Heavy haul railway construction safety risk assessment method and device

Abstract

The invention relates to a heavy haul railway construction safety risk assessment method and device. The method comprises the following steps: establishing a risk index system for heavy haul railway business line construction safety, and constructing a risk assessment network for describing risk propagation based on the risk index system; the method comprises the following steps: determining the importance of each index in a risk index system by using a cloud model fuzzy algorithm, and generating an index weight cloud model; determining the fuzzy condition probability of the risk assessment network by combining the quantitative weight and the qualitative importance of each index and the logic correlation among the indexes; the risk occurrence probability of each index is obtained, the risk occurrence probability is mapped into a fuzzy probability based on a triangular fuzzy number model, and the prior probability of the risk assessment network is obtained after defuzzification processing; and calculating the risk probability of each index in the risk assessment network based on the risk index system, the fuzzy condition probability and the prior probability to obtain a risk assessment result. By adopting the method, the accuracy and timeliness of security risk assessment can be effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing for safety evaluation, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for heavy-haul railway construction safety risk assessment. Background Art

[0002] As a vital piece of transport infrastructure, heavy-haul railways carry a vast amount of freight and play a vital role in economic development. In recent years, with the continued growth in freight demand, the workload of heavy-haul railway line renovation, expansion, and maintenance has become increasingly demanding. This has also led to an increase in safety incidents, seriously impacting railway transportation safety.

[0003] Traditional methods for on-site construction management in heavy-haul railway operations are often based on empiricism. These methods lack a deep understanding of the risks inherent in operating line construction and lack systematic and widespread application of scientific risk management and control methods. Developing risk assessment methods tailored to the specific characteristics of heavy-haul railway construction to improve railway safety is a pressing issue. Summary of the Invention

[0004] Based on this, it is necessary to provide a heavy-duty railway construction safety risk assessment method, device, computer equipment, computer-readable storage medium and computer program product that can improve the efficiency and effectiveness of risk assessment in response to the above technical problems.

[0005] In a first aspect, the present application provides a heavy-haul railway construction safety risk assessment method, the method comprising:

[0006] Establish a risk indicator system for construction safety of heavy-haul railway operating lines, and construct a risk assessment network based on the risk indicator system to describe risk propagation;

[0007] By using a cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined to generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0008] Determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0009] Obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification;

[0010] Based on the risk indicator system, the fuzzy conditional probability and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

[0011] In one embodiment, the cloud model fuzzy algorithm is used to determine the importance of each indicator in the risk indicator system and generate an indicator weight cloud model, including:

[0012] Obtaining the importance score results of each of the indicators according to the numerical range corresponding to the qualitative importance language variable;

[0013] Calculate the cloud model parameters of each indicator by using the cloud model fuzzy algorithm and the importance score results of each indicator;

[0014] The ambiguity parameter is determined according to the cloud model parameter of each of the indicators, and the indicator weight cloud model is obtained when the ambiguity parameter meets the ambiguity condition.

[0015] In one embodiment, the calculating of cloud model parameters of each of the indicators includes:

[0016] Determining cloud model parameters of the last-level indicators in the indicator hierarchy; the cloud model parameters include cloud model expectation, cloud model entropy, and cloud model hyperentropy;

[0017] According to the cloud model parameters of the last-layer indicators, the cloud model parameters of the upper-level indicators in the indicator hierarchy are calculated in sequence as the cloud model parameters of each of the indicators.

[0018] In one embodiment, before the step of determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators, the method further includes:

[0019] Calculation is performed based on the cloud model parameters and membership functions of the indicators to obtain the quantitative weight and qualitative importance of the indicators.

[0020] In one embodiment, the determining of the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators includes:

[0021] Determine the sub-indicators and parent indicators included in each indicator based on the logical relationships between the indicators;

[0022] Determine the conditional probability of occurrence of any indicator's sub-indicators through AND gate operations between the parent indicators corresponding to the sub-indicators of any indicator;

[0023] Based on the conditional probability of occurrence of the sub-indicators in each of the indicators, the fuzzy conditional probability of the risk assessment network is obtained.

[0024] In one embodiment, the step of obtaining the risk occurrence probability of each indicator, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification includes:

[0025] Obtaining an indicator occurrence probability language variable according to the risk description information of each of the indicators as the risk occurrence probability of each of the indicators;

[0026] By establishing a mapping relationship between linguistic variables and triangular fuzzy numbers, the risk occurrence probability of each indicator is converted into the fuzzy probability;

[0027] The membership function of triangular fuzzy numbers is used to obtain the prior probability of the risk assessment network through defuzzification.

