Gas cylinder safety risk prediction method and system
By combining Bayesian network model and game theory ideas, using cylinder safety risk factor scores and accident counts, weighted summing is achieved, and the accuracy of cylinder safety risk prediction is improved, and the problem of subjective dependence in the existing technology is solved. .
Patent Information
- Application Number
- CN202411071929.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-15
AI Technical Summary
The lack of data support in the prediction of gas cylinder safety risk in the prior art has led to the strong reliance on subjective consciousness and affecting the accuracy of the evaluation.
A gas cylinder safety risk prediction method is proposed. By obtaining the scores of various risk factors and the number of accidents, combining with Bayesian network model for positive reasoning, using game theory to determine the weighted coefficients, realizing the weighted sum of subjective and objective empowerment, and obtaining A prior probability, and then predict the safety risk of gas cylinders.
It improves the accuracy of gas cylinder safety risk prediction, avoids subjective dependence caused by expert experience and knowledge, and can accurately determine the risk factors that lead to gas cylinder accidents.
Smart Images

Figure CN118966776A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of gas cylinder safety risk prediction, and in particular to a gas cylinder safety risk prediction method and system. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] As a special equipment, gas cylinders have a wide range of flow and diverse use environments. Most of the filled media are flammable, explosive, toxic and corrosive. Once a gas cylinder leaks, burns or explodes, it will seriously threaten people's lives and property. Risk prediction is one of the most effective ways to prevent accidents. On the basis of identifying and evaluating the possible safety risks at the production site, understanding the safety status of the production site in real time, analyzing and predicting the possible accident risk status and risk level can provide a theoretical basis for gas cylinder safety risk management and accident prevention.
[0004] In the actual scenario of gas cylinder production and use, due to the lack of a large amount of historical data and the uncertainty of existing data, expert evaluation methods are often used to estimate the possibility of risk events. Expert evaluation method refers to the expert's assignment of values to event risk points based on his or her own domain knowledge and experience when relevant event data is insufficient. However, this method is highly dependent on subjective consciousness, and the expert's experience and opinions may be affected by personal preferences, experience limitations or emotional factors, thus affecting the accuracy of the evaluation. Summary of the invention
[0005] In order to solve the above problems, the present invention proposes a gas cylinder safety risk prediction method and system. When predicting the safety risk of gas cylinders, not only the score of each risk factor is considered, but also the objective weighting of each risk factor is considered, which improves the accuracy of gas cylinder safety risk prediction and can determine the dangerous factors that lead to gas cylinder accidents.
[0006] To achieve the above object, the present invention adopts the following technical solution:
[0007] First, a method for predicting gas cylinder safety risks is proposed, including:
[0008] Obtain the safety risk factor score of each gas cylinder, the number of safety accidents that occurred in the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that led to the number of safety accidents that occurred in the gas cylinder;
[0009] Divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents involving the gas cylinder to obtain the objective weighting of each gas cylinder safety risk factor;
[0010] Based on the scores, determine the subjective weighting of each gas cylinder safety risk factor;
[0011] The subjective weighting and the objective weighting are weighted and summed to obtain the prior probability of each gas cylinder safety risk factor;
[0012] According to the prior probability of each risk factor, the Bayesian network model constructed according to the gas cylinder safety risk index system is forward inferred to obtain the gas cylinder safety risk prediction results.
[0013] Furthermore, according to the score of each gas cylinder safety risk factor, the subjective weighting of each gas cylinder safety risk factor is determined based on triangular fuzzy numbers.
[0014] Furthermore, based on game theory, the weighting coefficients for weighted summation of subjective weighting and objective weighting are determined.
[0015] Furthermore, the gas cylinder safety risk index system includes primary indicators, secondary indicators and tertiary indicators, among which each gas cylinder safety risk factor is a tertiary indicator; the tertiary indicator is the specific content of the secondary indicator, and the secondary indicator is the specific content of the primary indicator;
[0016] The Bayesian network model constructed according to the gas cylinder safety risk index system takes each gas cylinder safety risk factor as the root node, the gas cylinder safety risk prediction result as the leaf node, the secondary index and the primary index as the intermediate node, and determines the directed arcs between the nodes according to the relationship between the indicators.
