A dynamic assessment method and device for multi-process flow interaction risks in gas transmission stations

Abstract

The present invention provides a dynamic risk assessment method and device for multi-process flow interaction in a gas transmission station yard. The method includes: analyzing risk factors and their mutual relationships in the operation process of the gas transmission station yard by using the multi-process flow interaction theory; screening out root nodes, intermediate nodes, and target nodes according to the selected risk factors and their mutual relationships, and establishing a multi-layer Bayesian network; determining the prior probability and conditional probability of nodes in the multi-layer Bayesian network by using the fuzzy cloud model, and establishing a risk assessment model; instantaneously inputting node evidence, determining the risk evolution path, and taking avoidance measures in a timely manner. The embodiments of the present application can solve the problems in the risk assessment of gas transmission station yards that it is difficult to describe the uncertainty, hierarchy, and interaction between risk factors.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of risk assessment for gas transmission stations, and particularly to a dynamic assessment method and device for multi-process flow interaction risks in gas transmission stations. Background Art

[0002] Gas transmission stations mainly play a role of control and regulation in the long-distance natural gas transmission process, and their stability is the guarantee for the safe operation of the entire gas transmission pipeline network.

[0003] Currently, traditional research in related fields mainly focuses on the analysis and identification of single or limited risk factors in gas transmission stations. However, in this complex system of gas transmission stations, the relationships between various risk factors have characteristics such as uncertainty, hierarchy, and interactivity. In terms of uncertainty reasoning, Bayesian networks have significant advantages compared to other risk assessment methods. But in the existing research applications, traditional Bayesian networks not only have strong subjectivity in determining probability parameters, but also for complex systems, it is difficult to clarify the interaction relationships between nodes, and the hierarchical relationships between each node are not clear. Therefore, the existing technology cannot accurately and reliably conduct risk assessment on gas transmission stations. Summary of the Invention

[0004] The purpose of the present invention is to provide a dynamic assessment method and device for multi-process flow interaction risks in gas transmission stations to solve the problems of uncertainty, hierarchy, and interactivity in describing the relationships between risk factors in gas transmission station risk assessment.

[0005] To solve the above technical problems, the present invention provides a dynamic assessment method for multi-process flow interaction risks in gas transmission stations, which is characterized by including the following steps:

[0006] Analyze the risk factors and their relationships in the operation process of gas transmission stations using the multi-process flow interaction theory;

[0007] Select root nodes, intermediate nodes, and target nodes based on the selected risk factors, and establish a multi-layer Bayesian network;

[0008] Use the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establish a risk assessment model;

[0009] Instantly input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner.

[0010] Furthermore, the multi-process flow interaction theory includes:

[0011] Establish the multi-process flow interaction theory;

[0012] Study the risk factors in the operation process from the perspective of "flow";

[0013] Establish IF-THEN inference rules through expert experience and accident cases to determine the mutual relationships between risk factors.

[0014] Further, screen out root nodes, intermediate nodes, and target nodes based on the selected risk factors and their mutual relationships, and establish a multi-level Bayesian network, including: determining root nodes, intermediate nodes, target nodes, and the mutual relationships between each node, that is, the directed edges in the multi-level Bayesian network.

[0015] Further, use the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-level Bayesian network, and establish a risk assessment model, including:

[0016] Classify the likelihood of risk occurrence to form standard clouds for each level;

[0017] Experts evaluate the probability parameters of each node to form evaluation clouds for each node;

[0018] Calculate the closeness between the evaluation cloud and the standard cloud to determine the risk level;

[0019] Determine the node probability parameters.

[0020] Further, immediately input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner, including: inputting the prior probability and conditional probability of each node into the multi-level Bayesian network risk assessment model established in the third step, the likelihood of a specified risk accident occurring can be obtained. When a fault occurs during the operation of the gas transmission station, update the status of each node in a timely manner, and the likelihood of the risk of a specified event occurring can be obtained for dynamic assessment of the gas transmission station risk.

