤ P < 0.01, it is considered risk-free;

[0029] If 0.01 ≤ P < 0.15, the risk level is determined to be low, and a green alert is issued.

[0030] If 0.1 ≤ P < 0.4, the risk level is determined to be medium, and a yellow warning is issued.

[0031] If 0.4 ≤ P, it is judged as high risk, and a red alert is issued;

[0032] Where P is the posterior risk probability of any single risk or the joint probability of the three risks;

[0033] When P ≥ 0.4, the required support level needs to be increased. To obtain a new support level , ;

[0034] Specifically, when 0.4 ≤ P < 0.6, k takes the value of 0.3; when P ≥ 0.6, k takes the value of 0.5. Round up to the nearest integer.

[0035] A probabilistic model construction system for identifying geological risks in tunnel engineering, applied in the aforementioned probabilistic model construction method, includes:

[0036] The data acquisition module collects and stores the corresponding surrounding rock characteristic parameters for each section;

[0037] The prior probability calculation module is used to establish the prior risk probability function under unsupported conditions and calculate the prior probability of various geological risks.

[0038] The posterior probability calculation module is used to calculate the risk reduction coefficient by combining the support matching degree, and then obtain the posterior risk probability of various risks.

[0039] The joint probability calculation module is used to calculate the joint probability of multiple risks by combining the posterior probability of various risks with the risk correlation coefficient.

[0040] The risk assessment output module is used to classify risk levels according to preset thresholds and output assessment results and support optimization suggestions.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] This invention presents a probabilistic model construction method and system for identifying geological risks in tunnel engineering. It establishes prior probability models for three major geological risks: collapse, water inrush, and large deformation of surrounding rock. It calculates the overall probability of at least one disaster by using a joint probability formula with correlation coefficient correction. Compared with existing technologies that only identify a single water inrush risk, this invention can quantify the coupling effect when multiple risks coexist, truly reflect the comprehensive danger level of a section, and solve the problem that single risk assessment can easily lead to insufficient measures.

[0043] Meanwhile, by introducing support matching degree, risk reduction coefficient and matching degree correction factor, a complete quantitative chain is established from surrounding rock characteristics to recommended support level, and then to the posterior risk probability under different actual support levels, so that users can compare the risks under different support levels and thus choose the optimal solution that balances economy and safety.

[0044] Furthermore, this invention also introduces a weakening factor to reduce the reduction coefficient for special geological bodies such as faults and karst, making the posterior risk probability closer to the actual engineering situation, avoiding the dangerous bias of overestimating the support effect under intact surrounding rock conditions, and improving the accuracy of risk assessment. Attached Figure Description

[0045] Figure 1 This is a flowchart of the method of the present invention;

[0046] Figure 2 This is a graph showing the effect of the support level on the risk reduction coefficient in this invention.

[0047] Figure 3 This is a joint probability surface diagram of the three risks in this invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] like Figures 1-3 As shown, this invention provides a technical solution: a method for constructing a probabilistic model for geological risk assessment in tunnel engineering, referring to... Figure 1 This includes the following steps:

[0050] S1. The tunnel is divided into several sections along the tunnel axis according to the preset segment length. In each section, the mean value of the surrounding rock characteristic parameters is obtained using at least one method among drilling-while-drilling measurement, machine vision scanning, and advanced geological prediction. These surrounding rock characteristic parameters include: rock quality indicators, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress, and indicators of special geological bodies. It should be noted that due to the significant spatial variability of geological conditions along the tunnel, this scheme adopts a segmented evaluation strategy. The segment length is determined based on pre-explored geological boundaries such as changes in surrounding rock grade, fault zone boundaries, and lithological boundaries. In sections with stable surrounding rock grades, the segment length is 50m-100m. In fault fracture zones, karst development areas, or sections with frequent changes in surrounding rock grade, the segment length is... With the tunnel length shortened to 20m-30m, in tunnel engineering practice, the aforementioned surrounding rock characteristic parameters can all be obtained through mature existing technologies. Specifically: rock quality indicators are obtained through borehole core sampling statistics; joint volume is calculated through geological sketching of the tunnel face or 3D laser scanning and machine vision recognition technology; rock mass integrity coefficient is calculated after obtaining the longitudinal wave velocity of the rock mass and rock blocks through sonic logging; groundwater seepage flow is determined through borehole pressure water test, pumping test, and on-site triangular weir measurement; tunnel wall tangential stress is converted through geostress measurement technologies such as water pressure fracturing method and hollow inclusion strain relief method; special geological body indicators are identified through comprehensive advanced geological prediction methods such as TSP seismic wave method, ground-penetrating radar method, and advanced drilling. The acquisition technologies of the above parameters have been widely used in this field.