[0028] In a second aspect, the present application further provides a heavy-haul railway construction safety risk assessment device, the device comprising:

[0029] A network construction module is used to establish a risk indicator system for the construction safety of heavy-haul railway operating lines, and to construct a risk assessment network for describing risk propagation based on the risk indicator system;

[0030] A cloud model generation module is used to determine the importance of each indicator in the risk indicator system by using a cloud model fuzzy algorithm and generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0031] A fuzzy conditional probability determination module is used to determine the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0032] A priori probability acquisition module is used to obtain the risk occurrence probability of each of the indicators, map the risk occurrence probability into fuzzy probability based on a triangular fuzzy number model, and obtain the priori probability of the risk assessment network after defuzzification;

[0033] The risk assessment result obtaining module is used to calculate the risk probability of each indicator in the risk assessment network based on the risk indicator system, the fuzzy conditional probability and the prior probability to obtain a risk assessment result.

[0034] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0036] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0037] The above-mentioned heavy-haul railway construction safety risk assessment method, device, computer equipment, computer-readable storage medium and computer program product establish a risk indicator system for the construction safety of heavy-haul railway operating lines, construct a risk assessment network for describing risk propagation based on the risk indicator system, and then use a cloud model fuzzy algorithm to determine the importance of each indicator in the risk indicator system and generate an indicator weight cloud model. The indicator weight cloud model meets the ambiguity condition under natural language processing. Combined with the quantitative weight and qualitative importance of each indicator, as well as the logical relationship between each indicator, the fuzzy conditional probability of the risk assessment network is determined to obtain the risk occurrence probability of each indicator. The risk occurrence probability is mapped to fuzzy probability based on a triangular fuzzy number model. After defuzzification, the prior probability of the risk assessment network is obtained. Then, based on the risk indicator system, fuzzy conditional probability and prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result, thereby achieving optimization of the heavy-haul railway construction safety risk assessment process, avoiding the influence of subjective judgment, effectively assessing the risk status, and improving the accuracy and timeliness of heavy-haul railway construction safety risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 1 is a flow chart of a heavy-haul railway construction safety risk assessment method according to an embodiment;

[0040] Figure 2 Schematic diagram of a comprehensive evaluation process based on fuzzy evaluation and Bayesian reasoning in one embodiment;

[0041] Figure 3 is a schematic diagram of importance scatter distribution in one embodiment;

[0042] Figure 4A schematic flow chart of a heavy-haul railway construction safety risk assessment method in another embodiment;

[0043] Figure 5 This is a structural block diagram of a heavy-haul railway construction safety risk assessment device in one embodiment;

[0044] Figure 6 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0046] In an exemplary embodiment, Figure 1 As shown, a method for assessing the safety risk of heavy-haul railway construction is provided. This embodiment uses the method applied to a terminal as an example. It is understood that the method can also be applied to a server, or to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps 101 to 105. Among them:

[0047] Step 101: establishing a risk indicator system for construction safety of heavy-haul railway operating lines, and constructing a risk assessment network for describing risk propagation based on the risk indicator system.

[0048] As an example, the risk assessment network can be obtained based on a Bayesian network structure.

[0049] In practical applications, such as Figure 2 As shown in Figure 2, an evaluation index system (i.e., risk index system) can be established based on the actual risks of heavy-haul railway operating line construction, as shown in Figure 2. , and can be classified according to their mutual relationships to establish a Bayesian network describing risk propagation (i.e., risk assessment network), which can be used to evaluate the risk level of sudden accidents on heavy-haul railways.

[0050] For example, the main safety risk factors that exist when the train operation section dispatches and coordinates the construction of heavy-load railway operating lines can be identified, and then a multi-level evaluation indicator system can be established according to the impact relationship.

[0051] Step 102: Determine the importance of each indicator in the risk indicator system by using a cloud model fuzzy algorithm, and generate an indicator weight cloud model.

[0052] Among them, the indicator weight cloud model meets the ambiguity conditions under natural language processing. If the value of ambiguity CD is within the preset range, it can be used to represent the condensation of expert opinions.

[0053] In the specific implementation, such as Figure 2 As shown in the figure, we can score the importance of each observation indicator (i.e., each indicator in the risk indicator system) to generate an observation indicator weight cloud model, and judge whether the ambiguity requirement is met to determine whether the expert opinions are cohesive.

