A mine regional safety risk assessment method

Abstract

The present application belongs to the technical field of mine safety assessment, and provides a mine regional safety risk assessment method, comprising: index system construction, DEMATEL method weight calculation, BWM method weight calculation, combination optimization, evaluation standard cloud construction, reverse cloud parameter calculation, comprehensive cloud parameter fusion calculation, forward cloud generation calculation, grade matching; the present application combines and optimizes the first and second weight sets through the weight optimization model, realizes the finding of the optimal solution combination with the minimum deviation based on the game theory principle, reduces the influence of subjectivity, and improves the reliability of the weight index; the present application performs fusion calculation on the evaluation index combination weight and the index cloud parameter through the comprehensive cloud parameter model, takes into account the fuzziness and randomness of the mine risk index, improves the accuracy of the risk grade determination, and avoids the grade misjudgment when close to the threshold.

Description

Technical Field

[0001] This invention relates to the field of mine safety assessment technology, and in particular to a method for regional safety risk assessment in mines. Background Technology

[0002] Monitoring weights originated from mathematical statistics and are commonly used in safety risk assessment processes. Weights reflect the importance of the assessed object, and their application ranges from large-scale system engineering optimization decisions to various everyday decision-making problems. There are various methods for determining weights, which can be categorized into three main types based on different computational data: subjective weighting methods, objective weighting methods, and combined weighting methods. Subjective weighting methods include: Delphi method, AHP method, chain ratio scoring method, point estimation method, attribute importance ranking method, fuzzy subset method, judgment matrix method, order relation analysis method, and network analysis method. Objective weighting methods include: PCA method, factor analysis method, standard deviation method, EWM method, coefficient of variation method, grey relational analysis method, multi-objective programming method, CRITIC weighting method, and TOPSIS method. Combined weighting methods include: additive composition method, multiplicative composition method, range maximization method, distance function method, and least squares combined weighting.

[0003] Subjective weighting methods assign weights based on the decision-maker's subjective judgment. Their drawback is their reliance on the evaluator's subjective judgment, which introduces a degree of subjectivity. Objective weighting methods can fully utilize the information inherent in the original data, but suffer from poor universality and the tendency for weight values ​​to vary with sample data. Combined weighting methods combine subjective and objective weighting methods, commonly including additive and multiplicative methods. In additive combined weighting, the coefficients reflect the proportions of the weights obtained from subjective and objective weighting methods in the combined weights. A smaller difference between the two coefficients indicates stronger consistency in weighting. However, determining the combined coefficients requires considering the consistency of weights from each method and expert preferences, leading to poor model reliability. Multiplicative combined weighting is sensitive to numerical changes during multiplication and division operations, easily resulting in excessively large or small weight values. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a regional safety risk assessment method for mines, which solves the problems of strong subjectivity, poor universality, poor reliability, and excessive sensitivity to data in the existing technology.

[0005] To achieve the above objectives, the present invention provides the following solution: A method for regional safety risk assessment in mines, comprising: Construct a safety risk assessment index system for underground metal mines; The first weight set is obtained by assigning weights to the indicators in the security risk assessment indicator system using the DEMATEL method. The weights in the safety risk assessment index system are assigned using the Brown-Wood Method (BWM) to obtain a second weight set. The first weight set and the second weight set are combined and optimized using a pre-built weight optimization model to obtain the combined weights of the evaluation indicators; Build an evaluation standard cloud; The parameters of the evaluation standard cloud are calculated using the reverse cloud algorithm to obtain the index cloud parameters; The combined weights of the evaluation indicators and the cloud parameters of the indicators are fused and calculated using a pre-constructed comprehensive cloud parameter model to obtain comprehensive cloud feature parameters; The evaluation standard cloud and the comprehensive cloud feature parameters are input into the positive cloud generator to generate the cloud, thus obtaining the evaluation standard cloud and the comprehensive evaluation cloud. The evaluation standard cloud and the comprehensive evaluation cloud are matched according to the established grading rules to obtain the regional safety risk assessment results of the mine.

[0006] Preferably, the safety risk assessment index system includes: one primary index set and four secondary index sets; the primary index set includes: personnel factors, equipment factors, underground environment, and safety management; each primary index in the primary index set corresponds to a set of secondary indexes; the secondary index set corresponding to personnel factors includes: job tenure, education level, safety personnel staffing status, certification status, safety awareness score, and personal protective equipment status; the secondary index set corresponding to equipment factors includes: hoisting equipment, drainage equipment, power supply equipment, communication equipment, fire-fighting equipment, and personnel positioning equipment; the secondary index set corresponding to the underground environment includes: hydrogeological conditions, roof stability score, safety passages, effective air volume, lighting conditions, target toxic and harmful gases, and safety warning signs; the secondary index set corresponding to the safety management includes: safety production responsibility system, safety culture, safety investment, safety education and training, safety management system, accident emergency plan, safe operating procedures, and rectification of safety hazards.

