Gas tunnel construction risk assessment method based on game theory combination weighting

Abstract

The invention discloses a gas tunnel construction risk assessment method based on game theory combination weighting, and relates to the field of tunnel engineering safety, the assessment method comprises the following steps: establishing a 10-index evaluation system covering four types of factors of coal seam occurrence, gas parameters, geological features and tunnel design, and constructing a grading standard; then, obtaining subjective weights by adopting an analytic hierarchy process (AHP), calculating objective weights by combining an anti-entropy weight method (AEW), and solving an optimal combination coefficient of the subjective and objective weights through a game theory combination weighting model to realize dynamic optimization of a weight system; a single-index continuous measurement function is constructed by introducing an unascertained measurement theory, and fuzzy transition characteristics of an index state are described through a measurement axiom of non-negative bounded, normalized and addible; the method can break through the static threshold limit of a traditional method, provides a more scientific and engineering-applicable method for gas tunnel risk evaluation, and provides a theoretical reference for risk evaluation of tunnel gas disasters.

Description

Technical Field

[0001] The present invention relates to the field of tunnel engineering safety, and in particular to a gas tunnel construction risk assessment method based on game theory combined weighting. Background Art

[0002] my country has relatively abundant and widely distributed coal and oil resources. With the development of transportation, tunnel construction requires traversing gas-bearing strata, particularly in southwest China. Many tunnels, particularly in the southwest, will traverse gas-bearing coal seams during construction. Gas tunnels face complex geological environments and challenging construction, making them prone to production safety accidents, posing a serious threat to the lives of construction workers and the safety of production equipment. However, existing research still faces the following limitations that need to be addressed: First, in terms of weight determination, some studies use a single subjective weighting method, which is easily influenced by expert subjective experience and can lead to weight assignments that deviate from actual project requirements. While some objective weighting methods rely on objective data calculations, they can produce extreme weights due to indicator sensitivity and ignore the empirical judgment of geological experts on key risks. This disconnect between subjective and objective methods makes it difficult to integrate engineering experience with data-driven approaches in the weighting system. Second, ambiguity and uncertainty are common in gas hazard assessment. While existing methods, such as fuzzy comprehensive evaluation and extension theory, can address some of this uncertainty, they often rely on discrete thresholds for grading, making it difficult to characterize the continuous transition of indicator states from low to high risk. Third, existing combination weighting methods often use linear weighting or simple normalization processing, which fails to effectively resolve the game conflict between subjective and objective weights.

[0003] Therefore, in order to solve the above problems, this paper proposes a risk assessment method for gas tunnel construction based on game theory combination empowerment. Summary of the Invention

[0004] The purpose of this invention is to design a risk assessment method for gas tunnel construction based on game theory combined weighting, which is suitable for quantitative assessment of risk levels in gas tunnel construction and solves the problems of imbalance in subjective and objective weight distribution and insufficient fuzzy processing in traditional methods.

[0005] In order to achieve the above technical effects, the present invention is implemented through the following technical solutions: a gas tunnel construction risk assessment method based on game theory combined weighting, characterized by comprising the following steps:

[0006] S1. Construct an evaluation index system: select coal seam occurrence, gas, geological characteristics, and tunnel design as evaluation indicators;

[0007] S2. Based on existing tunnel safety assessment guidelines and related research, establish quantitative indicators and grading standards. Classify the risk level into four levels according to the grading standards, and then classify the quantitative indicators into risk levels based on the classification standards.

[0008] S3. Construct a game theory combined weighting model based on the subjective weight of the analytic hierarchy process (AHP), the objective weighting of the anti-entropy weight method (AEW), and the game theory optimization combination;

[0009] S4. Construct an unascertained measurement theory model based on single-index measurement function and multi-index comprehensive evaluation;

[0010] S5. Utilize the game theory combined weighting model to solve the optimal combination coefficient of subjective and objective weights, and realize the dynamic optimization of the weight system. Use the "non-negative boundedness, normalization, and additive" measurement axioms of the unascertained measurement theory model to characterize the fuzzy transition characteristics of the indicator state. Finally, determine the tunnel gas disaster risk level through the maximum membership principle.