[0017] Furthermore, the prediction result of gas cylinder safety risk is set as the occurrence of a safety accident, and the Bayesian network model constructed according to the gas cylinder safety risk index system is reversely inferred to obtain the posterior probability of each gas cylinder safety risk factor;
[0018] According to the posterior probability of each gas cylinder safety risk factor, the risk factors that lead to safety accidents of gas cylinders are determined.
[0019] Furthermore, the ratio between the posterior probability and the prior probability of each gas cylinder safety risk factor is calculated, and the gas cylinder safety risk factor with the largest ratio is selected as the risk factor that causes the gas cylinder safety accident.
[0020] Secondly, a gas cylinder safety risk prediction system is proposed, including:
[0021] A data acquisition unit, used to obtain the safety risk factor score of each gas cylinder, the number of safety accidents of the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that leads to the number of safety accidents of the gas cylinder;
[0022] An objective weight determination unit is used to divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents that occurred in the gas cylinder to obtain an objective weight for each gas cylinder safety risk factor;
[0023] A subjective weight determination unit is used to determine the subjective weight of each gas cylinder safety risk factor based on the score;
[0024] A priori probability determination unit is used to perform weighted summation of subjective weighting and objective weighting to obtain the priori probability of each gas cylinder safety risk factor;
[0025] The forward reasoning unit is used to perform forward reasoning on the Bayesian network model constructed according to the gas cylinder safety risk index system according to the prior probability of each risk factor, so as to obtain the gas cylinder safety risk prediction result.
[0026] In a third aspect, a computer device is provided, the device comprising:
[0027] a processor adapted to execute a computer program;
[0028] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a gas cylinder safety risk prediction method proposed in the first aspect is implemented.
[0029] In a fourth aspect, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a gas cylinder safety risk prediction method proposed in the first aspect.
[0030] In a fifth aspect, a computer program product is proposed, which includes a computer program. When the computer program is executed by a processor, it implements a gas cylinder safety risk prediction method proposed in the first aspect.
[0031] Compared with the prior art, the present invention has the following beneficial effects:
[0032] 1. The present invention proposes a method and system for predicting the safety risk of gas cylinders. The method obtains the scores of the safety risk factors of each gas cylinder, determines the subjective weighting of each gas cylinder safety risk factor according to the scores, and also obtains the number of safety accidents of gas cylinders and the number of occurrences of each gas cylinder safety risk factor that leads to the number of safety accidents of gas cylinders. According to each number, the objective weighting of each gas cylinder safety risk factor is determined; based on game theory, the optimal weighting coefficient of subjective weighting and objective weighting is determined, and the subjective weighting and objective weighting are weighted by the weighting coefficient to obtain the prior probability of each gas cylinder safety risk factor; the prior probability is used to predict the safety risk of gas cylinders, which improves the accuracy of the safety risk prediction of gas cylinders and avoids the drawbacks of subjective dependence brought by expert experience knowledge in fuzzy theory. In addition, the method can not only accurately determine the results of gas cylinder safety risk prediction, but also determine the safety factors that lead to safety accidents of gas cylinders.
[0033] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.
[0035] Figure 1 A flowchart of a method disclosed for an embodiment;
[0036] Figure 2 The Bayesian network structure diagram constructed in the disclosed method of the embodiment. DETAILED DESCRIPTION
[0037] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0038] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.
[0039] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.
[0040] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0041] Example 1
[0042] Risk prediction is one of the most effective ways to prevent accidents. In terms of gas cylinder safety risk prediction, expert evaluation methods are often used to estimate the probability of risk events due to insufficient data caused by practical constraints or resource limitations. An expert is a person who has knowledge about the operation of a specific system and can evaluate the operating conditions of the system. Quantifying expert knowledge as a subjective probability is a method of synthesizing expert opinions when data is insufficient. Expert evaluation methods have been applied in various fields, including decision analysis, psychology, and risk analysis. Fuzzy theory can effectively deal with the uncertainty and ambiguity of data, provide flexibility and simplify complexity. It is applicable to various fields and can be used in combination with expert knowledge to improve the efficiency and accuracy of problem solving. Due to its inherent advantages, fuzzy logic has become an indispensable component of risk prediction methods.