[0007] Preferably, the expression for the DEMATEL method is: ;in, This is the first weight set; The centrality of the indicator.

[0008] Preferably, the indicators in the safety risk assessment indicator system are weighted using the Brown-Warshall Method (BWM) to obtain a second weight set, including: In the aforementioned safety risk assessment index system, optimal and worst-case indicators are set. The optimal indicators include: safety management, safety awareness score, power supply equipment, roof stability score, and safety management system. The worst-case indicators include: underground environment, length of service, hoisting equipment, lighting conditions, and safety culture. Construct a relative importance scale between the target optimal index and the indicators in the safety risk assessment index system excluding the target optimal index to obtain a first comparison vector; Construct a relative importance scale between the worst-case indicator of the target and the indicators in the safety risk assessment indicator system excluding the worst-case indicator of the target, and obtain a second comparison vector; The weights of the basic indicators are obtained by calculating the weights of the first comparison vector and the second comparison vector using the weight calculation formula. The weights of the basic indicators are calculated using a mathematical programming model to obtain the second weight set; the expression of the mathematical programming model includes: and ;in, The optimal indicator weight; Basic weights; It is the ratio of the optimal indicator to the target indicator; The worst-case indicator weight; It is the ratio of the target indicator to the worst-case indicator; This refers to the number of indicators.

[0009] Preferably, the expression of the weight optimization model includes: , , ;in, The combined weights of the evaluation indicators; , These are the weight allocation coefficients for the first combination and the weight allocation coefficients for the second combination, respectively. Weights for the DEMATEL method; The weights are determined by the BWM method. Assign coefficients to the i-th combination weights; Let these be the coefficients of the i-th optimal linear combination; This is the z-th weight.

[0010] Preferably, the expression for the evaluation criterion cloud is: ;in, This is the expected value; It is the entropy value; It is the hyperentropy value; , These are the upper and lower limits of the rating range, respectively; These are constant coefficients.

[0011] Preferably, the regional safety risk assessment results of the mine include: low risk level, general risk level, relatively high risk level, and major risk level; wherein, the scoring interval and standard cloud parameters of the low risk level are [90, 100] and (95, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters of the general risk level are [70, 90] and (80, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters of the relatively high risk level are [60, 70] and (65, 1.667, 0.1), respectively; and the scoring interval and standard cloud parameters of the major risk level are [0, 60] and (30, 10, 0.1), respectively.

[0012] Preferably, the expression for the integrated cloud parameter model includes: , , ;in, Let be the combined weight of the j-th indicator; Let j be the expected value of the j-th indicator; For the number of indicators; Let be the entropy of the j-th index; Let be the hyperentropy of the j-th index.

[0013] The present invention discloses the following technical effects: This invention provides a method for regional safety risk assessment in mines. By combining and optimizing the first and second weight sets through a weight optimization model, it solves the problems of strong subjectivity, poor universality, and poor reliability in existing technologies, and achieves the search for the optimal solution combination with the minimum deviation based on game theory. By integrating the weight combination of evaluation indicators and the indicator cloud parameters through a comprehensive cloud parameter model, it solves the problems of excessive sensitivity to data and misjudgment of levels when the data is close to the threshold in existing technologies, and achieves a balance between the fuzziness and randomness of mine risk indicators. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of a regional safety risk assessment process for mines provided in an embodiment of the present invention; Figure 2 Statistical graph of normal cloud model provided in embodiments of the present invention; Figure 3 The cloud generator provided in this embodiment of the invention; Figure 4 A flowchart of a safety risk assessment model for underground metal mines provided in an embodiment of the present invention; Figure 5 The chart shows the trend of weight ranking of evaluation indicators provided in the embodiments of the present invention. Detailed Implementation

[0015] Figure 1 This is a schematic diagram of a regional safety risk assessment process for mines provided in an embodiment of the present invention, such as... Figure 1 As shown, the present invention provides a method for regional safety risk assessment in mines, comprising: Step 100: Construct a safety risk assessment index system for underground metal mines; Step 200: Use the DEMATEL method to assign weights to the indicators in the security risk assessment indicator system to obtain the first weight set; Step 300: Use the BWM method to assign weights to the indicators in the safety risk assessment indicator system to obtain the second weight set; Step 400: Use a pre-built weight optimization model to perform combined optimization on the first weight set and the second weight set to obtain the combined weights of the evaluation indicators; Step 500: Construct an evaluation standard cloud; Step 600: Calculate the parameters of the evaluation standard cloud using the reverse cloud algorithm to obtain the indicator cloud parameters; Step 700: Use the pre-built comprehensive cloud parameter model to perform fusion calculation on the combined weights of the evaluation indicators and the cloud parameters of the indicators to obtain comprehensive cloud feature parameters; Step 800: Input the characteristic parameters of the evaluation standard cloud and the comprehensive cloud into the forward cloud generator to generate the cloud, and obtain the evaluation standard cloud and the comprehensive evaluation cloud; Step 900: Match the evaluation standard cloud and the comprehensive evaluation cloud according to the set grading rules to obtain the regional safety risk assessment results of the mine.