[0021] Further, the classification of the likelihood of risk occurrence to form standard clouds for each level includes:

[0022] Quantify the risk likelihood threshold by grading;

[0023] Through the bilateral constraint condition formula Calculate the parameters of the standard cloud model for each level interval;

[0024] Where: x max is the likelihood threshold, and k is a constant, which can be taken according to the fuzzy threshold of the variable. Usually, the values are 0.01, 0.5, 1. Here, for more intuitiveness and reducing errors, the preferred value is 0.01.

[0025] The experts evaluate the probability parameters of each node to form evaluation clouds for each node, including:

[0026] Experts score each node in the network one by one according to the determined grade intervals, and calculate the evaluation cloud model parameters according to the following formula.

[0027]

[0028] Where: X i (i = 1, 2, …, n) is the score value of the evaluation object; is the average value of the experts' scores.

[0029] The calculation of the closeness between the evaluation cloud and the standard cloud to determine the risk level includes:

[0030] Through the formula Calculate the closeness between the standard cloud and the evaluation cloud to determine the risk level of this node.

[0031] The determination of the node probability parameter includes:

[0032] Use the formula After standardizing the closeness, use the standard closeness as the weight of each risk level, and use the formula Calculate the node risk probability.

[0033] A gas transmission station multi-process flow interaction risk dynamic assessment device for implementing the above-mentioned gas transmission station multi-process flow interaction risk dynamic assessment method, including a memory, a processor, and a computer program stored on the memory, characterized in that it further includes:

[0034] A risk analysis module for analyzing risk factors and their interrelationships in the operation process of a gas transmission station using multi-process flow interaction theory;

[0035] A Bayesian network construction module for screening root nodes, intermediate nodes, and target nodes according to the selected risk factors and establishing a multi-layer Bayesian network;

[0036] A risk assessment model construction module for using the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network and establish a risk assessment model;

[0037] A dynamic assessment module for instantaneously inputting node evidence, determining the risk evolution path, and taking avoidance measures in a timely manner.

[0038] Furthermore, when the computer program is run by the processor, the following steps are executed:

[0039] Analyze risk factors and their interrelationships in the operation process of a gas transmission station using multi-process flow interaction theory;

[0040] Select root nodes, intermediate nodes, and target nodes according to the selected risk factors, and establish a multi-layer Bayesian network;

[0041] Use the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establish a risk assessment model;

[0042] Instantly input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner.

[0043] Furthermore, the computer program is stored on a computer storage medium, and when the computer program is executed by a processor on the computer storage medium, the following steps are implemented:

[0044] Use the multi-process flow interaction theory to analyze the risk factors and their interrelationships in the operation process of the gas transmission station yard;

[0045] Select root nodes, intermediate nodes, and target nodes according to the selected risk factors, and establish a multi-layer Bayesian network;

[0046] Use the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establish a risk assessment model;

[0047] Instantly input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner.

[0048] Compared with the prior art, the advantages of the present invention are as follows:

[0049] Using the multi-process flow in the present invention can solve the hierarchical problem, and divide the risk factors of the gas transmission station into three levels: behavior, information, and material; using the Bayesian network in the patent can well show the interactivity between risk factors; using the Mohu cloud model in the patent can solve the uncertainty problem. However, the prior art cannot distinguish the hierarchy, interactivity, and uncertainty between risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 It is a flowchart of a multi-process flow interaction risk dynamic assessment method for a gas transmission station yard according to an embodiment of the present specification;

[0052] Figure 2 It is a schematic diagram of the multi-process flow interaction theory of a multi-process flow interaction risk dynamic assessment method for a gas transmission station yard according to an embodiment of the present specification;

[0053] Figure 3 This is the multi - layer Bayesian network diagram for the dynamic risk assessment method of multi - process flow interaction in a gas transmission station yard according to an embodiment of this specification;

[0054] Figure 4 This is the multi - layer Bayesian network diagram taking the gas opening process as an example for the dynamic risk assessment method of multi - process flow interaction in a gas transmission station yard according to an embodiment of this specification;

[0055] Figure 5 This is the multi - layer Bayesian network inference diagram in the dynamic risk assessment method of multi - process flow interaction in a gas transmission station yard according to an embodiment of this specification;

[0056] Figure 6 This is the structural block diagram of a dynamic risk assessment device for multi - process flow interaction in a gas transmission station yard according to an embodiment of this specification. Detailed implementation manners

[0057] The accompanying drawings are only for illustrative purposes; for better illustrating this embodiment, some components in the drawings will be omitted, enlarged or reduced, which do not represent the dimensions of the actual product; for those skilled in the art, some well - known structures and their descriptions in the drawings may be omitted.