[0043] In the field of special equipment, various neural network models are often used in elevators and cranes to mine accident data and establish analysis models and prediction models. However, in the actual use scenarios of gas cylinders, the occurrence of accidents often involves a large number of environmental and human factors. The prediction of gas cylinder accidents often requires an explanation of the prediction results to facilitate the management department to make rectifications and prevention. Although the neural network model has good learning ability and is suitable for the study of complex models, the interpretability of the results is poor. The Bayesian Network (BN) based on Bayesian theory can perform causal reasoning on node relationships based on conditional probability. The clear network structure relationship chain has strong interpretability and can handle problems such as probabilistic reasoning with complex indicator factors and uncertain data.
[0044] In recent years, the fuzzy Bayesian Network (FBN), which combines fuzzy theory with Bayesian Network, has been proven to be effective in quantitative risk prediction and reliability analysis in the petroleum and chemical industries. In view of the problem that the indicators of gas cylinder safety risk factors are complex and the data is uncertain, this embodiment uses the advantages of fuzzy theory and Bayesian theory network in processing uncertain events, uses the fuzzy Bayesian network model to model the correlation pattern between influencing factors and gas cylinder safety risks, analyzes gas cylinder accidents from qualitative and quantitative perspectives, and establishes a risk factor indicator system that affects gas cylinder safety. At the same time, in order to avoid the subjective dependence disadvantages brought by expert experience knowledge in fuzzy theory, the probability statistics method is used to objectively weight the event influencing factors, and the multi-angle weighting algorithm is balanced based on the game theory idea to obtain the optimal combination weight.
[0045] A gas cylinder safety risk prediction method disclosed in this embodiment is described in detail.
[0046] like Figure 1As shown, a method for predicting safety risks of gas cylinders disclosed in this embodiment includes:
[0047] S1: Obtain the safety risk factor score of each gas cylinder, the number of safety accidents that occurred in the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that led to the number of safety accidents that occurred in the gas cylinder.
[0048] S2: Divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents involving the gas cylinder to obtain the objective weighting of each gas cylinder safety risk factor.
[0049] S3: Determine the subjective weighting of each gas cylinder safety risk factor based on the score.
[0050] S4: Perform weighted summation of subjective weighting and objective weighting to obtain the prior probability of each gas cylinder safety risk factor.
[0051] S5: According to the prior probability of each risk factor, the Bayesian network model constructed according to the gas cylinder safety risk index system is forward inferred to obtain the gas cylinder safety risk prediction results.
[0052] Before making an actual safety risk prediction for a gas cylinder, this embodiment first sorts out the investigation reports on gas cylinder safety accidents that have occurred, determines the safety risk factors that lead to safety accidents involving gas cylinders, and constructs a gas cylinder safety risk index system based on the safety risk factors of each gas cylinder; then, a Bayesian network model is constructed based on the gas cylinder safety risk index system.
[0053] Among them, the gas cylinder safety risk index system includes first-level indicators, second-level indicators and third-level indicators, among which each gas cylinder safety risk factor is a third-level indicator; the third-level indicator is the specific content of the second-level indicator, and the second-level indicator is the specific content of the first-level indicator.
[0054] In the specific implementation, a total of 280 gas cylinder accident investigation reports were collected from the official websites of provincial and municipal governments, emergency management bureaus, and special equipment inspection agencies, covering the period from 2002 to 2023. The investigation reports were analyzed in detail from the perspectives of environment, management, objects, and people, and the safety risk factors that lead to gas cylinder accidents were sorted out as three-level indicators, resulting in a gas cylinder safety risk indicator system as shown in Table 1. Among them, the first-level indicators include personnel risk Y1, material risk Y2, environmental risk Y3, and management risk Y4. The second-level indicators corresponding to personnel risk Y1 include employee qualification Y5, habitual behavior Y6, and one-time violation Y7; the second-level indicators corresponding to material risk Y2 include unqualified gas cylinders Y8, defects Y9, and gas mixing Y10. 10 , Oil pollution Y 11 、Safety Attachment Y 12 and equipment facilities 13 ; The secondary indicators corresponding to environmental risk Y3 include weather Y 14, ignition source Y 15 and environmental facilities 16 ; The secondary indicators corresponding to management risk Y4 include safety production training and education Y 17 and system management 18 Each secondary indicator includes the corresponding safety risk factors, as shown in Table 1.