[0051] S2. Establish the prior risk probability function under unsupported conditions, and calculate the prior probabilities of various geological risk types. In this scheme, the geological risk types include at least landslide, sudden water inrush, and large deformation of the surrounding rock. The expression of the prior risk probability function in this step is:

[0052] Where i represents three types of geological risk, i=1 represents landslide, i=2 represents sudden water inrush, i=3 represents large deformation of surrounding rock; X1-X6 are rock quality indicators, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress and special geological body indication, respectively. When a special geological body exists, X6 takes the value of 1, and when there is no special geological body, X6 takes the value of 0. The special geological body is a fault or karst within a detection depth of less than 25m. This is the intercept corresponding to the geological risk type. - The regression coefficients corresponding to the geological risk types were obtained by collecting no fewer than 100 tunnel engineering cases covering three types of disasters: landslides, sudden water inrushes, and large deformations of surrounding rock. Each case included six surrounding rock characteristic parameters for that section. For each type of risk i, a Logistic regression model was established with the occurrence of the disaster as the dependent variable. The intercept and regression coefficients that maximized the likelihood function were solved using the iterative weighted least squares method. Insignificant variables were eliminated through stepwise regression, and only significant variables were retained in the final model.

[0053] In a specific embodiment of this solution, the following representative values ​​were obtained through core drilling, sonic logging, water pressure testing, and geostress measurement: X1=45%, X2=12 lines / m3, X3=0.42, X4=25L(min⋅10m), X5=6.2MPa, X6=0.

[0054] The specific coefficients are shown in the table below:

[0055]

[0056] Substituting the above data into the prior risk probability function, we obtain... That is, the probability of collapse under unsupported conditions is 20.1%; That is, the prior probability of a sudden water inrush is 32.1%. The prior probability of large deformation is 0.3%. The reason for eliminating insignificant variables is that the dominant controlling factors of different geological hazards are different. Specifically, landslides are mainly controlled by rock quality indicators, joint volume, rock mass integrity coefficient, groundwater seepage flow, and special geological body indicators, while the tangential stress of the tunnel wall has no significant impact on landslides in ordinary deep tunnels; sudden water inrushes are mainly controlled by rock quality indicators, groundwater seepage flow, and special geological body indicators, while the joint volume, rock mass integrity coefficient, and tangential stress of the tunnel wall have no significant impact on sudden water inrushes in ordinary deep tunnels; large deformation of the surrounding rock is mainly controlled by rock quality indicators, rock mass integrity coefficient, tangential stress of the tunnel wall, and special geological body indicators, while the joint volume and groundwater seepage flow have no significant impact on large deformation of the surrounding rock in ordinary deep tunnels.

[0057] S3. Based on the surrounding rock characteristic parameters collected in the section, determine the support level recommended in the construction specification document, and collect the support level to be applied in the section, and calculate the support matching degree.

[0058] Specifically, based on the formula 100 + 3 × X1 + 250 × X3, the basic quality index of the surrounding rock is calculated to be 340. According to the "Highway Tunnel Design Code," this is determined to be Grade 3 surrounding rock (the highest being Grade 5). Therefore, the code recommends the following support level. The rating is 3, while the support level that the construction unit intends to apply is... If the value is 2, then the formula for calculating the support matching degree is... The calculated value of M is 0.667.