[0058] Next, the technical solutions in the embodiments of this specification will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this specification. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0059] In the process of implementing this application, the inventors of this application found that: when traditional risk assessment models evaluate complex systems such as gas transmission station yards, they ignore the influence of the uncertainty, interactivity, and hierarchy of the relationships between multiple risk factors on the occurrence of accidents, that is, they do not consider that the occurrence of a single risk factor is random and there are non - negligible causal relationships between multiple risk factors, which is likely to affect the accuracy of risk assessment. Moreover, the probability parameters of traditional risk assessment models are mostly subjectively determined by experts, making many data deviate from the actual situation to a certain extent. After further research, it was found that: the multi - layer Bayesian network can be combined with the risk assessment of gas transmission station yards. By using the multi - process flow interaction theory to identify the three - layer risk factors of information flow - behavior flow - material flow in the gas transmission station yard, a multi - layer Bayesian network can be established. By establishing the mutual relationships between nodes, the interactivity of risk factors can be characterized, and by assigning probability parameters to nodes, the uncertainty of risk factors can be characterized, making the risk assessment more real and reliable. And by using the fuzzy cloud model theory to fuzzify the probability parameters of nodes in the Bayesian network, the influence of human subjective factors can be weakened, making the risk assessment results more in line with the actual situation.

[0060] As Figure 1 shown, a dynamic risk assessment method for multi-process flow interaction in a gas transmission station yard of the present application, the execution subject of the method is a server. The specific implementation steps of the risk assessment method for the gas transmission station are as follows:

[0061] S101: Analyze the risk factors and their interrelationships in the operation process of the gas transmission station yard by using the multi-process flow interaction theory.

[0062] Establish the multi-process flow interaction theory;

[0063] As Figure 2 shown, regard the gas transmission station as a complex system, study the flow and transmission of information between personnel and substances from the perspective of "flow", abstract the safe production path as the interactive evolution of three process flows: information flow, behavior flow, and material flow, and establish a multi-process flow interaction theory including information flow - behavior flow - material flow.

[0064] As the commander, information is transmitted between personnel or flows between substances in the forms of humidity, temperature, flow rate, rotation speed, pressure, and physical and chemical characteristic parameters of the substance itself, controlling the evolution of the entire safety path; as the controller, behavior is the embodiment of the subjective initiative of personnel, and controls substances under the guidance of information; as the carrier, the substance is the carrier of information and behavior, making the entire safety path manifest.

[0065] Study the risk factors in the operation process from the perspective of "flow";

[0066] According to the multi-process flow interaction theory, break down the risk factors in the specific operation process into information factors, material factors, and behavior factors.

[0067] Establish IF-THEN inference rules through expert experience and accident cases to determine the interrelationships between risk factors.

[0068] S102: Select root nodes, intermediate nodes, and target nodes according to the selected risk factors, and establish a multi-layer Bayesian network.

[0069] The Bayesian network is essentially a probabilistic graphical model. It uses the Bayesian algorithm with relaxed conditional independence assumptions, represents the causal relationships between random variables through a directed acyclic graph, and organically combines the directed acyclic graph with probability theory, uses a conditional probability distribution table for parameter quantification, and can effectively combine prior knowledge and the current situation of the system for analysis and evaluation, that is, it adopts a method combining qualitative representation and quantitative evaluation.

[0070] The network structure of the Bayesian network is a directed acyclic graph, denoted as <X, A>, where X is the set of nodes in the network and A is the set of directed edges in the network. Nodes represent features or random variables (observable variables, or hidden variables, unknown parameters, etc.), and directed edges represent the causal relationships between variables.

