Heavy haul railway contact network risk assessment research based on causal Bayesian network

Abstract

The invention belongs to the technical field of overhead line system risk assessment, and particularly relates to a heavy haul railway overhead line system risk assessment method based on a Bayesian network, a medium and equipment, and the method comprises the steps: building a causal Bayesian network assessment model of the operation risk of the heavy haul railway overhead line system based on the operation historical data of the heavy haul railway overhead line system; deducing the conditional probability of each node in the causal Bayesian network evaluation model according to the statistical analysis of the historical operation data of the heavy haul railway overhead line system; and assuming that the heavy haul railway overhead line system has a system fault, reversely deducing the posterior probability of each node in the causal Bayesian network evaluation model, and identifying key risk factors having great influence on the operation risk of the heavy haul railway overhead line system. According to the method, the causal Bayesian network evaluation model of the operation risk of the heavy haul railway overhead line system is constructed, forward derivation and reverse derivation are carried out, then key risk factors having great influence on the operation risk of the heavy haul railway overhead line system are identified, safety risk prevention and control measures are enhanced, and the operation risk is reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of overhead contact network risk assessment, and in particular relates to a heavy-load railway overhead contact network risk assessment method, medium and equipment based on a Bayesian network. Background Art

[0002] Risk assessment is an important risk management task. It uses scientific methods and means to systematically analyze the threats and vulnerabilities faced by complex systems, evaluate the extent of harm that may be caused by security incidents, and thus propose targeted protection countermeasures and measures.

[0003] The contact network is an important component of the heavy-load railway traction power supply system and the only way for heavy-load trains to obtain electricity. Since the contact network has no backup equipment and plays a special role in heavy-load railway transportation, the safe and reliable operation of the contact network is crucial to the safe operation of heavy-load railways. However, the contact network is completely exposed to the outdoor environment, making it susceptible to weather conditions and equipment aging, which directly affects the safety and order of train operation and seriously affects heavy-load railway transportation. However, there are relatively few studies on risk assessment of heavy-load railway contact network systems, and it is impossible to identify key risk factors that have a greater impact on the operating risks of heavy-load railway contact networks. Therefore, it is impossible to propose targeted protective countermeasures and measures to reduce the operating risks of heavy-load railway contact networks. In view of the above-mentioned defects, the present invention has made improvements. Summary of the invention

[0004] In order to overcome the shortcomings of the background technology, the present invention provides a heavy-duty railway contact network risk assessment method, medium and equipment based on a Bayesian network. By identifying various risk factors of the heavy-duty railway contact network and combining methods such as FMEA to mine and explore the causal relationship between risk factors, a contact network risk propagation network diagram is established. By analyzing and extracting the contact network operation risk causal chain and key risk nodes, it is helpful to strengthen safety risk prevention and control measures and reduce the level of operation risk.

[0005] The technical solution adopted by the present invention is: a heavy-duty railway contact network risk assessment method based on Bayesian network, the method comprising:

[0006] Based on the historical operation data of heavy-haul railway contact network, a causal Bayesian network assessment model for the operation risk of heavy-haul railway contact network is constructed;

[0007] Derivation of the conditional probability of each node in the causal Bayesian network evaluation model based on the statistical analysis of the heavy-haul railway contact network operation historical data;

[0008] Assuming that a system failure occurs in the heavy-haul railway contact network, the posterior probability of each node in the causal Bayesian network evaluation model is reversely derived to identify the key risk factors that affect the operation risk of the heavy-haul railway contact network.

[0009] Further,

[0010] The heavy-duty railway contact network operation historical data includes three types of historical data on climatic conditions, vehicle loads, and equipment failures of the heavy-duty railway contact network operation.

[0011] Further,

[0012] The causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed based on the historical operation data of the heavy-haul railway contact network, including:

[0013] Conduct risk factor analysis based on the climate conditions, vehicle loads, and equipment failure history data to determine the risk factors that affect the safe operation of heavy-duty railway contact networks;

[0014] Based on the risk factor analysis, the risk model and causal relationship analysis of various risk factors are carried out in combination with the heavy-duty railway contact network fault tree model and FMEA method, the risk model and causal relationship of the various risk factors leading to the operation risk of the heavy-duty railway contact network are determined, and a risk model and causal relationship analysis table of the heavy-duty railway contact network is established;

[0015] According to the heavy-haul railway contact network fault tree model and the risk pattern and causal association analysis table, a causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed.