[0054] Optionally, the importance of the indicators can be scored by experts, and the cloud model fuzzy algorithm can be used to achieve quantitative weight scoring and qualitative importance evaluation of the risk indicator system; for example, based on the prior qualitative importance language variable The corresponding numerical range can be obtained by experts scoring the importance of the observation indicators. , by mapping Calculate the cloud model parameters and use the ambiguity CD to determine whether the expert opinions are condensed. If the result is no, score again. If the result is yes, the cloud model parameters of the upper-level indicators can be generated in sequence based on the cloud model parameters of the observed indicators to determine the quantitative weights of all indicators. and qualitative importance.

[0055] Step 103 : Determine the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators.

[0056] In one example, if Figure 2 As shown, the quantitative weights and qualitative importance of all indicators can be obtained. Based on the quantitative weight C of the indicator system m The interaction relationship between the parameters and the indicators can determine the fuzzy conditional probability CPT of the Bayesian network.

[0057] Step 104 , obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification.

[0058] Specifically, if Figure 2 As shown in the figure, the possibility of the occurrence of the observation indicators reported qualitatively by front-line personnel (i.e., the probability of risk occurrence) can be obtained. By converting the qualitative linguistic variables into fuzzy probabilities based on the triangular fuzzy number model, the fuzzy variables can be used as the prior probability of the Bayesian network after defuzzification. For example, based on the qualitative description of the risk of the observation indicators by front-line staff, the probability of the occurrence of the observation indicators can be obtained. , the prior fuzzy probability of the observation index can be obtained by mapping g(V) , after defuzzification, we get the prior probability A.

[0059] Step 105 : Based on the risk indicator system, the fuzzy conditional probability, and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

[0060] In one example, the risk indicator system U, fuzzy conditional probability CPT, and prior fuzzy probability , calculate the posterior exact probability B of each indicator in the Bayesian network, as well as the risk probability of the final indicator and the final evaluation result (i.e., risk assessment result), such as Figure 2 As shown in the figure, a Bayesian network is formed and a comprehensive safety evaluation is performed.

[0061] In an optional embodiment, the heavy-duty railway construction safety risk assessment method in this embodiment can effectively avoid the uncertainty brought by the direct use of natural language by using fuzzy methods in the early network indicator weight calculation and prior probability acquisition process, and through the targeted selection of experts to construct the network and the selection of front-line personnel to determine the prior probability, it can ensure the accuracy of the assessment network itself and the timeliness of obtaining prior information, thereby effectively improving the accuracy and timeliness of risk assessment.

[0062] Compared with traditional methods, the technical solution of this embodiment introduces a scientific and rapid assessment method for construction safety risks of operating lines. It quantifies the expert evaluation weight data and the qualitative data of on-site qualitative data through fuzzy methods, and calculates the probability of risk occurrence through Bayesian networks, effectively avoiding the influence of subjective judgment and being able to effectively assess the risk status. Compared with the traditional method of relying on experience for safety evaluation, it is more accurate and reliable, and improves the level of on-site safety management.

[0063] In the above-mentioned heavy-haul railway construction safety risk assessment method, a risk indicator system for the construction safety of heavy-haul railway operating lines is established, and a risk assessment network for describing risk propagation is constructed based on the risk indicator system. Then, by utilizing the cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined, and an indicator weight cloud model is generated. Combined with the quantitative weight and qualitative importance of each indicator, as well as the logical relationship between each indicator, the fuzzy conditional probability of the risk assessment network is determined, and the risk occurrence probability of each indicator is obtained. The risk occurrence probability is mapped to the fuzzy probability based on the triangular fuzzy number model, and the prior probability of the risk assessment network is obtained after defuzzification. Then, based on the risk indicator system, fuzzy conditional probability, and prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain the risk assessment result, thereby realizing the optimization of the heavy-haul railway construction safety risk assessment process, avoiding the influence of subjective judgment, effectively assessing the risk status, and improving the accuracy and timeliness of the heavy-haul railway construction safety risk assessment.

[0064] In an exemplary embodiment, determining the importance of each indicator in the risk indicator system by using a cloud model fuzzy algorithm and generating an indicator weight cloud model may include the following steps:

[0065] According to the numerical range corresponding to the qualitative importance linguistic variable, the importance score result of each of the indicators is obtained; the cloud model parameters of each of the indicators are calculated using the cloud model fuzzy algorithm and the importance score result of each of the indicators; the ambiguity parameter is determined according to the cloud model parameter of each of the indicators, and the indicator weight cloud model is obtained when the ambiguity parameter meets the ambiguity condition.