[0059] S4. For each type of geological risk, based on the proposed support level and support matching degree, calculate the risk reduction coefficient of the support scheme and the posterior risk probability of the support scheme. The formula for calculating the reduction coefficient is as follows:

[0060] ,in, The sensitivity coefficient of the i-th type of risk to the support is obtained through numerical simulation and engineering inversion, such as the FLAC3D model and the strength reduction method. The value ranges from 1.6 to 2.0. The value ranges from 0.7 to 1.0. The value ranges from 1.0 to 1.4. In one specific implementation of this scheme, The value is 1.8. The value is 0.9. The value is 1.2. The highest support level is represented by a value of 5. The matching degree correction factor, whose piecewise function is obtained from finite element parameter analysis, specifically: ,in, To ensure support matching, The incremental form reflects how strong support can further enhance risk reduction.

[0061] Reference Figure 2 It is a curve showing the impact of support level on risk reduction coefficient. In a specific embodiment of this scheme, when M=0.667, then... =0.9, =2, =0.4, therefore: , and .

[0062] Calculate the posterior risk probability for each risk using the formula for calculating posterior risk probability:

[0063] ;in, , When the geological body is special, the weakening coefficient of the i-th type of risk on the support effect is... The value ranges from 0.25 to 0.45. The value ranges from 0.2 to 0.35. The value ranges from 0.15 to 0.3. This indicates a special geological body, and its value is either 0 or 1. For the prior risk probability of risk type i, in this scheme... The value is 0.35. The value is 0.3. The value is 0.25. What we do know is that... The core reason for this introduction is that when tunnels pass through faults or special karst geological formations, the actual effectiveness of conventional support measures will be significantly reduced. If the reduction coefficient is still calculated based on the intact surrounding rock conditions, the ability of the support to suppress risks will be seriously overestimated, leading to a risk assessment result that is biased towards danger.

[0064] In one specific implementation of this plan, it is assumed that the tunnel does not pass through any special geological formations. =0, =1, calculated to get , , Posterior probabilities show that after applying Level 2 support, the risk of collapse decreased from 20.1% to 10.8%, the risk of sudden water inrush decreased from 32.1% to 23.4%, and the risk of large deformation decreased from 3% to 2%, demonstrating the risk reduction effect of support on risky water.

[0065] To verify The importance of this is given by the assumption that the tunnel passes through a special geological formation. The value is 1, obtained through calculation. , , It is evident that special geological bodies significantly weaken the support effect, leading to a marked increase in each type of risk. This indicates that special geological bodies significantly weaken the support effect, making risk assessment more consistent with engineering practice.

[0066] S5. Based on the posterior risk probabilities of the three risks, calculate the joint probability of the three risks. The formula for calculating the joint probability is as follows:

[0067] ,in, The correlation coefficient between risks is calculated based on statistics from no fewer than 100 multi-hazard cases, with a value ranging from 0 to 0.5. The posterior risk probability for each risk, specifically in this scheme, .

[0068] The calculation yielded: , , , , , Finally, the calculation yielded... =0.283, the joint probability of the three risks is referenced. Figure 3 The surface diagram shown.

[0069] S6. Based on the posterior risk probability and joint probability of each individual risk, determine the risk level according to the preset threshold range, and output the risk determination result and support optimization suggestions.

[0070] In this plan, the risk level is determined according to the following rules:

[0071] If 0 ≤ P < 0.01, it is considered risk-free;

[0072] If 0.01 ≤ P < 0.15, the risk level is determined to be low, and a green alert is issued.

[0073] If 0.15 ≤ P < 0.4, the risk level is determined to be medium, and a yellow warning is issued.

[0074] If 0.4 ≤ P, it is judged as high risk, and a red alert is issued;

[0075] For low-risk situations, the recommended support optimization includes at least the following: maintain the current support design, construct normally according to specifications, and no additional measures are required. For medium-risk situations, the recommended support optimization includes at least the following: increase the frequency of on-site monitoring, increase the density of system anchor bolts or adjust the thickness of shotcrete, increase radial grouting points if the dominant risk is sudden water inrush, and reduce the cycle advance length if the dominant risk is collapse or large deformation. For high-risk situations, the recommended support optimization is: stop construction and organize experts to conduct a special risk assessment.