[0055] Table 1 Gas cylinder safety risk index system
[0056]
[0057]
[0058]
[0059]
[0060] The Bayesian network model constructed according to the gas cylinder safety risk index system takes each gas cylinder safety risk factor as the root node, the gas cylinder safety risk prediction result as the leaf node, the secondary index and the primary index as the intermediate node, and determines the directed arcs between the nodes according to the relationship between the indicators. This topological structure takes the prior probability of each gas cylinder safety risk factor as input, calculates the probability of each intermediate node and the probability of the leaf node through the conditional probability calculation formula, and takes the gas cylinder safety risk prediction result as output, and the gas cylinder safety risk prediction result is expressed by the probability of a gas cylinder safety accident.
[0061] Bayesian network (BN), also known as belief network or directed acyclic graph model, is used to describe probabilistic causal reasoning in scenarios with incomplete data expression, fuzzy uncertainty, etc. In the prediction of gas cylinder safety risks, the correlation between risk factors and levels can be established based on the clear network structure relationship chain of BN, and the interpretability of risk prediction can be enhanced by using probability-based causal reasoning models. Based on the risk indicator system that has been constructed, the risk indicators are converted into Bayesian network nodes, and the relationships between the indicators are converted into directed arcs between Bayesian network nodes. In the Bayesian network model, the tail of the arrow is the root node, pointing to the child node. In this embodiment, the Bayesian network model constructed according to the gas cylinder safety risk indicator system has a topological structure as follows: Figure 2 As shown, X1~X 54 As the root node, Y1~Y 18 is the intermediate node, T is the leaf node, and T is the gas cylinder safety risk prediction result.
[0062] After constructing a Bayesian network model based on the gas cylinder safety risk index system, when it was used to predict the gas cylinder safety risk, the scores of each gas cylinder safety risk factor, the number of gas cylinder safety accidents number(M), and the number of times each gas cylinder safety risk factor caused number(M) gas cylinder safety accidents occurred were obtained.
[0063] Among them, the gas cylinder safety risk factor scores are obtained by experts based on their experience and knowledge. The specific gas cylinder safety risk factor scores include very low (V), low (L), low (ML), medium (M), high (MH), high (H) and very high (VH).
[0064] This embodiment determines the subjective weighting of each gas cylinder safety risk factor based on the triangular fuzzy number according to the score of each gas cylinder safety risk factor.
[0065] As a special fuzzy set, triangular fuzzy number (TFN) is a quantitative method for fuzzy language evaluation of information describing uncertainty, ambiguity and partial truth by using the experience and knowledge of domain experts. The membership function of fuzzy number is as follows:
[0066]
[0067] Where x represents a specific element in the domain U, which refers to each security risk factor in this embodiment. For any x∈U, there is a number μ A (x)∈[0,1] corresponds to μ A (x) is called the membership degree of x to U. a and b represent the upper and lower bounds of the fuzzy number, respectively, and m represents the value with a greater probability.
[0068] The 7-level linguistic variables are used to convert the qualitative language of the expert judgment into a fuzzy interval number to determine the consistency S(A, B) between the expert opinions. The linguistic variables and the corresponding triangular fuzzy numbers are shown in Table 2.
[0069]
[0070] Among them, A(i) and A(j) are the triangular fuzzy numbers of the i-th and j-th experts respectively.
[0071] A(i)=(a i ,m i ,b i ),A(j)=(a j ,m j ,b j ),S(A(i),A(j))∈[0,1].