[0066] In practical applications, the set of language variables C for the importance of observation indicators can be: C: {c1 = not very important (0-0.2), c2 = less important (0.2-0.4), c3 = general importance (0.4-0.6), c4 = relatively important (0.6-0.8), c5 = very important (0.8-1.0),}. Based on the qualitative importance language variable C and the corresponding numerical range, the observation indicators can be initially scored. Then, based on the reverse cloud model fuzzy algorithm, the cloud model parameter E of each observation indicator can be reversely calculated from the expert's importance score x. x 、E n and H e , and calculate the confusion CD value.

[0067] For example, mapping The cloud model parameter E can be calculated as follows: x 、E n and H e :

[0068] (1)

[0069] (2)

[0070] (3)

[0071] Among them, E x For the cloud model expectation, E n is entropy, H e It is super entropy.

[0072] The calculation method of confusion CD is as follows:

[0073] (4)

[0074] When the value of ambiguity CD is within the range of (0.2, 0.5044], it can be considered that the experts have reached a consensus on the importance of a certain indicator, that is, the concept has been condensed. Otherwise, a new round of scoring and measurement process will continue.

[0075] In one example, the cloud model expectation E for each indicator under the same category can be xi Perform normalization to obtain the weight W i , can be calculated as follows:

[0076] (5)

[0077] In an exemplary embodiment, the step of calculating the cloud model parameters of each indicator may include the following steps:

[0078] Determine the cloud model parameters of the last-level indicators in the indicator hierarchy; the cloud model parameters include cloud model expectation, cloud model entropy, and cloud model super entropy; based on the cloud model parameters of the last-level indicators, calculate the cloud model parameters of the upper-level indicators in the indicator hierarchy in sequence as the cloud model parameters of each of the indicators.

[0079] Specifically, the cloud model parameter E of a certain layer indicator in the network can be used to calculate the x 、E n and H e , calculate the cloud model parameter E of its parent indicator x '、E n ' and H e ', after the calculation of the cloud model parameters of the observation indicators of the last layer is completed, the cloud model parameters of the upper-level indicators can be calculated in the following way:

[0080] (6)

[0081] (7)

[0082] (8)

[0083] The weight calculation method of each superior indicator is the same as shown in formula (5), and the quantitative weight of all indicators can be obtained in the end.

[0084] In an exemplary embodiment, before the step of determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators, the following steps may be further included:

[0085] Calculation is performed based on the cloud model parameters and membership functions of the indicators to obtain the quantitative weight and qualitative importance of the indicators.

[0086] In an optional embodiment, although only quantitative weights can complete the reasoning function of the Bayesian network and provide the final risk assessment results, on-site personnel are far away from the expert scoring process and cannot directly grasp the importance of each indicator. It is difficult to estimate the sensitivity of key indicators that may cause the final risk in advance in daily work. This is a waste of the knowledge consensus condensed by experts. Therefore, this embodiment supplements the qualitative importance of the indicators. The specific calculation method is as follows:

[0087] (1) Standard cloud model generation of importance comments: linguistic variables that can be used to evaluate qualitative importance for experts And the corresponding numerical range, set the standard fuzzy cloud parameters, as shown in Table 1 below:

[0088] Table 1 Importance level range, qualitative language description and corresponding fuzzy cloud parameters

[0089]

[0090] According to the model parameters in Table 1, each qualitative variable can be processed with the help of fuzzy cloud forward process to obtain the following Figure 3 The importance scatter plot based on the importance standard cloud model in [1] can represent the fuzzy importance of the "wrongly transmitted command" indicator. The specific fuzzy cloud forward process is as follows:

[0091] Step 1: Randomly generate a n For expectation, H e 2 is a random number with normal distribution and variance E nn ;

[0092] Step 2: Randomly generate a x is the expectation, E in step 1 nn 2 is a random number x with normal variance, which completes the generation of a cloud droplet in the cloud model;

[0093] Step 3: Repeat steps 1 and 2 for a total of N times, and finally generate the corresponding cloud diagrams for each standard qualitative description. Figure 3 , where the five qualitative descriptions are highlighted.