[0076] In the rule-based discrimination, P represents the threshold range in which the posterior risk probability of any single risk or the joint probability of the three risks falls. Specifically, in the specific implementation of this scheme, the posterior probability of landslide (10.8%) is in the low-risk range; the posterior probability of sudden water inrush (23.4%) is in the medium-risk range; the posterior probability of large deformation (0.2%) is in the no-risk range, and the system does not consider it when discriminating and outputting risks; the joint probability (28.3%) is greater than any single probability and is in the medium-risk range.

[0077] In this approach, when individual risk probabilities are independent, the coupling effect of multiple risks coexisting may be overlooked, potentially underestimating the overall risk level of a section. For example, if the probability of landslide is in the low-risk range and the probability of water inrush is in the medium-risk range, judging them separately might lead to the mistaken belief that the overall risk is controllable, and that only medium-risk local grouting and monitoring measures for water inrush would suffice. However, simply recognizing that the landslide risk is in the low-risk range ignores its potential escalation. After a water inrush occurs, it may escalate into a medium-risk event due to softening of the surrounding rock and lubrication of the structural surface. Furthermore, the possibility of the two disasters mutually aggravating each other cannot be expressed by a single probability. Therefore, the conclusion of a medium-risk level with a combined probability of 28.3% more objectively reflects the true overall risk of at least one disaster occurring. This suggests that on-site management should not only focus on controlling water inrush but also simultaneously strengthen landslide prevention and control to achieve coordinated risk management and avoid insufficient or omitted measures due to isolated judgments.

[0078] When P ≥ 0.4, the required support level needs to be increased. To obtain a new support level , ,in, For the highest support level, the value is 5. k is an empirical coefficient; when 0.4 ≤ P < 0.6, k is 0.3; when P ≥ 0.6, k is 0.5. Round up to the nearest integer. For example... If the value of k is 2, and the probability of any single risk or joint risk is greater than or equal to 0.4 and less than 0.6, then k is assigned a value of 0.3. Rounding up to the nearest integer, the result is 3. Therefore, the recommended support level is level 3. Based on the recommended support level, the posterior risk probability and joint probability are recalculated, and the updated risk assessment result and support optimization suggestions are output. If the probability still does not meet the requirement (P≥0.4), the above process is repeated to adjust and improve the support level until the requirement of P<0.4 is met, and the final support optimization suggestions are output. Thus, the probabilistic model for geological risk assessment in tunnel engineering is completed. The model comprehensively considers the differences of different risk-dominant factors, the weakening effect of special geological bodies on support, and the coupling effect between multiple risks. It can output risk level assessment results that conform to the actual engineering situation and implementable support optimization suggestions.

[0079] In addition, this solution also discloses a probabilistic model construction system for geological risk assessment in tunnel engineering, which is applied in probabilistic model construction methods, including:

[0080] The data acquisition module collects and stores the corresponding surrounding rock characteristic parameters for each section;

[0081] The prior probability calculation module is used to establish the prior risk probability function under unsupported conditions and calculate the prior probability of various geological risks.

[0082] The posterior probability calculation module is used to calculate the risk reduction coefficient by combining the support matching degree, and then obtain the posterior risk probability of various risks.

[0083] The joint probability calculation module is used to calculate the joint probability of multiple risks by combining the posterior probability of various risks with the risk correlation coefficient.

[0084] The risk assessment output module is used to classify risk levels according to preset thresholds and output assessment results and support optimization suggestions.