[0072] Table 2 Triangular fuzzy numbers corresponding to linguistic variables
[0073]
[0074] The centroid method is used to defuzzify the triangular fuzzy number to obtain the fuzzy possibility score (FPS), which is the supervisor's weight.
[0075]
[0076] Using statistical methods, the proportion of each gas cylinder safety risk factor in the accident data is calculated as the objective weight of each safety risk factor. Traverse the gas cylinder safety accident data, count the number of times each risk factor that causes number (M) gas cylinder safety accidents occurs, and calculate the objective weight of the gas cylinder safety risk factor.
[0077]
[0078] Where P(A) represents the objective weighting of the safety risk factor A of the gas cylinder, number(A) represents the number of times the safety risk factor A of the gas cylinder occurs in number(M) safety accidents of the gas cylinder, and number(M) represents the number of safety accidents of the gas cylinder.
[0079] This embodiment performs weighted summation of subjective weighting and objective weighting to obtain the prior probability W of each gas cylinder safety risk factor.
[0080] W=ε1W1+ε2W2+...+ε k W k (5)
[0081] Among them, W represents the prior probability of the risk factor, ε k Represents the kth weight W j The weighting coefficient, W j represents the kth weighting, k=2, W1 and W2 represent subjective weighting and objective weighting respectively.
[0082] In order to balance the difference between subjective weighting and objective weighting of each gas cylinder safety risk factor, thereby obtaining the prior probability of each gas cylinder safety risk factor, this embodiment determines the weighting coefficient ε when performing weighted summation of subjective weighting and objective weighting based on game theory.
[0083] Game theory is a theoretical method used to study the cooperation, conflict or decision-making scenarios of two or more parties. It seeks the optimal solution to the problem by integrating the analysis results of both parties or more parties to meet the goal of maximizing mutual benefits. Therefore, this embodiment applies the idea of game theory to weigh the difference between the subjective and objective weightings of risk factors to obtain the optimal weights that cause gas cylinder safety risks, that is, the prior probability of each gas cylinder safety risk factor. Assuming that there are k weighting methods for the indicator, there are coefficients ε that are not all zero. The prior probability W is expressed as a linear combination of the weights solved by each calculation method, as shown in formula (5). The weight coefficient optimization function is used to determine the weight coefficient ε when the subjective weighting and the objective weighting are weighted and summed; the weight coefficient optimization function is shown in formula (6).
[0084]
[0085] Among them, ε j Denotes the jth weight W j The weighting coefficient of , j ranges from 1 to k, k=2.
[0086] Combined with the properties of differential optimization, the weighted coefficient optimization function is transformed into a combination of linear equations that can be solved using the first-order derivative, as shown in formula (7). j Formula (8) is used for normalization, and formula (9) is used to calculate the final prior probability of the risk factor.
[0087]
[0088]
[0089]
[0090] Among them, W' represents the prior probability of the final risk factor, ε' j Denotes the jth weight W j The standardization coefficient of , j ranges from 1 to k, and k=2.
[0091] The prior probability of each gas cylinder safety risk factor calculated in this embodiment is the prior probability of the root node of the constructed Bayesian network model.
[0092] The conditional probability of each intermediate node in the Bayesian network model is calculated according to the conditional probability formula. For any random variable v in BN = (v1, v2, ..., v n ), the joint probability density distribution is shown in formula (10), and the conditional probability expression is shown in formula (11).
[0093]
[0094]
[0095] Among them, P(v1,v2,...,v n ) represents the joint probability density distribution, P(v i |v1,v2,...,v i-1 ) represents node v i The conditional probability, v i represents the i-th node, and n is the total number of nodes.
[0096] The Bayesian network is based on the Bayesian theory. The process of using prior probability and conditional probability to reason about the probability of the target node layer by layer is called forward reasoning. According to the forward reasoning algorithm of the Bayesian network and formula (12), the probability P(T) of the leaf node T, i.e. the gas cylinder accident, can be further calculated, thereby realizing the prediction of the gas cylinder safety risk.