[0094] (2) Calculation of membership: Step 1, generate N cloud droplets x according to the cloud models generated for each indicator, for example, the cloud model parameter (E x , E n , H e )=(0.89, 0.0153, 0.0018), then the cloud droplets generated according to the model parameters are as follows: Figure 3As shown in ; Step 2, use the membership formula to calculate the membership of all cloud droplets to each comment cloud in the standard cloud model of the importance comment. For example, the qualitative importance of each indicator is calculated using the given standard cloud model and membership function:

[0095] (9)

[0096] The results are averaged to obtain the average membership of the indicator for each qualitative comment, and according to the maximum membership principle, the comment corresponding to the maximum value of the five average values ​​is taken as the qualitative evaluation result of the indicator importance.

[0097] Taking "mistransmitted or missed dispatch commands" as an example, we can generate 2,000 cloud droplets based on its cloud model parameters. We can calculate the membership of each of the 2,000 cloud droplets to the five comment clouds and calculate the average. The results are as follows:

[0098]

[0099] Since the degree of affiliation for the comment "very important" is the highest, the qualitative rating of "incorrect transmission or omission of dispatching commands" is "very important". Therefore, after normalization, the weights may have smaller values. However, a small weight value does not mean that the indicator itself is unimportant to the occurrence of safety risks. In order to avoid the situation where only weights are considered without considering importance, and to better present the expert's early evaluation information to on-site safety assessors, this embodiment introduces a qualitative importance evaluation on the basis of quantitative importance weights. This makes the evaluation scale closer to the staff's cognition, easier to understand and perceive, and can effectively improve on-site safety assessors' understanding of risk indicators and promote the development of safety work.

[0100] In an exemplary embodiment, the determining of the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators may include the following steps:

[0101] According to the logical relationships between the indicators, the sub-indicators and parent indicators contained in each indicator are determined; through AND gate operations between the parent indicators corresponding to the sub-indicators of any indicator, the conditional probability of occurrence of the sub-indicators of any indicator is determined; based on the conditional probability of occurrence of the sub-indicators in each indicator, the fuzzy conditional probability of the risk assessment network is obtained.

[0102] Optionally, each sub-u of each indicator U k The calculation method of the conditional probability CPT of the indicator can be: its parent indicator u ki The AND gate operation between them means that as long as one of the parent indicators occurs, the child indicator will occur.

[0103] In specific implementation, the probability of security risk can be obtained by determining the CPT and performing Bayesian network reasoning:

[0104] (1) Based on the established risk indicator system, the logical relationship between indicators is analyzed to obtain the conditional probability (CPT) of the Bayesian network. For example, the risk indicator system obtained in advance by a certain train service section is shown in Table 2.

[0105] Table 2 Construction safety risk index system for operating lines of heavy-haul railway operation sections

[0106]

[0107] Analysis of Table 2 shows that each final-level observed indicator influences the higher-level indicator in an "OR" fashion: that is, if any observed indicator occurs, the higher-level indicator will also occur. The influence of the sub-item indicators on the comprehensive indicator, as well as the impact of the comprehensive indicator on the final risk assessment, all follow an "OR" relationship. Table 2 can be used to represent the mapping of the fault tree to the Bayesian network model. The CPT for each indicator pair is the weight of each indicator multiplied by the probability of occurrence.

[0108] (2) Based on the Bayesian network, prior probability, weight, and conditional probability, the probability of occurrence of the final risk indicator can be obtained, which can provide assistance for the safety work of the operating line construction. The specific calculation method can adopt the Bayesian reasoning calculation method.

[0109] In an exemplary embodiment, obtaining the risk occurrence probability of each indicator, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification may include the following steps:

[0110] According to the risk description information of each of the indicators, a linguistic variable of the indicator occurrence probability is obtained as the risk occurrence probability of each of the indicators; by establishing a mapping relationship between the linguistic variable and the triangular fuzzy number, the risk occurrence probability of each of the indicators is converted into the fuzzy probability; using the membership function of the triangular fuzzy number, the prior probability of the risk assessment network is obtained after defuzzification processing.

[0111] In practical applications, for the construction process of an operating line, the various safety risk indicators and their logical relationships are relatively fixed. Safety risk indicators and Bayesian network diagrams can be determined through consultation and analysis, and the weighted relationships between indicators can also be determined through the aforementioned expert scoring method. However, the prior probability of the final observation indicator in the Bayesian network, as a key indicator reflecting the specific on-site conditions and the starting point for all subsequent calculations, is significantly affected by the dynamic changes in the on-site situation.