[0085] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. A method for constructing a probabilistic model for geological risk assessment in tunnel engineering, characterized in that, Includes the following steps: S1. Divide the tunnel into several sections along the tunnel axis according to the preset section length. In each section, use at least one of the following methods to obtain the surrounding rock characteristic parameters of the section: drilling while drilling measurement, machine vision scanning, and advanced geological prediction. Among them, the characteristic parameters of the surrounding rock include: rock quality index, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress and special geological body indication; S2. Establish a prior risk probability function under unsupported conditions, and calculate the prior probability of various geological risk types respectively. The geological risk types include at least landslide, sudden water inrush and large deformation of surrounding rock. S3. Based on the surrounding rock characteristic parameters collected in the section, determine the support level recommended in the construction specification document, and collect the support level to be applied in the section, and calculate the support matching degree. S4. For each type of geological risk, calculate the risk reduction coefficient of the support scheme and the posterior risk probability of the support scheme based on the proposed support level and support matching degree. S5. Based on the posterior risk probabilities of the three risks, calculate the joint probability of the three risks. S6. Based on the posterior risk probability and joint probability of each individual risk, determine the risk level according to the preset threshold range, and output the risk determination result and support optimization suggestions. In step S3, the formula for calculating the support matching degree is: ,in, The proposed support level is... The recommended support level is specified in the standard document; In step S4, the formula for calculating the reduction coefficient is: ,in, Let be the sensitivity coefficient of the i-th type of risk to the support structure, obtained through numerical simulation and engineering inversion. Values ​​range from 1.6 to 2.

0. The value ranges from 0.7 to 1.

0. Values ​​range from 1.0 to 1.

4. The highest support level is represented by a value of 5. This is a matching degree correction factor; In step S4, the formula for calculating the posterior risk probability is: ; in, , When the geological body is special, the weakening coefficient of the i-th type of risk on the support effect is... The value ranges from 0.25 to 0.

45. The value ranges from 0.2 to 0.

35. The value ranges from 0.15 to 0.

3. This indicates a special geological body, and its value is either 0 or 1. Let be the prior risk probability of risk type i.

2. The method for constructing a probabilistic model for geological risk assessment in tunnel engineering according to claim 1, characterized in that, In step S2, the expression for the prior risk probability function is: Where i represents three types of geological risk, i=1 represents landslide, i=2 represents sudden water inrush, i=3 represents large deformation of surrounding rock; X1-X6 are rock quality indicators, joint volume number, rock mass integrity coefficient, groundwater seepage flow, tunnel wall tangential stress and special geological body indication, respectively. When a special geological body exists, X6 takes the value of 1, and when there is no special geological body, X6 takes the value of 0. and - These are fixed constants corresponding to the geological risk type.

3. The method for constructing a probabilistic model for geological risk assessment in tunnel engineering according to claim 1, characterized in that, In the formula for calculating the reduction coefficient The function is defined as: ,in, For support matching degree.

4. The method for constructing a probabilistic model for geological risk assessment in tunnel engineering according to claim 1, characterized in that, In step S5, the formula for calculating the joint probability is: ,in, The correlation coefficient between risks ranges from 0 to 0.

5. The posterior risk probability for each type of risk.

5. The method for constructing a probabilistic model for geological risk assessment in tunnel engineering according to claim 1, characterized in that, The risk level is determined according to the following rules: If 0 ≤ P < 0.01, it is considered risk-free; If 0.01 ≤ P < 0.15, the risk level is determined to be low, and a green alert is issued. If 0.15 ≤ P < 0.4, the risk level is determined to be medium, and a yellow warning is issued. If 0.4 ≤ P, it is judged as high risk, and a red alert is issued; Where P is the posterior risk probability of any single risk or the joint probability of the three risks; When P ≥ 0.4, the required support level needs to be increased. To obtain a new support level , ; in, For the highest support level, the value is 5. k is an empirical coefficient; when 0.4 ≤ P < 0.6, k is 0.3; when P ≥ 0.6, k is 0.

5. Round up to the nearest integer.

6. A probabilistic model construction system for judging geological risks in tunnel engineering, applied in the probabilistic model construction method according to any one of claims 1-5, characterized in that, include: The data acquisition module collects and stores the corresponding surrounding rock characteristic parameters for each section; The prior probability calculation module is used to establish the prior risk probability function under unsupported conditions and calculate the prior probability of various geological risks. The posterior probability calculation module is used to calculate the risk reduction coefficient by combining the support matching degree, and then obtain the posterior risk probability of various risks. The joint probability calculation module is used to calculate the joint probability of multiple risks by combining the posterior probability of various risks with the risk correlation coefficient. The risk assessment output module is used to classify risk levels according to preset thresholds and output assessment results and support optimization suggestions.

Images