[0011] Furthermore, in S1, the coal seam occurrence, gas content, geological characteristics, and tunnel design are as follows:

[0012] Coal seam occurrence refers to coal seam strike, coal seam dip, coal seam thickness and coal seam inclination. Coal seam thickness is easy to quantify and measure, so the coal seam thickness is selected as an indicator to characterize coal seam occurrence.

[0013] Gas content refers to the gas content expressed by three indicators: average gas concentration of return air flow, coal seam gas content and relative gas emission volume;

[0014] Geological characteristics are expressed in terms of stratum lithology and geological structure. Since gas is less likely to explode in a permeable rock mass, but more likely to explode in a permeable rock mass, rock type is used to characterize stratum lithology. Geological structure is primarily characterized by cracks, fractures, surfaces, and linear structures in the rock mass. Furthermore, the connectivity and sealing of the rock mass directly affects the permeability and stability of the tunnel, so connectivity and sealing is used to characterize the geological structure.

[0015] Tunnel design refers to tunnel depth, tunnel span, tunnel length and ventilation wind speed.

[0016] Furthermore, in S2, the quantitative indicators established include: coal seam thickness, coal seam gas content, average gas concentration of return air flow, relative gas outburst volume, rock type, connectivity and sealing, tunnel depth, tunnel length, tunnel span and face wind speed.

[0017] Furthermore, in S2, the risk levels of quantitative indicators are divided based on the classification criteria, as follows:

[0018] Coal seam thickness: 3.5m and above is level one risk; 1.3-3.5m is level two risk; 0.5-1.3m is level three risk; 0-0.5m is level four risk;

[0019] Coal seam gas content: 9m3 / t and above are level one risks; 6-9m 3 / t is the second level risk, 3-6m 3 / t is the third level risk, 0-3m 3 / t is level 4 risk;

[0020] Average gas concentration in return air: 0.85% and above is level 1 risk; 0.7%-0.85% is level 2 risk; 0.6%-0.7% is level 3 risk; 0-6% is level 4 risk;

[0021] Relative gas outflow: 10m 3 / t and above are level one risks; 7-10m 3 / t is the second level risk, 5-7m 3 / t is the third level risk, 0-5m 3 / t is level 4 risk;

[0022] Rock type: Shale and mudstone are classified as level 1 risk; limestone is classified as level 2 risk; silt and fine sand are classified as level 3 risk; medium and coarse sand are classified as level 4 risk;

[0023] Connectivity and closure: Closed structures are classified as level one risk; undisturbed structures are classified as level two risk; connected structures are classified as level three risk; and tensile structures are classified as level four risk.

[0024] Tunnel depth: 500m and above is level 1 risk; 300-500m is level 2 risk; 100-300m is level 3 risk; 0-100m is level 4 risk;

[0025] Tunnel length: 14km and above is level 1 risk; 7-14km is level 2 risk; 2-7km is level 3 risk; 0-2km is level 4 risk;

[0026] Tunnel span: 0-9m is level 1 risk; 9-14m is level 2 risk; 14-18m is level 3 risk; 18m and above is level 4 risk;

[0027] Wind speed at the palm face: 0-0.15m / s is level one risk; 0.15-0.25m / s is level two risk; 0.25-0.5m / s is level three risk; 0.5m / s and above is level four risk.

[0028] Furthermore, in S3, the subjective weights of the analytic hierarchy process (AHP), the objective weights of AEW, and the game theory optimization are specifically combined as follows:

[0029] Subjective weights of the analytic hierarchy process (AHP): The importance of indicators is compared pairwise based on expert experience, a judgment matrix is ​​constructed, and the weights are calculated using the sum-product method and a consistency test is performed (CR < 0.1);

[0030] Objective weighting of the Anti-Entropy Weighting (AEW) method: After standardizing the original data, the indicator weight is calculated using the anti-entropy value to reduce the sensitivity of data fluctuations;

[0031] Game theory optimization combination: A cross programming model is introduced to solve the optimal combination coefficients (θ1 and θ2) of subjective and objective weights to achieve dynamic balance of weights.