[0016] Preferably, the safety risk assessment index system includes: one primary index set and four secondary index sets; the primary index set includes: personnel factors, equipment factors, underground environment, and safety management; each primary index in the primary index set corresponds to a set of secondary indexes; the secondary index set corresponding to personnel factors includes: job tenure, education level, safety personnel staffing status, certification status, safety awareness score, and personal protective equipment status; the secondary index set corresponding to equipment factors includes: hoisting equipment, drainage equipment, power supply equipment, communication equipment, fire-fighting equipment, and personnel positioning equipment; the secondary index set corresponding to the underground environment includes: hydrogeological conditions, roof stability score, safety passages, effective air volume, lighting conditions, target toxic and harmful gases, and safety warning signs; the secondary index set corresponding to the safety management includes: safety production responsibility system, safety culture, safety investment, safety education and training, safety management system, accident emergency plan, safe operating procedures, and rectification of safety hazards.

[0017] Specifically, the expression for the DEMATEL method is: ;in, This is the first weight set; The centrality of the indicator.

[0018] Furthermore, the indicators in the aforementioned safety risk assessment indicator system are weighted using the Brown-Warshall Method (BWM) to obtain a second weight set, which includes: In the aforementioned safety risk assessment index system, optimal and worst-case indicators are set. The optimal indicators include: safety management, safety awareness score, power supply equipment, roof stability score, and safety management system. The worst-case indicators include: underground environment, length of service, hoisting equipment, lighting conditions, and safety culture. Construct a relative importance scale between the target optimal index and the indicators in the safety risk assessment index system excluding the target optimal index to obtain a first comparison vector; Construct a relative importance scale between the worst-case indicator of the target and the indicators in the safety risk assessment indicator system excluding the worst-case indicator of the target, and obtain a second comparison vector; The weights of the basic indicators are obtained by calculating the weights of the first comparison vector and the second comparison vector using the weight calculation formula. The weights of the basic indicators are calculated using a mathematical programming model to obtain the second weight set; the expression of the mathematical programming model includes: and ;in, The optimal indicator weight; Basic weights; It is the ratio of the optimal indicator to the target indicator; The worst-case indicator weight; It is the ratio of the target indicator to the worst-case indicator; This refers to the number of indicators.

[0019] Specifically, the expression of the weight optimization model includes: , , ;in, The combined weights of the evaluation indicators; , These are the weight allocation coefficients for the first combination and the weight allocation coefficients for the second combination, respectively. Weights for the DEMATEL method; The weights are determined by the BWM method. Assign coefficients to the i-th combination weights; Let these be the coefficients of the i-th optimal linear combination; This is the z-th weight.

[0020] Furthermore, the expression for the evaluation criterion cloud is: ;in, This is the expected value; It is the entropy value; It is the hyperentropy value; , These are the upper and lower limits of the rating range, respectively; These are constant coefficients.

[0021] Specifically, the regional safety risk assessment results for the mine include: low risk level, general risk level, relatively high risk level, and major risk level; wherein, the scoring interval and standard cloud parameters for the low risk level are [90, 100] and (95, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters for the general risk level are [70, 90] and (80, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters for the relatively high risk level are [60, 70] and (65, 1.667, 0.1), respectively; and the scoring interval and standard cloud parameters for the major risk level are [0, 60] and (30, 10, 0.1), respectively.

[0022] Furthermore, the expression for the integrated cloud parameter model includes: , , ;in, Let be the combined weight of the j-th indicator; Let j be the expected value of the j-th indicator; For the number of indicators; Let be the entropy of the j-th index; Let be the hyperentropy of the j-th index.

[0023] Specifically, the Dematel method is used for weighting. Decision Laboratory Analysis (DEMATEL) calculates the influence and affected degree of each factor on other factors by analyzing the logical relationships and direct influence matrices among the factors in a system, thereby determining the causal relationships between factors and the position of each factor in the system. When using the Dematel method, analysis is primarily based on the data and logical relationships within the system, using mathematical and statistical methods to calculate the influence relationships between factors. Therefore, from this perspective, the Dematel method is a relatively objective method. Its application is influenced by the researcher's understanding of the system, professional knowledge and experience, and the accuracy and completeness of the data. Weights are calculated based on the indicator centrality obtained from the Dematel method. w D =( w 1, w 2, , w n ): .