[0112] For example, the probability of occurrence of various observation indicators in the same train section at different times may vary. The prior probabilities used in Bayesian network calculations can be provided through regular reports from field staff. To simplify the operational process and better reflect the staff's direct understanding, the field staff's reports can be presented in the form of a qualitative semantic set V. To facilitate subsequent calculations, a triangular fuzzy model can be used to map the ambiguous qualitative evaluations to probability values ​​in the (0, 1) range. This can be done as follows:

[0113] (1) Determine the standard time occurrence probability semantic value and the corresponding triangular fuzzy number, as shown in Table 3:

[0114] Table 3: Semantic values ​​of risk indicator occurrence probability and corresponding triangular fuzzy numbers

[0115]

[0116] (2) On-site staff evaluate the frequency of occurrence of the terminal observation indicators of the Bayesian network based on the semantic value set V. Fuzzy operations are performed on the evaluation results of all people to obtain the final fuzzy result, which can then be defuzzified to obtain a specific probability value.

[0117] For example, by establishing a mapping relationship between semantic values ​​and triangular fuzzy numbers through g(V), the semantic evaluation results of front-line staff can be converted into prior fuzzy probabilities:

[0118]

[0119] For another example, the membership function of triangular fuzzy numbers is as follows:

[0120] (10)

[0121] Among them, x is the probability value (the value of the exact probability), a, b and m are triangular fuzzy number parameters, a is the lower limit of the probability corresponding to a certain qualitative comment, m is the maximum possible probability value corresponding to a certain qualitative comment, and b is the upper limit of the probability value corresponding to a certain qualitative comment. g(V) calculation results Follow The fuzzy number operation rules under the provisions, and The defuzzification method follows the mean area method. The calculation rules and defuzzification operation methods between related fuzzy numbers are calculated according to the relevant rules corresponding to the membership function (10).

[0122] In one example, taking the final-level observation indicator "mistransmitted or omitted dispatch orders," three on-site security personnel rated its probability of occurrence as: common (0.3, 0.5, 0.7), rare (0.5, 0.7, 0.9), and common (0.3, 0.5, 0.7). The final fuzzy triangular fuzzy number is (0.36, 0.57, 0.77). Defuzzifying the fuzzy probability of "mistransmitted or omitted dispatch orders" using the mean area method yields a precise probability value. For example, the probability of "mistransmitted or omitted dispatch orders" is 0.66, which can be directly substituted into the Bayesian network as the prior probability for the final-level indicator.

[0123] In an exemplary embodiment, Figure 4 As shown, a flow chart of another method for heavy-haul railway construction safety risk assessment is provided. In this embodiment, the method includes the following steps:

[0124] In step 401, a risk indicator system for construction safety on heavy-haul railway operating lines is established. A risk assessment network is constructed based on this risk indicator system to describe risk propagation. In step 402, the importance score of each indicator is obtained based on the numerical range of the qualitative importance linguistic variable. The cloud model parameters for each indicator are calculated using the cloud model fuzzy algorithm and the importance score results. In step 403, the ambiguity parameter is determined based on the cloud model parameters of each indicator. If the ambiguity parameter meets the ambiguity condition, the indicator weight cloud model is obtained. In step 404, the quantitative weight and qualitative importance of each indicator are calculated based on the cloud model parameters and the membership function of each indicator. In step 405, the fuzzy conditional probability of the risk assessment network is determined by combining the quantitative weight and qualitative importance of each indicator, as well as the logical relationships between the indicators. In step 406, the indicator occurrence probability linguistic variable is obtained based on the risk description information of each indicator. This variable serves as the risk occurrence probability of each indicator. By establishing a mapping relationship between the linguistic variable and triangular fuzzy number, the risk occurrence probability of each indicator is converted into a fuzzy probability. In step 407, the membership function of the triangular fuzzy number is used to defuzzify the prior probability of the risk assessment network. In step 408, the risk probability of each indicator in the risk assessment network is calculated based on the risk indicator system, fuzzy conditional probability, and prior probability to obtain the risk assessment result.

[0125] It should be noted that the specific limitations of the above steps can be found in the above specific limitations of a heavy-haul railway construction safety risk assessment method, which will not be repeated here.