[0097]
[0098] Where P a (x i ) is the node x i The parent node of P(x i |P a (x i )) represents the probability of the child node occurring under the condition of the parent node event. Bayesian network can use the logical relationship between nodes to decompose and simplify, and derive the probability of leaf nodes from the prior probability of the root node.
[0099] The present embodiment discloses a method for predicting safety risks of gas cylinders. Based on the obtained results of safety risk prediction of gas cylinders, risk factors leading to safety accidents of gas cylinders can also be determined. Specifically, the risk factors are:
[0100] The gas cylinder safety risk prediction result is set as the occurrence of a safety accident, and the gas cylinder safety prediction model is reversely inferred to obtain the posterior probability of each gas cylinder safety risk factor;
[0101] According to the posterior probability of each gas cylinder safety risk factor, the risk factors that lead to safety accidents of gas cylinders are determined.
[0102] The ratio between the posterior probability and the prior probability of each gas cylinder safety risk factor is calculated, and the gas cylinder safety risk factor with the largest ratio is selected as the risk factor that causes the gas cylinder safety accident.
[0103] The reverse diagnosis of risk nodes is to improve the efficiency of pre-emptive prevention. After setting the probability of leaf node events to 100% in the Bayesian network, the posterior probability of each root node can be calculated by formula (13). The conditional probability of the intermediate nodes can be updated by combining the conditional probability formula of the Bayesian network. The degree of change of the root node probability is analyzed. The degree of change is the ratio between the posterior probability of the root node and the prior probability. Through the degree of change of the root node probability, we can reversely identify which risk factors are the key to the occurrence of accidents and are most sensitive to the possibility of accidents. Finding the most sensitive risk events for key supervision or governance can reduce the probability of events.
[0104] P(x i |T)=P(x i ,T) / P(T) (13)
[0105] In the formula, P(x i ,T) / P(T) means the root node x when the leaf node T event occurs i The posterior probability of an event. Given a leaf node state, the Bayesian network can implement bidirectional reasoning based on the Bayesian axiom to obtain the posterior probability of the node.
[0106] The present embodiment discloses a method for predicting the safety risk of gas cylinders. Based on the gas cylinder accident report, the risk factors leading to gas cylinder safety accidents are comprehensively analyzed from the perspectives of people, objects, environment and management. The people, objects, environment and management factors are used as primary indicators, and detailed risk factors are used as secondary and tertiary indicators to establish a gas cylinder safety risk indicator system. In view of the drawbacks of subjective dependence brought about by expert experience and knowledge in fuzzy theory, in the present embodiment, the probability statistics method is used to objectively weight the event influencing factors, and the multi-angle weighting algorithm is balanced based on the game theory idea to obtain the optimal combination weight, that is, to obtain the optimal prior probability of the root node, so as to minimize the subjectivity inherent in the expert evaluation, thereby more accurately predicting the probability of the accident.
[0107] Example 2
[0108] In this embodiment, a gas cylinder safety risk prediction system is disclosed, including:
[0109] A data acquisition unit, used to obtain the safety risk factor score of each gas cylinder, the number of safety accidents of the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that leads to the number of safety accidents of the gas cylinder;
[0110] An objective weight determination unit is used to divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents that occurred in the gas cylinder to obtain an objective weight for each gas cylinder safety risk factor;
[0111] A subjective weight determination unit is used to determine the subjective weight of each gas cylinder safety risk factor based on the score;
[0112] A priori probability determination unit is used to perform weighted summation of subjective weighting and objective weighting to obtain the priori probability of each gas cylinder safety risk factor;
[0113] The forward reasoning unit is used to perform forward reasoning on the Bayesian network model constructed according to the gas cylinder safety risk index system according to the prior probability of each risk factor, so as to obtain the gas cylinder safety risk prediction result.
[0114] The present invention also discloses a computer device, which includes:
[0115] a processor adapted to execute a computer program;
[0116] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a gas cylinder safety risk prediction method disclosed in Example 1 is implemented.
[0117] The present invention also discloses a computer-readable storage medium, which stores a computer program. The computer program is suitable for being loaded by a processor and executing a gas cylinder safety risk prediction method disclosed in Example 1.