[0032] Furthermore, in S4, the single-index measurement function and multi-index comprehensive evaluation are as follows:

[0033] Single indicator measurement function: Based on the indicator level standard, a continuous measurement function is constructed to characterize the degree to which the indicator belongs to a certain risk level (0≤μ≤1), satisfying the "non-negative, bounded, normalized, and additive" axioms; the specific axioms are as follows:

[0034] When the single indicator measurement evaluation matrix When expressing evaluation indicators; Belongs to level k C k The degree of the above mentioned conditions must meet the following requirements (1)-(3):

[0035]

[0036] Among them, formula (1) is called "non-negative boundedness", formula (2) is called "normalization", and formula (3) is called "additivity". When μ satisfies (1)-(3) at the same time, it is called an unascertained measure, or simply measure.

[0037] In formulas (1)-(3), is a single indicator measurement evaluation matrix; k is a constant; μ is an unascertained measure, indicating the degree to which the indicator belongs to a certain risk level; is the observed value of the i-th indicator of the j-th evaluation object; C k is the kth risk level (k = 1, 2, 3, 4, corresponding to level I (very high), level II (high), level III (medium), level IV (low)); U is the evaluation space, that is, the union of all risk levels.

[0038] Comprehensive evaluation of multiple indicators: The single indicator measurement values ​​are weighted and summed by comprehensive weights to obtain the comprehensive measurement vector μ = [μ1, μ2, μ3, μ4], which corresponds to extremely high risk level I, high risk level II, medium risk level III and low risk level IV.

[0039] The beneficial effects of the present invention are:

[0040] This paper comprehensively considers the factors affecting gas disasters, combines the existing research status, and starts from the four aspects of coal seam occurrence, gas, geological characteristics, and tunnel design. In combination with the existing data, it selects 10 risk factors affecting gas disasters in tunnels, establishes a tunnel gas disaster risk assessment index system, and classifies each indicator.

[0041] The game theory combined weighting method adopted in this invention achieves a dynamic balance between subjective experience and objective data through cross-planning optimization of AHP and anti-entropy weighting method. It not only retains the emphasis of engineering experience on deep burial risks, but also corrects the deviation of single subjective judgment through data anti-entropy, making the weighting system more robust. The unascertained measure theory transforms the fuzzy level of gas disaster evaluation index into a continuous probability distribution through the measurement axiom of "non-negative boundedness-normalization-additivity", breaking through the discretization limitation of traditional fuzzy evaluation and intuitively reflecting the transition characteristics of the indicator state.

[0042] The gas tunnel construction risk assessment method based on game theory combined empowerment described in the present invention was used to evaluate gas disaster risks in actual tunnel projects. The risk assessment results were compared with the on-site investigation results. It was found that the evaluation results were basically consistent with the actual situation, indicating that this method has good applicability in tunnel gas risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a flow chart of the risk assessment method for gas tunnel construction based on game theory combined weighting according to the present invention;

[0045] Figure 2 It is a single indicator measurement function diagram of each evaluation indicator of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0047] Example 1

[0048] A risk assessment method for gas tunnel construction based on game theory combined weighting is characterized by comprising the following steps:

[0049] Step 1: Construct an evaluation index system: Select coal seam occurrence, gas, geological characteristics, and tunnel design as evaluation indicators. Coal seam occurrence refers to coal seam strike, coal seam dip, coal seam thickness, and coal seam inclination. Coal seam thickness is easy to quantify and measure, so coal seam thickness is selected as an indicator to represent coal seam occurrence.

[0050] Gas content refers to the gas content expressed by three indicators: average gas concentration of return air flow, coal seam gas content and relative gas emission volume;

[0051] Geological characteristics are expressed in terms of stratum lithology and geological structure. Since gas is less likely to explode in a permeable rock mass, but more likely to explode in a permeable rock mass, rock type is used to characterize stratum lithology. Geological structure is primarily characterized by cracks, fractures, surfaces, and linear structures in the rock mass. Furthermore, the connectivity and sealing of the rock mass directly affects the permeability and stability of the tunnel, so connectivity and sealing is used to characterize the geological structure.

[0052] Tunnel design refers to tunnel depth, tunnel span, tunnel length and ventilation wind speed.

[0053] Step 2: Establish quantitative indicators based on existing tunnel safety assessment guidelines and related research. These indicators include coal seam thickness, coal seam gas content, average return air flow gas concentration, relative gas emission, rock type, connectivity and sealing, tunnel depth, tunnel length, tunnel span, and face wind speed.