[0126] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0127] Based on the same inventive concept, embodiments of the present application also provide a heavy-haul railway construction safety risk assessment device for implementing the aforementioned heavy-haul railway construction safety risk assessment method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the heavy-haul railway construction safety risk assessment device provided below can be found in the above-mentioned limitations of the heavy-haul railway construction safety risk assessment method and will not be further elaborated here.

[0128] In an exemplary embodiment, Figure 5 As shown, a heavy-haul railway construction safety risk assessment device is provided, comprising:

[0129] A network construction module 501 is used to establish a risk indicator system for construction safety of heavy-haul railway operating lines, and to construct a risk assessment network for describing risk propagation based on the risk indicator system;

[0130] The cloud model generation module 502 is configured to determine the importance of each indicator in the risk indicator system by using a cloud model fuzzy algorithm and generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0131] A fuzzy conditional probability determination module 503 is configured to determine the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0132] A priori probability acquisition module 504 is used to obtain the risk occurrence probability of each indicator, map the risk occurrence probability into fuzzy probability based on a triangular fuzzy number model, and obtain the priori probability of the risk assessment network after defuzzification;

[0133] The risk assessment result obtaining module 505 is used to calculate the risk probability of each indicator in the risk assessment network based on the risk indicator system, the fuzzy conditional probability, and the prior probability to obtain a risk assessment result.

[0134] In one embodiment, the cloud model generation module 502 is specifically used to obtain the importance score results of each of the indicators based on the numerical range corresponding to the qualitative importance linguistic variable; use the cloud model fuzzy algorithm and the importance score results of each of the indicators to calculate the cloud model parameters of each of the indicators; determine the ambiguity parameters based on the cloud model parameters of each of the indicators, and obtain the indicator weight cloud model when the ambiguity parameters meet the ambiguity conditions.

[0135] In one embodiment, the cloud model generation module 502 is further used to determine the cloud model parameters of the last-level indicators in the indicator hierarchy; the cloud model parameters include cloud model expectation, cloud model entropy, and cloud model super entropy; based on the cloud model parameters of the last-level indicators, the cloud model parameters of the upper-level indicators in the indicator hierarchy are calculated in sequence as the cloud model parameters of each of the indicators.

[0136] In one embodiment, the apparatus further comprises:

[0137] The membership calculation module is used to calculate the quantitative weight and qualitative importance of each indicator based on the cloud model parameters and membership function of each indicator.

[0138] In one embodiment, the fuzzy conditional probability determination module 503 is specifically used to determine the sub-indicators and parent indicators contained in each of the indicators based on the logical relationships between the indicators; determine the conditional probability of occurrence of the sub-indicators of any indicator through AND gate operations between the parent indicators corresponding to the sub-indicators of any indicator; and obtain the fuzzy conditional probability of the risk assessment network based on the conditional probability of occurrence of the sub-indicators in each of the indicators.

[0139] In one embodiment, the prior probability acquisition module 504 is specifically used to obtain the indicator occurrence probability linguistic variable based on the risk description information of each indicator as the risk occurrence probability of each indicator; by establishing a mapping relationship between the linguistic variable and the triangular fuzzy number, the risk occurrence probability of each indicator is converted into the fuzzy probability; using the membership function of the triangular fuzzy number, the prior probability of the risk assessment network is obtained after defuzzification processing.

[0140] Each module in the heavy-haul railway construction safety risk assessment device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0141] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a heavy-load railway construction safety risk assessment method.

[0142] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0143] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0144] Establish a risk indicator system for construction safety of heavy-haul railway operating lines, and construct a risk assessment network based on the risk indicator system to describe risk propagation;

[0145] By using a cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined to generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0146] Determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0147] Obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification;

[0148] Based on the risk indicator system, the fuzzy conditional probability and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

[0149] In one embodiment, when the processor executes the computer program, it also implements the steps of the heavy-haul railway construction safety risk assessment method in other embodiments described above.

[0150] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0151] Establish a risk indicator system for construction safety of heavy-haul railway operating lines, and construct a risk assessment network based on the risk indicator system to describe risk propagation;

[0152] By using a cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined to generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0153] Determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0154] Obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification;

[0155] Based on the risk indicator system, the fuzzy conditional probability and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

[0156] In one embodiment, when the computer program is executed by the processor, the steps of the heavy-haul railway construction safety risk assessment method in the other embodiments described above are also implemented.