[0024] The Brown-Warner Method (BWM) weighting method allows decision-makers to determine their most and least concerned indicators based on their own circumstances. By comparing the relationships between these indicators and other indicators, an evaluation value can be derived. This method is well-suited for mine safety risk assessments involving complex production processes, variable environments, and numerous risk factors. It decomposes various risk factors into different levels, thus systematically addressing the problem. The commonly used Analytic Hierarchy Process (AHP) mathematizes the decision-making process through pairwise comparisons of indicators. However, when the number of evaluation indicators (n>9) increases, the number of comparisons becomes (n-1) / 2, significantly increasing the computational load and subjective randomness, leading to biased evaluation results. Compared to AHP, the BWM method also compares indicators, but through a structured comparison approach, the number of pairwise comparisons is reduced to 2n-3, simplifying the calculation process and resulting in more consistent results. The specific operational steps are as follows: 1) Determine the set of indicators C =( C 1, C 2, , C n The evaluation index system is divided into 4 primary indicators as 1 index set and 28 secondary indicators as 4 index sets, for a total of 5 index sets for solution, as shown in Table 1.

[0025] 2) Select the optimal indicator based on expert opinions. C B and worst-case indicators C W This embodiment identifies the optimal and worst-performing indicators from the five indicator sets, as shown in Table 1.

[0026] Table 1

[0027] 3) Scoring was conducted using a scale from level 1 to level 9, as shown in Table 2. Experts compared the optimal indicator with each of the remaining indicators to determine the relative importance scale of the optimal indicator compared to the other indicators, thus constructing a comparison vector. A B = ( a B1 , a B2 , , a Bn Experts then compared the remaining indicators with the worst-case indicator one by one to determine the relative importance scale of the other indicators compared to the worst-case indicator, and constructed a comparison vector.A W =( a W1 , a W2 , , a Wn The relative importance scale, i.e., the comparison of indicator weights, requires that the optimal weights satisfy the following condition: for any indicator i, the weights... w i have: .

[0028] Table 2

[0029] 4) Using the following mathematical programming model, obtain the optimal index weights. w B =( w 1, w 2, ..., w n ).

[0030] .

[0031] .

[0032] use CR The relative importance consistency ratio (RQR) measures the rationality of the expert relative importance scale selection, and is calculated as follows: .

[0033] in, CR ∈[0,1], CR The closer to 0, the higher the consistency. ξ The values ​​can be obtained from a linear programming model, as shown in Table 3. a BW Take different values ​​of the consistency index ( CI ).

[0034] Table 3

[0035] Preferably, the weights of the evaluation indicators are determined based on game theory. Game theory uses mathematical tools to abstract game models into mathematical problems, helping decision-makers find the optimal strategy to maximize their benefits in multi-party games. In recent years, game theory has been widely applied in determining indicator weights. Aiming at Nash equilibrium, it comprehensively considers the information between various indicators, finding consistency and compromise among multiple weighting methods to eliminate linear relationships between indicators and reduce the influence of subjectivity. This embodiment first uses the Decision Laboratory Analysis (DEMATEL) method and the Best-Worst Method (BWM) to determine the weights of the underground metal mine safety risk assessment indicators. Then, it uses the game theory combination weighting method to find the optimal solution combination with the smallest deviation, achieving an optimized combination of weights to obtain the combined weight values ​​of the indicators, making the determination of indicator weights more reasonable and scientific. This embodiment constructs a game theory weight optimization model, with the specific steps as follows: 1) Combine and optimize the weights obtained by the Dematel and Brownian methods to minimize the deviation of the weights, which is equivalent to transforming it into the optimization of the first derivative form of a system of linear equations, i.e.: .

[0036] 2) Normalize the calculated optimal linear combination coefficients using the previous formula to obtain the combination weight allocation coefficients. β 1 and β 2: .

[0037] 3) Calculate the combined weights of the evaluation indicators. W : .

[0038] In the formula: W For combined weights, w D For the DEMATEL method weights, w B The weights are determined by the BWM method.

[0039] Furthermore, the cloud model theory for safety risk assessment. 1) The concept and numerical characteristics of the cloud model. In practical risk assessment, complex dynamic problems involving nonlinear multivariates are frequently encountered. Many risk factors exhibit high uncertainty, manifesting as randomness and fuzziness. Randomness refers to the uncertainty of whether an event will occur under certain objective environmental conditions, described by the probability of its occurrence; fuzziness refers to the uncertainty of events that have already occurred but are difficult to describe precisely, measured by membership degrees. Fuzzy theory uses membership degrees to describe the degree of fuzziness of events; however, when membership degrees are set to precise values, fuzziness disappears, thus presenting limitations. To compensate for the shortcomings of fuzzy theory and more efficiently handle problems involving fuzziness and randomness, researchers proposed the cloud model theory in 1995 based on fuzzy mathematics and probability theory. This theory establishes a mapping relationship between qualitative concepts and quantitative characteristics to achieve the mutual conversion between qualitative concepts and quantitative values. In the safety risk assessment of underground metal mines, the cloud model allows risk levels to be presented in a more intuitive and clear cloud map form, providing an effective mathematical model for solving the uncertainty problem in the assessment process. The cloud model uses digital features to quantitatively describe qualitative concepts, namely: 1) Expectations ( Ex ):expect Ex It is the center value of the cloud droplet distribution on the universe of discourse U, and is the point that best expresses the qualitative concept. The membership degree of this position is 1.