[0054] Based on the current construction status, geological conditions, and risk characteristics of gas tunnels, and in conjunction with the Technical Specifications for Railway Gas Tunnels, the Tunnel Safety Assessment Guide, and related research, the gas disaster risk during tunnel construction is divided into four levels. Specifically, the classification and manifestation of tunnel gas disaster risk levels are shown in Table 1.

[0055]

[0056] Table 1

[0057] The indicators are classified into different levels based on the tunnel gas disaster risk level classification, as follows:

[0058] Coal seam thickness: 3.5m and above is level one risk; 1.3-3.5m is level two risk; 0.5-1.3m is level three risk; 0-0.5m is level four risk;

[0059] Coal seam gas content: 9m 3 / t and above are level one risks; 6-9m 3 / t is the second level risk, 3-6m 3 / t is the third level risk, 0-3m 3 / t is level 4 risk;

[0060] Average gas concentration in return air: 0.85% and above is level 1 risk; 0.7%-0.85% is level 2 risk; 0.6%-0.7% is level 3 risk; 0-6% is level 4 risk;

[0061] Relative gas outflow: 10m 3 / t and above are level one risks; 7-10m 3 / t is the second level risk, 5-7m 3 / t is the third level risk, 0-5m 3 / t is level 4 risk;

[0062] Rock type: Shale and mudstone are classified as level 1 risk; limestone is classified as level 2 risk; silt and fine sand are classified as level 3 risk; medium and coarse sand are classified as level 4 risk;

[0063] Connectivity and closure: Closed structures are classified as level one risk; undisturbed structures are classified as level two risk; connected structures are classified as level three risk; and tensile structures are classified as level four risk.

[0064] Tunnel depth: 500m and above is level 1 risk; 300-500m is level 2 risk; 100-300m is level 3 risk; 0-100m is level 4 risk;

[0065] Tunnel length: 14km and above is level 1 risk; 7-14km is level 2 risk; 2-7km is level 3 risk; 0-2km is level 4 risk;

[0066] Tunnel span: 18m and above is level 1 risk; 14-18m is level 2 risk; 9-14m is level 3 risk; 0-9m is level 4 risk;

[0067] Wind speed at the face: 0-0.15m / s is level 1 risk; 0.15-0.25m / s is level 2 risk; 0.25-0.5m / s is level 3 risk; 0.5m / s and above is level 4 risk;

[0068] Based on the above content, the evaluation indicators and grade standards are established as shown in Table 2;

[0069]

[0070]

[0071] Table 2

[0072] Step 3: Construct a game theory combined weighting model based on the subjective weight of the analytic hierarchy process (AHP), the objective weighting of the anti-entropy weight method (AEW), and the game theory optimization combination;

[0073] The subjective weight of the analytic hierarchy process (AHP) refers to the process of comparing the influence of each element in the same level on the indicators of the previous level based on the risk assessment index system, calculating its characteristic value to determine the priority (i.e. weight), usually using the nine-point scale. The main process is as follows:

[0074] The risk factors of tunnel gas disasters are compared pairwise according to the 9-level scaling method, and the judgment matrix A is constructed as shown in formula (1):

[0075]

[0076] Where: b is the risk factor;

[0077] Use the sum-product method to solve the subjective weight

[0078] 1) Normalize the judgment matrix A, as shown in formula (2):

[0079] Column normalization

[0080] Row normalization, sum the matrix by row The vector is shown in formula (3):

[0081] W 主 =(W1,W2,…W n ) T (3)

[0082] By normalizing the vector W 主 , calculate the weight vector W', as shown in formula (4):

[0083]

[0084] 2) Consistency test

[0085] Calculate the maximum eigenvalue of matrix A, as shown in formula (5):

[0086]

[0087] Where: C R is the consistency ratio; C I Consistency index; R I Random consistency index;

[0088] C R When it is less than 0.1, it indicates that the weight distribution is reasonable; C R When it is greater than 0.1, the judgment matrix needs to be adjusted and recalculated until a reasonable result is achieved.

[0089] The objective weighting of the anti-entropy weight method (AEW) refers to the calculation of indicator weights through anti-entropy values ​​after standardizing the original data to reduce the sensitivity of data fluctuations. The details are as follows:

[0090] 1) Create the original matrix

[0091] Suppose there are S evaluation objects, x j ∈X(j=1,2,…,S) is the evaluation object, and each evaluation object has n evaluation indicators X1,X2,…,X n , X ji is the measurement value of the i-th evaluation index of the j-th evaluation object; the matrix is ​​shown in Formula 8.