[0071] As Figure 3 shown, the nodes in the multi-layer Bayesian network are the risk factors identified through the multi-process flow interaction theory, and the information flow, behavior flow, and material flow are respectively used as the three levels in the multi-layer Bayesian network through the multi-process flow interaction theory. To facilitate better quantification of the multi-layer Bayesian network in the next step, considering the actual situation at the accident scene, all node states are set to be two-state. True indicates that the risk factor corresponding to this node occurs, and False indicates that the risk factor corresponding to this node does not occur. The mutual relationships between each node are determined by the IF-THEN inference rules summarized from expert experience and accident cases. Using the obtained nodes and the mutual relationships between nodes, a multi-layer Bayesian network is established for the next step of analysis.

[0072] S103: Use the fuzzy cloud model to determine the prior probability and conditional probability of the nodes in the multi-layer Bayesian network, and establish a risk assessment model.

[0073] The probability parameters corresponding to each node in the network include the prior probability and the conditional probability. For a node without any parent nodes, its probability parameter is called the prior probability, which is mostly obtained by expert assignment or accident statistics. For a node with one or more parent nodes, its probability parameter is called the conditional probability. The conditional probability distribution table of any node relative to its parent node set is a quantitative representation of the probability dependence degree between the parent and child nodes, and can be obtained through prior probability inference. However, it is too difficult, cumbersome, and the accuracy cannot be guaranteed to conduct quantitative analysis one by one, so expert assignment is carried out uniformly.

[0074] The fuzzy cloud model can fully reflect expert wisdom, and three cloud model parameters (E x , E n , and H e ) are defined. E x is used to describe the qualitative concept and represents the expected value of the risk probability, usually equal to the average of the interval numbers. E n is called entropy, which describes the fuzzy degree of the qualitative concept and determines the span of the cloud curve. H e can be regarded as the entropy of entropy, representing the dispersion of cloud droplets. The possibility of risk occurrence is output in the form of a cloud distribution, realizing the transformation from risk qualitative to risk quantification, and having the characteristic of visualization.

[0075] Since the determination of probability parameters in the expert assignment process is too subjective, the fuzzy cloud model is used to fuzzify and randomize the expert assignment in order to improve the accuracy of expert judgment on concepts.

[0076] 1. Classify the possibility of risk occurrence and form standard clouds for each level.

[0077] In risk assessment, three values ​​are used (E x , E n , andH e ) is used to quantify the descriptive risk probability level represented by the interval threshold, and this model is called the standard cloud model.

[0078] In order to facilitate the quantification of the possibility of risk occurrence, the risk possibility threshold is graded, and the standard cloud model parameters of each grade interval are calculated through the bilateral constraint formula (1) to generate the standard cloud.

[0079]

[0080] Where: x max is the possibility threshold, k is a constant and can be set according to the fuzzy threshold of the variable. It is usually 0.01, 0.5, or 1. Here, 0.01 is taken for more intuitiveness and to reduce errors.

[0081] 2. Experts evaluate the probability parameters of each node to form an evaluation cloud for each node;

[0082] During risk assessment, experts use the risk probability interval threshold (x min ,x max ) Score the prior probability and conditional probability of each node. Assuming there are n experts, an evaluation cloud is formed based on the fuzzy cloud theory. The evaluation cloud model parameters are calculated using formula (2).

[0083]

[0084] Where: X i (i=1,2,…,n) is the score value of the evaluation object. is the average of the experts' scores.

[0085] 3. Calculate the progress of the evaluation cloud and the standard cloud to determine the risk level.

[0086] After risk assessment of the node, a standard cloud is generated. The risk level of the node is determined by calculating the proximity between the standard cloud and the evaluation cloud using formula (3).

[0087]

[0088] 4. Determine node probability parameters and establish a risk assessment model

[0089] For the convenience of quantitative reasoning of the Bayesian network, the risk levels of the nodes determined in the previous step are converted into node risk probabilities. After calculating the membership degrees of any node belonging to each risk level, the membership degrees are standardized using formula (4), and the standardized membership degrees are used as the weights of each risk level to which the node belongs. The node risk probability is calculated using formula (5). The prior probabilities and conditional probabilities of each node are input into the multi-layer Bayesian network established in S102 to obtain a risk assessment model.