[0040] 2) Entropy ( En ):entropy En It is a mathematical description of the range of uncertainty of a qualitative concept, representing the degree of dispersion of cloud droplets. It is determined by the randomness and fuzziness of the qualitative concept. The larger the value, the larger the range of cloud droplets that the qualitative concept can satisfy, and the more fuzzy the qualitative concept becomes.

[0041] 3) Hyperentropy ( He ): Hyperentropy He It is the entropy of entropy, determined by the randomness and fuzziness of entropy. He This reflects the uncertainty of entropy, the cohesiveness of cloud droplet uncertainty in the qualitative concept, and determines the thickness and dispersion of the cloud. The larger the value, the greater the dispersion of the cloud, the greater the randomness of the membership degree, and the greater the thickness of the cloud.

[0042] Specifically, the normal cloud model. The normal cloud model is the most important cloud model, such as... Figure 2 As shown, it is based on the universality of the normal membership function, and its expected curve is a normal curve, defined as follows. y =exp[-( x - Ex ) 2 / 2( En ) 2 ] is a normal cloud (X , Y The expected curve of the normal cloud model. The steps for generating the normal cloud model are as follows: 1) Generation Ex For expected value, En Normal random numbers with standard deviation x i ; 2) Generation En For expected value, He Normal random numbers with standard deviation En i ; 3) Calculation y i =exp[-( x i - Ex ) 2 / 2( En i ) 2 ],make( x i , y i () represents cloud droplets.

[0043] 4) Repeat the above steps until generated. n Until a single cloud droplet.

[0044] refer to Figure 3 A cloud generator is a cloud generation algorithm that can be implemented through software and hardware. Functionally, it can be divided into forward cloud generators and reverse cloud generators. A forward cloud generator is a forward process that generates the required three cloud digital features (…). Ex , En , He ) and cloud droplet count n Input a cloud generator, output the coordinates of the cloud droplets in the universe of discourse. x membership degree μ Ã ( x The inverse cloud generator is the reverse process of the forward cloud generator. It takes cloud droplets conforming to a certain distribution as input and outputs three numerical features corresponding to that cloud model. Ex , En , He ).

[0045] Furthermore, the steps for using cloud models to conduct security risk assessments are as follows: 1) Construct an evaluation standard cloud: By consulting with experts in the field, the evaluation criteria levels were determined and the scoring ranges were divided. Q min , Q max ] Calculate standard cloud parametersC =( Ex , En , He ): .

[0046] In the formula: Q max , Q min These represent the upper and lower limits of the rating range, respectively. k It is a constant, and is adjusted according to the degree of ambiguity of the indicator.

[0047] 2) Calculate the cloud parameters of the indicator: The expected value of each indicator is calculated using the reverse cloud algorithm. Ex j Entropy En j hyperentropy He j : .

[0048] .

[0049] .

[0050] In the formula: q The number of samples for the indicator. The sample variance of the indicator.

[0051] 3) Calculate the comprehensive cloud parameters: Based on the indicator cloud feature parameters ( Ex j , En j , He j ) and combined weights W j Synthesize the data to obtain comprehensive cloud feature parameters ( Ex , En , He ): .

[0052] 4) Drawing cloud maps and comparing cloud maps: Using MATLAB software, standard cloud parameters and comprehensive cloud parameters are input into the forward cloud generator to generate evaluation standard cloud and evaluation comprehensive cloud. The generated cloud map can be used to intuitively determine the safety risk level.

[0053] Preferably, the evaluation indicators are classified into safety risk levels. Based on on-site investigations of underground metal mine safety management, fundamental data from actual mine production, and monitoring data, and considering various factors affecting the safety risk level of underground metal mines, the evaluation indicators are qualitatively and quantitatively classified, referencing relevant research findings on mine safety risk level classification and some industry risk grading standards, to better study the safety risk level of underground metal mines. The safety risk levels are divided into four levels: low risk (Level I), general risk (Level II), relatively high risk (Level III), and major risk (Level IV). For the six qualitative indicators F5 (safety awareness), F14 (hydrological conditions), F15 (roof stability), F19 (toxic and harmful gases), F21 (safety production responsibility system), and F22 (safety culture), qualitative language is used for grading. Mine safety management personnel and industry experts classify the safety risk level of these indicators based on actual conditions. For the remaining 22 quantitative indicators, values ​​are assigned and graded based on actual values. Referring to relevant literature, this embodiment proposes a method for classifying the safety risk level of underground metal mines, as shown in Table 4.