[0092]

[0093] 2) Standardize the original data

[0094] For the indicators of the larger the better type,

[0095]

[0096] The smaller the better the index is

[0097]

[0098] Then we can get the normalized matrix

[0099] 3) Calculate the anti-entropy value of the i-th indicator. The formula is shown in formula (12):

[0100]

[0101] Where: E i represents the anti-entropy value of the i-th indicator; S represents the total number of evaluation objects;

[0102] Among them, P ji As shown in formula (13):

[0103]

[0104] Where: P ji Represents the standardized value y of the i-th indicator of the j-th evaluation object ji The proportion of the sum of the standardized values ​​of all evaluation objects of the indicator;

[0105] 4) According to the obtained anti-entropy value of each indicator, the weight of the i-th indicator is calculated as follows:

[0106]

[0107] Where: w irepresents the objective weight of the i-th indicator;

[0108] Game theory optimization combination refers to the introduction of a cross-programming model to solve the optimal combination coefficients (θ1 and θ2) of subjective and objective weights to achieve dynamic balance of weights, as follows:

[0109] When weighting the indicators extracted from gas disasters, in order to resolve the contradictory relationship between subjective and objective decision values ​​with different weights, find the minimum difference between the weights of each indicator and obtain the most coordinated decision weight indicator value, a comprehensive weighting method based on the weight aggregation idea of ​​game theory (GT) is used to construct the optimal combination state AHP-AEW of the two weighting methods, find the deviation between the minimized weight and other weights, and determine the optimal weighting result.

[0110] If θ1 and θ2 are combination coefficients, then the comprehensive weight value W 综 for

[0111]

[0112] Optimize θ1 and θ2 and introduce the cross programming model:

[0113]

[0114] The optimized first-order derivative equation is

[0115]

[0116] Get the combination coefficient and normalize it:

[0117]

[0118] Finally, the optimized comprehensive weight vector W of each decision indicator is obtained 综 :

[0119]

[0120] Step 4: Construct an unascertained measurement theory model based on single-index measurement function and multi-index comprehensive evaluation;

[0121] The single indicator measurement function refers to a continuous measurement function constructed according to the indicator grade standard, which describes the degree to which the indicator belongs to a certain risk grade (0≤μ≤1) and satisfies the axioms of "non-negative, bounded, normalized, and additive". The specific axioms are as follows:

[0122] When the single indicator measurement evaluation matrix When expressing evaluation indicators; Belongs to level k C k The degree of the above mentioned conditions must meet the following requirements (1)-(3):

[0123]

[0124]

[0125] Among them, formula (1) is called "non-negative boundedness", formula (2) is called "normalization", and formula (3) is called "additivity". When μ satisfies (1)-(3) at the same time, it is called an unascertained measure, or simply measure.

[0126] In formulas (1)-(3), is a single indicator measurement evaluation matrix; k is a constant; μ is an unascertained measure, indicating the degree to which the indicator belongs to a certain risk level; is the observed value of the i-th indicator of the j-th evaluation object; C k is the kth risk level (k = 1, 2, 3, 4, corresponding to level I (very high), level II (high), level III (medium), level IV (low)); U is the evaluation space, that is, the union of all risk levels.

[0127] Multi-indicator comprehensive evaluation refers to the weighted summation of single indicator measurement values ​​through comprehensive weights to obtain the comprehensive measurement vector μ = [μ1, μ2, μ3, μ4], which corresponds to extremely high risk level I, high risk level II, medium risk level III, and low risk level IV. The comprehensive measurement vector is as follows:

[0128] Let μ ijk =μ(x ij ∈C k ) indicates tunnel engineering i Belongs to the kth risk level C k To the extent that:

[0129]

[0130] Where: is the comprehensive weight vector consisting of j columns.

[0131] If the measure μ is unknown jk Satisfies: 0≤μ jk ≤1, Then we call μ j1 ,μ j2 ,…,μ jq ] for tunnel engineering i The multi-index evaluation vector of .

[0132] Step 5: Use the game theory combined weighting model to solve the optimal combination coefficient of subjective and objective weights to achieve dynamic optimization of the weight system; use the "non-negative bounded, normalized, and additive" measurement axioms of the unascertained measurement theory model to characterize the fuzzy transition characteristics of the indicator state; finally, determine the tunnel gas disaster risk level using the maximum membership principle.