[0090]

[0091] S104: Instantly input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner.

[0092] When a failure occurs during the operation of the gas transmission station, the status of each node is updated in a timely manner. Inputting the node evidence into the risk assessment model established in S103 can obtain the risk occurrence probability of a specified event, and dynamically evaluate the risk of the gas transmission station.

[0093] In the above embodiments of the present application, the gas transmission station yard is regarded as a complex system, and the risk factors of the gas transmission station yard are divided from the perspective of multi-process flow interaction. The risk factors belonging to different process flows, such as information flow (pressure information, etc.), behavior flow (operation behaviors of staff, etc.), and material flow (equipment, natural gas, etc.), are nested within one operation process. A multi-layer Bayesian network is established, and the combination of fuzzy cloud model and expert assignment evaluation is used to realize risk quantification, weakening the influence of expert subjectivity on the evaluation results. Instantly input node evidence under different conditions to determine the probability of an event occurring under the updated evidence state, so as to conduct a reliable risk assessment of the gas transmission station yard to facilitate taking measures to prevent accidents.

[0094] Although the process flow described above includes multiple operations that appear in a specific order, it should be clearly understood that these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel (for example, using a parallel processor or a multi-threaded environment).

[0095] To facilitate a clearer understanding of the multi-process flow interaction risk dynamic assessment method of the gas transmission station yard of the present application, the above multi-process flow interaction risk dynamic assessment system of the gas transmission station yard will be described below in conjunction with an exemplary embodiment. However, it should be noted that this specific embodiment is only for better illustrating the present application and does not constitute an improper limitation of the present application.

[0096] In this exemplary embodiment, the gas opening operation process in the gas transmission station yard is taken as the research object for analysis.

[0097] I. Identifying Risk Factors

[0098] The gas opening operation of a gas transmission station is the first step after the gas transmission station is closed or renovated and resumes operation. The safety of the startup process depends on the coupling effect of many factors, such as the interaction between personnel operation, equipment status, command transmission, gas pressure and other factors. According to the multi-process flow interaction theory, the risk factors and their causal relationships for each step in the gas opening process are determined, as shown in Table 1, where IF indicates that the risk factor belongs to the information flow, BF indicates that the risk factor belongs to the behavior flow, and MF indicates that the risk factor belongs to the material flow.

[0099] Table 1 Risk Identification in the Gas Opening Process of Gas Transmission Stations

[0100]

[0101]

[0102] II. Establishing a Multilayer Bayesian Network

[0103] In a multilayer Bayesian network, nodes represent features or random variables (observable variables, or hidden variables, unknown parameters, etc.), and directed edges represent the causal relationships between variables.

[0104] As can be seen from Table 1, there are complex interactions between the risk triggering factors and risk factors of different process flows. Specifically, risks can spread within the same flow or across different flows. Taking the first case as an example, if the wires are aging (MF), there may be an ignition source on site (MF). Another example is that improper valve operation (BF) may cause personal injury (BF). On the other hand, risks spread across different flows. For example, poor communication with the central control room (IF) may lead to control system failures (MF) and human error operations (BF), and aging of wires (MF) may lead to electric shock (BF), information transmission interruption (IF), and the presence of an ignition source on site (IF).

[0105] As Figure 4 shown, in the multilayer Bayesian network of the multi-process flow interaction risk dynamic assessment method, the nodes are the risk factors identified according to the multi-process flow interaction. Each node has been given in Table 2. In order to establish the interaction relationship between nodes, based on expert experience and accident cases, the IF-THEN inference rules between risk nodes as shown in Table 3 are determined. After establishing the nodes and their relationships, a multilayer Bayesian network is obtained.