[0118] The present invention also discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a gas cylinder safety risk prediction method disclosed in Example 1.
[0119] The method disclosed in Example 1 can be directly embodied as a hardware processor, or can be executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.
[0120] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0121] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.
Claims
1. A method for predicting gas cylinder safety risks, characterized in that: include: Obtain the safety risk factor score of each gas cylinder, the number of safety accidents that occurred in the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that led to the number of safety accidents that occurred in the gas cylinder; Divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents involving the gas cylinder to obtain the objective weighting of each gas cylinder safety risk factor; Based on the scores, determine the subjective weighting of each gas cylinder safety risk factor; The subjective weighting and the objective weighting are weighted and summed to obtain the prior probability of each gas cylinder safety risk factor; According to the prior probability of each risk factor, the Bayesian network model constructed according to the gas cylinder safety risk index system is forward inferred to obtain the gas cylinder safety risk prediction results.
2. A method for predicting safety risks of gas cylinders as claimed in claim 1, characterized in that: According to the score of each gas cylinder safety risk factor, the subjective weight of each gas cylinder safety risk factor is determined based on triangular fuzzy numbers.
3. A method for predicting safety risks of gas cylinders as claimed in claim 1, characterized in that: Based on game theory, the weighting coefficients for weighted summation of subjective weighting and objective weighting are determined.
4. A method for predicting safety risks of gas cylinders as claimed in claim 1, characterized in that: The gas cylinder safety risk index system includes primary, secondary and tertiary indicators, among which each gas cylinder safety risk factor is a tertiary indicator; the tertiary indicator is the specific content of the secondary indicator, and the secondary indicator is the specific content of the primary indicator; The Bayesian network model constructed according to the gas cylinder safety risk index system takes each gas cylinder safety risk factor as the root node, the gas cylinder safety risk prediction result as the leaf node, the secondary index and the primary index as the intermediate node, and determines the directed arcs between the nodes according to the relationship between the indicators.
5. A method for predicting safety risks of gas cylinders as claimed in claim 1, characterized in that: The prediction result of gas cylinder safety risk is set as the occurrence of a safety accident, and the Bayesian network model constructed according to the gas cylinder safety risk index system is reversely inferred to obtain the posterior probability of each gas cylinder safety risk factor; According to the posterior probability of each gas cylinder safety risk factor, the risk factors that lead to safety accidents of gas cylinders are determined.
6. A method for predicting safety risks of gas cylinders as claimed in claim 5, characterized in that: The ratio between the posterior probability and the prior probability of each gas cylinder safety risk factor is calculated, and the gas cylinder safety risk factor with the largest ratio is selected as the risk factor that causes the gas cylinder safety accident.
7. A gas cylinder safety risk prediction system, characterized in that: include: A data acquisition unit, used to obtain the safety risk factor score of each gas cylinder, the number of safety accidents of the gas cylinder, and the number of occurrences of each gas cylinder safety risk factor that leads to the number of safety accidents of the gas cylinder; An objective weight determination unit is used to divide the number of occurrences of each gas cylinder safety risk factor by the number of safety accidents that occurred in the gas cylinder to obtain an objective weight for each gas cylinder safety risk factor; A subjective weight determination unit is used to determine the subjective weight of each gas cylinder safety risk factor based on the score; A priori probability determination unit is used to perform weighted summation of subjective weighting and objective weighting to obtain the priori probability of each gas cylinder safety risk factor; The forward reasoning unit is used to perform forward reasoning on the Bayesian network model constructed according to the gas cylinder safety risk index system according to the prior probability of each risk factor, so as to obtain the gas cylinder safety risk prediction result.
8. An electronic device, characterized in that: The device comprises: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a gas cylinder safety risk prediction method according to any one of claims 1 to 6 is implemented.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a gas cylinder safety risk prediction method according to any one of claims 1 to 6.
10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements a gas cylinder safety risk prediction method as described in any one of claims 1-6.
Citation Information
Cited By
Water-power engineering anti-terrorism risk analysis method and system based on Bayesian network
CN122414842A