[0133] Example 2

[0134] The specific application of the risk assessment method for gas tunnel construction based on game theory combination empowerment is as follows:

[0135] The Zhaotong Tunnel with a total length of 16.26 km was selected. The tunnel belongs to the high and medium mountain landform area in the plateau slope zone. The terrain is steep and undulating. The ground elevation is 1390~2900m, the relative height difference is 1500m, and the maximum burial depth is about 990m. It is a deep-buried extra-long tunnel; the DK377km+410~740m section of the tunnel passes through the coal-bearing strata of the Jiusi section of the Datang stage of the Lower Carboniferous System. The advance drilling revealed that the coal seams are located in or near the tunnel body. A total of 9 coal seams were revealed, numbered from top to bottom as M1-9~M1-7 (first group), M1-6~M1-5 (second group), M1-4~M1-3 (third group) and M1-2~M1-1 (fourth group).

[0136] Four groups of gas tunnel risk assessment samples were obtained, and the practicality and effectiveness of the method were verified by combining them with the tunnel section (T1) of the Zhaotong Tunnel when uncovering the second coal seam. The data are shown in Table 3.

[0137]

[0138] Table 3 Gas tunnel data information

[0139] Substitute the data information of various indicators into Figure 2 In the corresponding indicator measurement function diagram, the single indicator measurement matrix of each sample can be obtained; taking T1 as an example, the indicator measurement matrix is ​​constructed:

[0140]

[0141] The subjective weighting method of AHP and the objective weighting method of IEW are combined with the GT method to determine the comprehensive weight of each indicator, as shown in Table 4:

[0142]

[0143] Table 4 Tunnel gas disaster evaluation index weights

[0144] According to the single-index evaluation matrix of T1 and the comprehensive weight of each evaluation index, the multi-index measurement evaluation vector of T1 is [0.381, 0.134, 0.296, 0.189], and its risk level is determined according to the maximum membership principle. Similarly, the multi-index measurement evaluation vectors of other tunnel samples can be obtained, as shown in Table 5.

[0145]

[0146]

[0147] Table 5 Evaluation results of unascertained measure theory

[0148] The analysis results show that the gas hazard risk distribution for each sample tunnel is highly consistent with the actual geological conditions and construction status. The results for T1 indicate that its risk level is Level I, primarily due to the extremely high tunnel depth (weight 0.154) and tunnel length (weight 0.168), which together contribute approximately 32% of the weight. This indicates that the geological characteristics of deep, long tunnels significantly increase the risk of gas accumulation. Furthermore, although the coal seam thickness of 0.88m falls within the Level III range, the average gas concentration in the return air is only 0.00621% (well below the Level III threshold), reflecting the significant dilution effect of the ventilation system (ventilation speed 0.32m / s, weight 0.089) on low-concentration gas. However, the potential risk of gas outburst under deep burial conditions remains a high priority. The risk level of SD5 is level IV (low risk). Although its tunnel span is 21m (weight 0.100) and belongs to the level I interval, the ventilation wind speed is 0.54m / s (exceeding the level I threshold) and the relative gas outflow volume is 8.4m 3 The combined effect of the three factors (level III) reduces the extreme impact of a single indicator through game theory weighting. This demonstrates that the model, through the dynamic balance of subjective and objective weights, can effectively avoid the over-reliance on a single factor in traditional subjective weighting methods, demonstrating the scientific nature of combined weighting.

[0149] Comparing the indicator weight distribution of each sample, the comprehensive weights of tunnel depth and tunnel length reached 0.154 and 0.168 respectively, which were significantly higher than other indicators, confirming the engineering consensus that "deep and long tunnels have higher gas disaster risks"; and the difference in weights of rock type and connectivity and sealing reveals the control of stratum lithology on gas permeability - T1's rock type score is 55 (silt and fine sand, level III), and its connectivity is better than T2's 30 points (limestone, level II), but the dominant role of depth value makes its risk level still relatively high, indicating that the coupling effect of different geological factors needs to be quantified and analyzed through models.