[0106] Table 2 Risk Nodes

[0107]

[0108]

[0109] IF-THEN Inference Rules between Risk Nodes:

[0110] (1) IF the inlet pressure gauge fails, THEN no display or incorrect information is shown on the pressure gauge, THEN the pressure of pipelines and equipment continuously rises;

[0111] (2) IF there is no communication with the central control room or the communication is insufficient, THEN control system failure OR human error;

[0112] (3) IF the walkie-talkie fails, THEN communication is not timely, THEN human error;

[0113] (4) IF there is no alarm when natural gas leaks, AND there is an ignition source on-site, AND natural gas leaks, THEN fire and explosion;

[0114] (5) IF the valve is opened too quickly OR the operation of opening the balance valve is skipped, THEN the valve is damaged;

[0115] (6) IF employees do not wear anti-static work clothes OR static electricity is not released, THEN there is an ignition source on-site;

[0116] (7) IF the combustible gas alarm fails, THEN no display or incorrect information is shown on the pressure gauge;

[0117] (8) IF the valve opening sequence is incorrect, THEN the instrument is damaged by gas;

[0118] (9) IF electric shock OR improper valve operation OR wire aging, THEN personal injury;

[0119] (10) IF wire aging, THEN electric shock OR information transmission interruption OR there is an ignition source on-site;

[0120] (11) IF improper valve operation, THEN the valve is locked OR the handwheel flies out;

[0121] (12) IF the integrity of equipment and facilities is insufficient, OR flange connection leaks, OR the valve is locked, OR the handwheel flies out, OR high-pressure gas damages the equipment, OR the valve is damaged, OR the pressure of pipelines and equipment continuously rises, OR control system failure, OR human error, OR the exhaust valve is not closed, THEN natural gas leaks.

[0122] III. Establish a risk assessment model.

[0123] To determine the parameters in the multi-layer Bayesian network, first divide the risk possibility into five levels and define the interval thresholds. Then calculate the standard cloud model parameters (Ex, En, He) according to the standard cloud equation, as shown in Table 3.

[0124] Table 3 Standard Cloud Model Parameters

[0125]

[0126] Then, experts score the possibility of each node event occurring, calculate the fuzzy cloud model parameters using the formula, and then generate the evaluation cloud. Taking the leakage at the flange connection in on-site production (M 6 = True) and no alarm when natural gas leaks, there is an ignition source on-site, and in the case of natural gas leakage, a fire and explosion occur (I 4 = True, M 7 = True, M 8 = True → M 15 = True) and no alarm when natural gas leaks, there is an ignition source on-site, but there is no natural gas leakage, and a fire and explosion occur (I 4 = True, M 7 = True, M 8 = False → M 15 = True) as three accident scenarios as examples, calculate the prior probability of M 6 = True and the conditional probabilities of local networks such as I 4 = True, M 7 = True, M 8 = True → M 15 = True, I 4 = True, M 7 = True, M 8 = False → M 15 = True respectively. Table 4 shows the expert scores and fuzzy evaluation cloud model parameters.

[0127] Table 4 Evaluation Cloud Model Parameters

[0128]

[0129]

[0130] It can be intuitively seen that the three evaluation clouds are not exactly the same as the standard cloud. Therefore, calculate the similarity between the evaluation cloud and the standard cloud according to formulas (3)-(4).

[0131] Finally, standardize the similarity between the evaluation cloud and each standard cloud as its weight belonging to each risk level, and calculate the risk probability using formula (5), as shown in Table 5.

[0132] Table 5 Affinity Degree and Risk Probability of Evaluation Cloud

[0133]

[0134] Thus, the leakage at the flange connection in on-site production (M 6The prior probability P = (M = True) = 0.224; and for node M 6 = True) = 0.224; and for node M 15 The local network conditional probabilities are shown in Table 6.

[0135] Table 6 for node M 15 Local network conditional probability table

[0136]

[0137] IV. Dynamic risk assessment

[0138] Using the aforementioned method, the prior probabilities of each node and the conditional probabilities in each local network are calculated and then input into the multi-layer Bayesian network risk assessment model established in the third step, and the possibility of a specified risk accident occurring can be obtained. Taking the gas opening process of a gas transmission station as an example, an inference graph as shown in Figure 5 is obtained. From the graph, the risk value of the final fire and explosion accident occurring is 0.36, which belongs to level II in the risk classification in the previous text, and the verbal description is "unlikely", indicating that in actual on-site situations, if safety management is in place and measures are appropriate, the probability of abnormal failures is very small.