[0054] Table 4

[0055] refer to Figure 4 An example application of a safety risk assessment model based on a game theory-based combinatorial weighted cloud model. The safety assessment process for underground metal mines using the constructed game theory-based combinatorial weighted cloud model is as follows: Figure 4As shown. Mine Overview: This example uses an underground gold mine as an example to conduct a safety risk assessment. The mine has a complete production management system. It is surrounded by numerous large and medium-sized mines and comprises three independent underground mining systems. The total permitted production capacity is 2.97 million tons per year, with nearly 3,000 employees. The mine operates 330 days a year, with three shifts per day, each shift lasting 8 hours. The terrain of the mining area slopes from east to west, with a mild climate. Rainfall is concentrated between July and September, with an average annual rainfall of 650 mm. The seismic intensity is VII, indicating a relatively stable crustal region. There are no large water systems, but incised gullies are well-developed. Rainfall occurs during the rainy season, while the soil is often dry during the dry season, having minimal impact on the mining area. The mine employs an upward horizontal stratified approach tailings cemented backfilling mining method for underground gold mining, producing gold concentrate. The mine's development and hoisting systems utilize vertical shafts and auxiliary inclined ramps. The transportation system includes inclined ramps and trackless transportation systems, as well as horizontal tunnel transportation using rail transport and belt conveyors. The system employs a centralized drainage system, with a pump house and water tank constructed at -630m. It includes a blasting material storage area with explosives chambers and detonator chambers. A mechanical exhaust ventilation system with a two-wing diagonal ventilation configuration is used. The power supply system loads include: the main drainage pump, main ventilation system, blind shaft hoist, ore recovery shaft hoist, skip loading facilities, mining machinery, and underground lighting. The fire-fighting water supply system uses a centralized system, with fire-fighting water stored in a high-level surface pool. The blasting material storage area and other locations are equipped with fire-fighting equipment such as fire-fighting sand and fire extinguishers. The six major safety and disaster prevention systems are well-established. The monitoring and control system includes toxic and harmful gas monitoring, ventilation system monitoring, and video surveillance. The personnel positioning system can realize real-time and accurate positioning of underground personnel and provide rescue alarms in case of safety accidents. The emergency refuge system includes disaster refuge chambers and self-rescue devices. The compressed air self-rescue system is built in conjunction with the production air supply system. The water supply rescue system relies on the production and domestic water supply network. The communication system mainly includes wired communication system, wireless communication system, and emergency broadcast communication system. A safety production management organization has been established and full-time safety production management personnel have been assigned, all of whom are certified and have a safety production responsibility system, safety management system, and job safety operating procedures. There is a dual prevention system of risk classification and control and hidden danger investigation and management.

[0056] Furthermore, the evaluation criteria were determined by consulting with experts in the field and mine safety management personnel, and scoring ranges were divided according to four levels of safety risk. He 0 = 0.1, and the standard cloud parameters are obtained, as shown in the table below. Input the standard cloud parameters corresponding to the four risk levels into the positive cloud generator, setting N = 3000, to generate the evaluation standard cloud. See Table 5 for the safety risk level scoring range and standard cloud parameters.

[0057] Table 5

[0058] The range of security risk levels shown by the cloud model is quite intuitive. The security risk levels of the cloud in the risk assessment standard are low risk (Level I), general risk (Level II), relatively high risk (Level III), and major risk (Level IV).

[0059] Furthermore, the combined weights are calculated. The weight calculation results of the BWM method are shown in Table 6.

[0060] Table 6

[0061] As shown in the table, in the weighting calculation results of the BM method, the five secondary indicators F24 (safety education and training), F23 (safety investment), F5 (safety awareness), F6 (personal safety protection), and F4 (certification for employment) rank in the top five, indicating they are in a relatively important position. The three secondary indicators F18 (lighting conditions), F7 (lifting equipment), and F20 (safety warning signs) rank last, suggesting they are not considered important by decision-makers.

[0062] The index weights were obtained using the Demetal method and the Brown method, and the combined weight allocation coefficients were calculated using the formula. β 1 = 0.224 β 2=0.776, and the final combined weights are shown in Table 7. The trend of the ranking of evaluation index weights is shown in the figure below.

[0063] Table 7

[0064] like Figure 5 As shown, the trend of the combined weights effectively balances the results of the DEMATEL and BWM weights, preventing the weight ranking from being too high or too low and achieving a more balanced result. For example, F28 (safety hazard rectification) ranks 4th in the DEMATEL method and 14th in the BWM method. After combining the weights of the two methods, the weight ranking is 12th. In actual production activities, safety hazard rectification is of great importance, demonstrating that game theory combined weights can yield more scientific weight values.