[0150] From a methodological perspective, the unascertained measure theory effectively addresses ambiguity in gas hazard assessment by characterizing the continuity of a single-indicator measurement function. For example, the T1 coal seam thickness is 0.88m, which falls within the Level III evaluation interval [0.5, 1.3), but is close to the lower limit of the Level II interval (1.3m). Traditional thresholding methods would directly classify it as Level III, ignoring the indicator's gradual progression toward higher risk levels. However, the unascertained measure theory, by constructing a continuous measurement function, calculates the indicator's values ​​at Levels III and II to be 0.475 and 0.525, respectively, indicating that its risk status is not entirely confined to Level III and that there is a potential for transition to Level II.

[0151] Further combined with the comprehensive effect of other indicators, although the low-level measurement of a single indicator (such as coal seam thickness) accounts for a certain proportion, the Level I measurement of high-weight indicators (tunnel depth and length) dominates the comprehensive evaluation results through linear weighting, and ultimately the risk level of T1 is determined to be Level I; this process embodies the two major advantages of the unascertained measurement theory: continuity representation: breaking through the mechanical division of traditional discrete thresholds, and using probability measurement to characterize the fuzzy transition of indicator states, which is more in line with the physical nature of gas disasters of "gradual accumulation and sudden outbreak"; systematic integration: through a comprehensive weight system (game theory combination empowerment), the nonlinear fusion of multi-indicator information is realized, avoiding the risk level deviation caused by misjudgment of a single indicator, and finally outputting a gradual evaluation conclusion that conforms to the actual engineering practice.

[0152] This methodology provides a new paradigm for the refined assessment of tunnel gas disaster risks, which is particularly suitable for complex geological conditions under the coupling of multiple factors. It effectively makes up for the limitations of traditional evaluation methods in dealing with "boundary ambiguity" and "system integrity".

Claims

1. A risk assessment method for gas tunnel construction based on game theory combined weighting is characterized by: The following steps are involved: S1. Construct an evaluation index system: select coal seam occurrence, gas, geological characteristics, and tunnel design as evaluation indicators; S2. Establish quantitative indicators and grading standards based on existing tunnel safety assessment guidelines and related research. Divide the risk level into four levels based on the grading standards, and categorize the quantitative indicators into risk levels based on the grading standards. S3. Construct a game theory combined weighting model based on the subjective weight of the analytic hierarchy process (AHP), the objective weighting of the anti-entropy weight method (AEW), and the game theory optimization combination; S4. Construct an unascertained measurement theory model based on single-index measurement function and multi-index comprehensive evaluation; S5. Utilize the game theory combined weighting model to solve the optimal combination coefficient of subjective and objective weights, and realize the dynamic optimization of the weight system. Use the "non-negative boundedness, normalization, and additive" measurement axioms of the unascertained measurement theory model to characterize the fuzzy transition characteristics of the indicator state. Finally, determine the tunnel gas disaster risk level through the maximum membership principle.

2. The method for risk assessment of gas tunnel construction based on game theory combined weighting according to claim 1 is characterized in that: In S1, the coal seam occurrence, gas content, geological characteristics, and tunnel design are as follows: Coal seam occurrence refers to coal seam strike, coal seam dip, coal seam thickness and coal seam inclination. Coal seam thickness is easy to quantify and measure, so the coal seam thickness is selected as an indicator to characterize coal seam occurrence. Gas content refers to the gas content expressed by three indicators: average gas concentration of return air flow, coal seam gas content and relative gas emission volume; Geological characteristics are expressed in terms of stratum lithology and geological structure. Since gas is less likely to explode in a permeable rock mass, but more likely to explode in a permeable rock mass, rock type is used to characterize stratum lithology. Geological structure is primarily characterized by cracks, fractures, surfaces, and linear structures in the rock mass. Furthermore, the connectivity and sealing of the rock mass directly affects the permeability and stability of the tunnel, so connectivity and sealing is used to characterize the geological structure. Tunnel design refers to tunnel depth, tunnel span, tunnel length and ventilation wind speed.

3. The method for risk assessment of gas tunnel construction based on game theory combined weighting according to claim 1 is characterized in that: In S2, the quantitative indicators established include: coal seam thickness, coal seam gas content, average gas concentration of return air flow, relative gas outburst rate, rock type, connectivity and sealing, tunnel depth, tunnel length, tunnel span and face wind speed.