[0139] Of course, the external environmental conditions on-site are constantly changing, and whether it is information, substances, or behavioral factors, they are all in dynamic motion. The occurrence of a failure in any one factor within a certain domain may affect the normal operation of the entire gas transmission station. Therefore, it is particularly important to update the node evidence in a timely manner for the risk assessment of the gas transmission station yard.

[0140] Taking the gas opening process of a gas transmission station yard as an example, when performing step (2), if the integrity of the equipment and instruments is poor (M 5 = True) and the exhaust valve is not closed (B 2 = True), then the risk value P(M8 = True) of gas leakage (M8 = True) will increase from 0.30 to 0.42. At this time, the risk level develops from "unlikely" to "possible", which can effectively guide managers and operators to take timely preventive measures.

[0141] By taking the gas opening process of a gas transmission station yard as an example for analysis and verification, the analysis results are consistent with the on-site actual situation, which proves the practicability and effectiveness of a multi-process flow interaction risk dynamic assessment method for gas transmission station yards proposed in this paper. Corresponding measures can be taken according to the risk assessment results to ensure safe production.

[0142] As shown in Figure 6 a multi-process flow interaction risk dynamic assessment device for a gas transmission station yard according to an embodiment of the present application may include:

[0143] A risk analysis module, configured to analyze risk factors and their mutual relationships in the operation process of a gas transmission station yard by using the multi-process flow interaction theory;

[0144] A Bayesian network construction module, configured to screen out root nodes, intermediate nodes, and target nodes according to the selected risk factors, and establish a multi-layer Bayesian network;

[0145] A risk assessment model construction module, configured to use a fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establish a risk assessment model;

[0146] A dynamic assessment module, configured to instantaneously input node evidence, determine the risk evolution path, and take avoidance measures in a timely manner.

[0147] The device in the embodiment of the present application corresponds to the method in the above embodiment. Therefore, for details of the device of the present application, please refer to the method in the above embodiment, which will not be elaborated herein.

[0148] For the convenience of description, when describing the above device, it is described by dividing it into various units according to functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0149] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0150] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0151] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1Steps of the functions specified in one or more boxes.

[0152] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0153] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0154] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage, or other magnetic storage devices, or any other non-transitory media that can store information accessible by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0155] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0156] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0157] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

[0158] Each embodiment in this specification is described in a progressive manner, and for the identical or similar parts among the embodiments, reference may be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and for the relevant parts, reference may be made to the partial description of method embodiments.

[0159] Certainly, the above description is not a limitation on the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions, or substitutions made by those skilled in the art within the scope of the essence of the present invention should also fall within the protection scope of the present invention.

Claims

1. A dynamic assessment method for multi-process flow interaction risks in gas transmission stations, characterized in that, it includes the following steps: Using the multi-process flow interaction theory to analyze the risk factors and their interrelationships in the operation process of gas transmission stations; Selecting root nodes, intermediate nodes, and target nodes based on the selected risk factors, and establishing a multi-layer Bayesian network; Using the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establishing a risk assessment model; Instantly inputting node evidence, determining the risk evolution path, and taking avoidance measures in a timely manner; Among them, the multi-process flow interaction theory includes: Establishing the multi-process flow interaction theory; Studying risk factors in the operation process from the perspective of "flow"; Establishing IF-THEN inference rules through expert experience and accident cases to determine the interrelationships between risk factors.

2. The dynamic assessment method for multi-process flow interaction risks in gas transmission stations according to claim 1, characterized in that, The step of selecting root nodes, intermediate nodes, and target nodes based on the selected risk factors and their interrelationships, and establishing a multi-level Bayesian network includes: establishing the root nodes, intermediate nodes, target nodes, and the interrelationships between each node, that is, the directed edges in the multi-layer Bayesian network.

3. The dynamic assessment method for multi-process flow interaction risks in gas transmission stations according to claim 1, characterized in that, The step of using the fuzzy cloud model to determine the prior probability and conditional probability of nodes in the multi-layer Bayesian network, and establishing a risk assessment model includes: Classifying the possibility of risk occurrence to form standard clouds for each level; Experts evaluate the probability parameters of each node to form evaluation clouds for each node; Calculating the closeness between the evaluation cloud and the standard cloud to determine the risk level; Determining the node probability parameters.