[0065] Specifically, a safety risk assessment based on a cloud model is conducted. Utilizing the safety risk assessment index system for underground metal mines and the calculated weighted values ​​of the evaluation index combinations, this embodiment will verify the safety risk assessment model based on a game theory-based weighted cloud model for this underground gold mine. Quantitative index evaluation values ​​are derived from data collected during on-site surveys of the underground gold mine. Three experts related to underground gold mines are invited to score the qualitative evaluation indicators according to grading standards, resulting in qualitative index evaluation values. The cloud parameters of the indicators are calculated using a formula, and the data is summarized in Table 8. From the secondary index cloud parameters, it can be seen that the evaluation scores for F1 (job tenure), F2 (educational level), F5 (safety awareness), F13 (monitoring facilities), and F18 (lighting conditions) are within the range of [70, 80], indicating that the safety risk level of the above indicators leans towards general risk and requires attention in daily management. The comprehensive cloud parameters (…) are calculated using a formula. Ex , En , He = (87.139, 2.765, 1.110), input the comprehensive cloud parameters into the positive cloud generator in MATLAB to generate the evaluation comprehensive cloud, and compare it with the evaluation standard cloud, as shown in Table 8.

[0066] Table 8

[0067] The security risk levels of standard clouds, from left to right, are low risk (Level I), moderate risk (Level II), significant risk (Level III), and major risk (Level IV). The cloud droplets of integrated clouds are mainly within the evaluation score range of [75, 95], and are concentrated in […]. Ex The area around 87.139 largely overlaps with Standard Cloud II. Based on the principle of maximum membership, the safety risk level of this underground gold mine is determined to be general risk (Level II), which is consistent with the actual situation of the mine. This demonstrates that using a safety risk assessment model based on a game theory-based combinatorial weighted cloud model to evaluate the safety risk of underground metal mines is feasible.

[0068] Furthermore, the fuzzy comprehensive evaluation method is compared and verified. The fuzzy comprehensive evaluation method (FCE) is a comprehensive evaluation method based on fuzzy mathematics theory. It transforms qualitative evaluation into quantitative evaluation and uses membership theory for judgment. It is simple to operate and successfully solves the problem of complex indicator evaluation systems, and is widely used in many fields. However, it requires expert judgment to handle membership degrees and membership functions, thus the determination of evaluation indicator weights is subjective. The fuzzy comprehensive evaluation method has broad applicability in solving multi-factor, multi-level problems. Therefore, this embodiment uses the fuzzy comprehensive evaluation method, as a traditional evaluation method, and compares the results with the cloud model analysis to verify the effectiveness of the method. The analysis steps of the fuzzy comprehensive evaluation method are as follows: 1) Establish the factor set: Based on the factors influencing safety risks in underground metal mines, a fuzzy comprehensive evaluation factor set is constructed: .

[0069] 2) Establish the weight set: The weights of each evaluation factor are used W express: .

[0070] 3) Establish an evaluation set for risk assessment: .

[0071] 4) Single-factor evaluation: For a single factor in factor set U, analyze the membership degree of subsets within the factor and construct the evaluation matrix of the subset. R , is represented as: .

[0072] 5) Fuzzy comprehensive evaluation: Utilize the corresponding weights of evaluation factors at each level W Membership matrix of evaluation factors X Multiplying them together yields the fuzzy comprehensive membership matrix. Y After normalization, the final comprehensive risk assessment membership matrix is ​​obtained. .

[0073] Y = W × X。

[0074] .

[0075] 6) Analyze the fuzzy comprehensive evaluation vector: The commonly used method is the maximum membership principle, but when the number of influencing factors on the evaluated object is large, some known information may be lost, potentially leading to inaccurate evaluation results. This embodiment uses the final comprehensive risk assessment membership vector... As a weighting and commentary set V T Multiply by the product to obtain the final fuzzy comprehensive evaluation score. Z As shown in the following formula, all evaluation information can be effectively utilized, thereby overcoming the shortcomings of the maximum membership principle.

[0076] .

[0077] The membership degree function is obtained by using an empirical formula based on triangular membership functions to calculate the membership degree of the secondary evaluation indicators to the safety risk level, as shown in Table 9. Then, the membership degree matrix of the four primary indicators is obtained by multiplying the weights of the secondary indicators by the membership degrees. X As shown below: .

[0078] After calculation = (0.303, 0.566, 0.131, 0, 0), the fuzzy comprehensive evaluation comment set is obtained based on the median of the risk level interval. V ={95, 85, 75, 65, 30}, the fuzzy comprehensive evaluation value is calculated using the formula. Z =86.716. Comparing this to the safety risk level range in the table below, the fuzzy comprehensive evaluation method indicates a low risk, consistent with the cloud model evaluation result, thus verifying the accuracy of the game theory-based weighted cloud model evaluation result. However, since the membership function of the fuzzy comprehensive evaluation method is an empirical formula with significant subjectivity, it can reduce the accuracy of the evaluation result. This embodiment utilizes cloud model theory to determine the membership function, effectively addressing the problem of membership randomness and making the evaluation result more reliable.

[0079] Table 9

[0080] The beneficial effects of this invention are as follows: This invention optimizes the combination of the first and second weight sets using a weight optimization model. Based on game theory principles, it finds the optimal solution combination with the minimum deviation, reducing the influence of subjectivity and improving the reliability of the weight indicators. By integrating the evaluation indicator combination weights and indicator cloud parameters using a comprehensive cloud parameter model, it takes into account the fuzziness and randomness of mine risk indicators, improves the accuracy of risk level determination, and avoids misjudgment of the level when it is close to the threshold.