4. The method for risk assessment of gas tunnel construction based on game theory combined weighting according to claim 1 is characterized in that: In S2, the risk levels of quantitative indicators are divided based on the classification criteria, as follows: Coal seam thickness: 3.5m and above is level one risk; 1.3-3.5m is level two risk; 0.5-1.3m is level three risk; 0-0.5m is level four risk; Coal seam gas content: 9m 3 / t and above are level one risks; 6-9m 3 / t is the second level risk, 3-6m 3 / t is the third level risk, 0-3m 3 / t is level 4 risk; Average gas concentration in return air: 0.85% or above is considered a Level 1 risk; 0.7%-0.85% is level 2 risk, 0.6%-0.7% is level 3 risk, and 0-6% is level 4 risk; Relative gas outflow: 10m 3 / t and above are level one risks; 7-10m 3 / t is the second level risk, 5-7m 3 / t is the third level risk, 0-5m 3 / t is level 4 risk; Rock type: Shale and mudstone are classified as level 1 risk; limestone is classified as level 2 risk; silt and fine sand are classified as level 3 risk; medium and coarse sand are classified as level 4 risk; Connectivity and closure: Closed structures are considered a level one risk; Undisturbed structures are classified as level 2 risk, connected structures as level 3 risk, and tensile structures as level 4 risk; Tunnel depth: 500m and above is level 1 risk; 300-500m is level 2 risk; 100-300m is level 3 risk; 0-100m is level 4 risk; Tunnel length: 14km and above is level 1 risk; 7-14km is level 2 risk; 2-7km is level 3 risk; 0-2km is level 4 risk; Tunnel span: 0-9m is level 1 risk; 9-14m is level 2 risk; 14-18m is level 3 risk; 18m and above is level 4 risk; Wind speed at the face: 0-0.15m / s is considered a level 1 risk; 0.15-0.25m / s is a level 2 risk, 0.25-0.5m / s is a level 3 risk, and 0.5m / s and above is a level 4 risk.

5. The method for risk assessment of gas tunnel construction based on game theory combined weighting according to claim 1 is characterized in that: In S3, the subjective weights of the analytic hierarchy process (AHP), the objective weights of AEW, and the game theory optimization combination are as follows: Subjective weights of the analytic hierarchy process (AHP): The importance of indicators is compared pairwise based on expert experience, a judgment matrix is ​​constructed, and the weights are calculated using the sum-product method and a consistency test is performed (CR < 0.1); Objective weighting of the Anti-Entropy Weighting (AEW) method: After standardizing the original data, the indicator weight is calculated using the anti-entropy value to reduce the sensitivity of data fluctuations; Game theory optimization combination: A cross programming model is introduced to solve the optimal combination coefficients (θ1 and θ2) of subjective and objective weights to achieve dynamic balance of weights.

6. The method for risk assessment of gas tunnel construction based on game theory combined weighting according to claim 1 is characterized in that: In S4, the single-index measurement function and multi-index comprehensive evaluation are as follows: Single indicator measurement function: Based on the indicator grade standard, a continuous measurement function is constructed to characterize the degree to which the indicator belongs to a certain risk grade (0≤μ≤1), satisfying the axioms of "non-negative, bounded, normalized, and additive". The specific axioms are as follows: When the single indicator measurement evaluation matrix When expressing evaluation indicators; Belongs to level k C k The degree of the above mentioned conditions must meet the following requirements (1)-(3): Among them, formula (1) is called "non-negative boundedness", formula (2) is called "normalization", and formula (3) is called "additivity". When μ satisfies (1)-(3) at the same time, it is called an unascertained measure, or simply measure. In formulas (1)-(3), is a single indicator measurement evaluation matrix; k is a constant; μ is an unascertained measure, which indicates the degree to which the indicator belongs to a certain risk level; is the observed value of the i-th indicator of the j-th evaluation object; C k is the kth risk level (k = 1, 2, 3, 4, corresponding to level I (very high), level II (high), level III (medium), level IV (low)); U is the evaluation space, that is, the union of all risk levels; Comprehensive evaluation of multiple indicators: The single indicator measurement values ​​are weighted and summed by comprehensive weights to obtain the comprehensive measurement vector μ = [μ1, μ2, μ3, μ4], which corresponds to extremely high risk level I, high risk level II, medium risk level III and low risk level IV.

Images