4. The dynamic assessment method for multi-process flow interaction risks in gas transmission stations according to claim 1, characterized in that, The step of instantly inputting node evidence, determining the risk evolution path, and taking avoidance measures in a timely manner includes: inputting the prior probability and conditional probability of each node into the multi-layer Bayesian network risk assessment model established in the third step, the possibility of a specified risk accident occurring can be obtained. When a fault occurs during the operation of the gas transmission station, the status of each node is updated in a timely manner, and the risk occurrence possibility of the specified event can be obtained, and the risks of the gas transmission station are dynamically evaluated.

5. The dynamic assessment method for multi-process flow interaction risks in gas transmission stations according to claim 3, characterized in that, The step of classifying the possibility of risk occurrence to form standard clouds for each level includes: Quantifying the risk possibility threshold by grading; Through the bilateral constraint condition formula Calculate the parameters of the standard cloud model for each grade interval where: x max is the possibility threshold, and k is a constant that can be assigned according to the fuzzy threshold of the variable, and usually can be assigned values of 0.01, 0.5, 1; The step of experts evaluating the probability parameters of each node to form evaluation clouds for each node includes: Experts score each node in the network one by one according to the determined grade interval, and calculate the evaluation cloud model parameters according to the following formula; Where: X i (i = 1, 2, …, n) is the score value of the evaluation object; is the average value of the expert scores; The step of calculating the closeness between the evaluation cloud and the standard cloud to determine the risk level includes: Determine the risk level of this node by calculating the fitting progress between the standard cloud and the evaluation cloud through the formula ​ The step of determining the node probability parameters includes: Using the formula After standardizing the paste progress and using the standard paste progress as the weight of each risk level, use the formula to calculate the node risk probability.

6. The dynamic assessment method for multi-process flow interaction risks in gas transmission stations according to claim 5, characterized in that, Bilateral constraint condition formula In this formula, k is a constant with a value of 0.

01.

7. A dynamic risk assessment device for multi-process flow interaction in a gas transmission station yard, which is used to implement the dynamic risk assessment method for multi-process flow interaction in a gas transmission station yard according to any one of claims 1 to 6, including a memory, a processor, and a computer program stored on the memory. Characterized in that, It further includes: A risk analysis module, which is used to analyze the risk factors and their mutual relationships in the operation process of the gas transmission station yard by using the multi-process flow interaction theory; A Bayesian network construction module, which is used to screen out root nodes, intermediate nodes, and target nodes according to the selected risk factors and establish a multi-layer Bayesian network; A risk assessment model construction module, which is used to determine the prior probability and conditional probability of the nodes in the multi-layer Bayesian network by using the fuzzy cloud model and establish a risk assessment model; A dynamic assessment module, which is used to input node evidence immediately, determine the risk evolution path, and take avoidance measures in a timely manner; Among them, when the computer program is run by the processor, the following steps are executed: Analyze the risk factors and their mutual relationships in the operation process of the gas transmission station yard by using the multi-process flow interaction theory; Screen out root nodes, intermediate nodes, and target nodes according to the selected risk factors and establish a multi-layer Bayesian network; Determine the prior probability and conditional probability of the nodes in the multi-layer Bayesian network by using the fuzzy cloud model and establish a risk assessment model; Input node evidence immediately, determine the risk evolution path, and take avoidance measures in a timely manner.

8. The dynamic risk assessment device for multi-process flow interaction in a gas transmission station yard according to claim 7, Characterized in that, The computer program is stored on a computer storage medium, and when the computer program is executed by the processor on the computer storage medium, the following steps are realized: Analyze the risk factors and their mutual relationships in the operation process of the gas transmission station yard by using the multi-process flow interaction theory; Screen out root nodes, intermediate nodes, and target nodes according to the selected risk factors and establish a multi-layer Bayesian network; Determine the prior probability and conditional probability of the nodes in the multi-layer Bayesian network by using the fuzzy cloud model and establish a risk assessment model; Input node evidence immediately, determine the risk evolution path, and take avoidance measures in a timely manner.

Images