Claims

1. A method for regional safety risk assessment in mines, characterized in that, include: Construct a safety risk assessment index system for underground metal mines; The first weight set is obtained by assigning weights to the indicators in the security risk assessment indicator system using the DEMATEL method. The weights in the safety risk assessment index system are assigned using the Brown-Wood Method (BWM) to obtain a second weight set. The first weight set and the second weight set are combined and optimized using a pre-built weight optimization model to obtain the combined weights of the evaluation indicators; Build an evaluation standard cloud; The parameters of the evaluation standard cloud are calculated using the reverse cloud algorithm to obtain the index cloud parameters; The combined weights of the evaluation indicators and the cloud parameters of the indicators are fused and calculated using a pre-constructed comprehensive cloud parameter model to obtain comprehensive cloud feature parameters; The evaluation standard cloud and the comprehensive cloud feature parameters are input into the positive cloud generator to generate the cloud, thus obtaining the evaluation standard cloud and the comprehensive evaluation cloud. The evaluation standard cloud and the comprehensive evaluation cloud are matched according to the established grading rules to obtain the regional safety risk assessment results of the mine.

2. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The safety risk assessment index system includes: one primary index set and four secondary index sets; the primary index set includes: personnel factors, equipment factors, underground environment, and safety management; each primary index in the primary index set corresponds to a set of secondary indexes; the secondary index set corresponding to personnel factors includes: job tenure, education level, safety personnel staffing status, certification status, safety awareness score, and personal protective equipment status; the secondary index set corresponding to equipment factors includes: hoisting equipment, drainage equipment, power supply equipment, communication equipment, fire-fighting equipment, and personnel positioning equipment; the secondary index set corresponding to the underground environment includes: hydrogeological conditions, roof stability score, safety passages, effective air volume, lighting conditions, target toxic and harmful gases, and safety warning signs; the secondary index set corresponding to the safety management includes: safety production responsibility system, safety culture, safety investment, safety education and training, safety management system, accident emergency plan, safe operating procedures, and rectification of safety hazards.

3. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The expression for the DEMATEL method is: ;in, This is the first weight set; The centrality of the indicator.

4. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The second weight set is obtained by assigning weights to the indicators in the aforementioned safety risk assessment indicator system using the Brown-Warshall Method (BWM), including: In the aforementioned safety risk assessment index system, optimal and worst-case indicators are set. The optimal indicators include: safety management, safety awareness score, power supply equipment, roof stability score, and safety management system. The worst-case indicators include: underground environment, length of service, hoisting equipment, lighting conditions, and safety culture. Construct a relative importance scale between the target optimal index and the indicators in the safety risk assessment index system excluding the target optimal index to obtain a first comparison vector; Construct a relative importance scale between the worst-case indicator of the target and the indicators in the safety risk assessment indicator system excluding the worst-case indicator of the target, and obtain a second comparison vector; The weights of the basic indicators are obtained by calculating the weights of the first comparison vector and the second comparison vector using the weight calculation formula. The weights of the basic indicators are calculated using a mathematical programming model to obtain the second weight set; the expression of the mathematical programming model includes: and ;in, The optimal indicator weight; Basic weights; It is the ratio of the optimal indicator to the target indicator; The worst-case indicator weight; It is the ratio of the target indicator to the worst-case indicator; This refers to the number of indicators.

5. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The expression for the weight optimization model includes: , , ;in, The combined weights of the evaluation indicators; , These are the weight allocation coefficients for the first combination and the weight allocation coefficients for the second combination, respectively. Weights for the DEMATEL method; The weights are determined by the BWM method. Assign coefficients to the i-th combination weights; Let these be the coefficients of the i-th optimal linear combination; This is the z-th weight.

6. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The expression for the evaluation standard cloud is: ;in, This is the expected value; It is the entropy value; It is the hyperentropy value; , These are the upper and lower limits of the rating range, respectively; These are constant coefficients.

7. The method for regional safety risk assessment in mines according to claim 1, characterized in that, The regional safety risk assessment results for the mine include: low risk level, general risk level, relatively high risk level, and major risk level; wherein, the scoring interval and standard cloud parameters for the low risk level are [90, 100] and (95, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters for the general risk level are [70, 90] and (80, 1.667, 0.1), respectively; the scoring interval and standard cloud parameters for the relatively high risk level are [60, 70] and (65, 1.667, 0.1), respectively; and the scoring interval and standard cloud parameters for the major risk level are [0, 60] and (30, 10, 0.1), respectively.

8. The method for regional safety risk assessment in mines according to claim 6, characterized in that, The expression for the integrated cloud parameter model includes: , , ;in, Let be the combined weight of the j-th indicator; Let j be the expected value of the j-th indicator; For the number of indicators; Let be the entropy of the j-th index; Let be the hyperentropy of the j-th index.

Images