[0157] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0158] Establish a risk indicator system for construction safety of heavy-haul railway operating lines, and construct a risk assessment network based on the risk indicator system to describe risk propagation;

[0159] By using a cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined to generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing;

[0160] Determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators;

[0161] Obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification;

[0162] Based on the risk indicator system, the fuzzy conditional probability and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

[0163] In one embodiment, when the computer program is executed by the processor, the steps of the heavy-haul railway construction safety risk assessment method in the other embodiments described above are also implemented.

[0164] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0165] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0166] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A heavy-haul railway construction safety risk assessment method, characterized in that: The method comprises: Establish a risk indicator system for construction safety of heavy-haul railway operating lines, and construct a risk assessment network based on the risk indicator system to describe risk propagation; By using a cloud model fuzzy algorithm, the importance of each indicator in the risk indicator system is determined to generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing; Determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators; Obtaining the risk occurrence probability of each of the indicators, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification; Based on the risk indicator system, the fuzzy conditional probability and the prior probability, the risk probability of each indicator in the risk assessment network is calculated to obtain a risk assessment result.

2. The method according to claim 1, characterized in that The cloud model fuzzy algorithm is used to determine the importance of each indicator in the risk indicator system and generate an indicator weight cloud model, including: Obtaining the importance score results of each of the indicators according to the numerical range corresponding to the qualitative importance language variable; Calculate the cloud model parameters of each indicator by using the cloud model fuzzy algorithm and the importance score results of each indicator; The ambiguity parameter is determined according to the cloud model parameter of each of the indicators, and the indicator weight cloud model is obtained when the ambiguity parameter meets the ambiguity condition.

3. The method according to claim 2, characterized in that The cloud model parameters for calculating the indicators include: Determining cloud model parameters of the last-level indicators in the indicator hierarchy; the cloud model parameters include cloud model expectation, cloud model entropy, and cloud model hyperentropy; According to the cloud model parameters of the last-layer indicators, the cloud model parameters of the upper-level indicators in the indicator hierarchy are calculated in sequence as the cloud model parameters of each of the indicators.

4. The method according to claim 2, characterized in that Before the step of determining the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators, the method further includes: Calculation is performed based on the cloud model parameters and membership functions of the indicators to obtain the quantitative weight and qualitative importance of the indicators.

5. The method according to claim 1, wherein The step of combining the quantitative weights and qualitative importance of the indicators, as well as the logical relationships between the indicators, to determine the fuzzy conditional probability of the risk assessment network includes: Determine the sub-indicators and parent indicators included in each indicator based on the logical relationships between the indicators; Determine the conditional probability of occurrence of any indicator's sub-indicators through AND gate operations between the parent indicators corresponding to the sub-indicators of any indicator; Based on the conditional probability of occurrence of the sub-indicators in each of the indicators, the fuzzy conditional probability of the risk assessment network is obtained.

6. The method according to claim 1, characterized in that The step of obtaining the risk occurrence probability of each indicator, mapping the risk occurrence probability into a fuzzy probability based on a triangular fuzzy number model, and obtaining the prior probability of the risk assessment network after defuzzification processing includes: Obtaining an indicator occurrence probability language variable according to the risk description information of each of the indicators as the risk occurrence probability of each of the indicators; By establishing a mapping relationship between linguistic variables and triangular fuzzy numbers, the risk occurrence probability of each indicator is converted into the fuzzy probability; The membership function of triangular fuzzy numbers is used to obtain the prior probability of the risk assessment network through defuzzification.

7. A heavy-haul railway construction safety risk assessment device, characterized in that: The device comprises: A network construction module is used to establish a risk indicator system for the construction safety of heavy-haul railway operating lines, and to construct a risk assessment network for describing risk propagation based on the risk indicator system; A cloud model generation module is used to determine the importance of each indicator in the risk indicator system by using a cloud model fuzzy algorithm and generate an indicator weight cloud model; the indicator weight cloud model satisfies the ambiguity condition under natural language processing; A fuzzy conditional probability determination module is used to determine the fuzzy conditional probability of the risk assessment network by combining the quantitative weights and qualitative importances of the indicators and the logical relationships between the indicators; A priori probability acquisition module is used to obtain the risk occurrence probability of each of the indicators, map the risk occurrence probability into fuzzy probability based on a triangular fuzzy number model, and obtain the priori probability of the risk assessment network after defuzzification; The risk assessment result obtaining module is used to calculate the risk probability of each indicator in the risk assessment network based on the risk indicator system, the fuzzy conditional probability and the prior probability to obtain a risk assessment result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images