[0016] Further,

[0017] The risk factor analysis includes data statistical analysis and association rule mining.

[0018] Further,

[0019] The causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed according to the heavy-haul railway contact network fault tree model and the risk pattern and causal association analysis table, including:

[0020] Encoding the various risk factors, using the risk factor codes to represent the risk factors as child nodes, and using the heavy-duty railway contact network system failure as the root node to construct a causal Bayesian network assessment model for the heavy-duty railway contact network operation risk;

[0021] Among them, for the sub-nodes of equipment failure type, yes and no are used to represent the failure and normal operating states of the contact network components corresponding to the sub-nodes respectively; for the sub-nodes of climate conditions and vehicle load types, yes and no are used to represent whether the risk factors corresponding to the sub-nodes have an impact or no impact on the contact network operation; for the root node, yes and no are used to represent the failure and normal states of the heavy-load railway contact network system respectively.

[0022] Further,

[0023] The conditional probability of each node in the causal Bayesian network evaluation model is derived based on the statistical analysis of the heavy-haul railway contact network operation history data, using the following expression:

[0024]

[0025] Where M is the number of risk factors related to climate conditions, P is the total number of overhead line equipment, J is the number of risk factors related to vehicle load, P(C M →E P ) is the probability of equipment failure caused by risk factors related to climate conditions; is the number of equipment failures caused by risk factors related to climate conditions; P(T J →E P ) is the probability of equipment failure caused by risk factors related to vehicle load; N is the number of equipment failures caused by risk factors related to vehicle load; E is the total number of equipment failure events.

[0026] Further,

[0027] The construction of the causal Bayesian network evaluation model and the calculation of the probability of each node in the model are based on the Bayesian network analysis software Netica.

[0028] Further,

[0029] After identifying the key risk factors that have an impact on the operational risks of heavy-duty railway contact networks, targeted risk prevention and control measures are proposed based on the key risk factors.

[0030] Based on the same inventive concept, the present invention also provides a computer-readable storage medium storing one or more programs, which, when executed, can implement the aforementioned Bayesian network-based heavy-duty railway contact network risk assessment method.

[0031] Based on the same inventive concept, the present invention also provides an electronic device, comprising a processor, a communication interface, a computer-readable storage medium as described above, and a communication bus; wherein the processor, the communication interface, and the computer-readable storage medium communicate with each other via the communication bus; and the processor is used to execute a program stored in the computer-readable storage medium.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. The constructed causal Bayesian network evaluation model can comprehensively reflect the causal relationship between various risk factors that affect the operation of the heavy-duty railway contact network system. It can well analyze the risks of the heavy-duty railway contact network system through forward deduction, and can well identify the key risk factors that have a greater impact on the risks of the heavy-duty railway contact network system through reverse deduction. It can provide good guidance for the subsequent formulation of risk prevention and control measures, help strengthen safety risk prevention and control measures, and reduce operational risks.

[0034] 2. Climate conditions, vehicle loads and equipment failures (i.e. equipment failures) are selected as the main factors affecting the operating risks of heavy-duty railway contact networks. The historical data on the operation of heavy-duty railway contact networks used mainly include three types of historical data: climate conditions, vehicle loads, and equipment failures of the operation of heavy-duty railway contact networks. This can better determine the risk factors that affect the safe operation of heavy-duty railway contact networks.

[0035] 3. Through statistical analysis of historical data and mining of association rules, combined with the heavy-duty railway contact network fault tree model and FMEA method, risk patterns and causal correlation analysis of various risk factors can be carried out, which can well determine the various factors and scenarios that may lead to system failures or accidents, and then determine the risk patterns and causal correlations of the various risk factors leading to the operation risks of the heavy-duty railway contact network, and establish a risk pattern and causal correlation analysis table for the heavy-duty railway contact network. Finally, according to the heavy-duty railway contact network fault tree model and risk pattern and causal correlation analysis table, a causal Bayesian network evaluation model for the operation risk of the heavy-duty railway contact network is constructed, which can facilitate further risk assessment.

[0036] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings.

[0037] The present invention will be further described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 A schematic diagram of the steps of a heavy-duty railway contact network risk assessment method based on a Bayesian network according to an embodiment of the present invention;

[0040] Figure 2It is a schematic diagram of the overall process of a heavy-duty railway contact network risk assessment method based on a Bayesian network according to an embodiment of the present invention;

[0041] Figure 3 A schematic diagram of a heavy-duty railway overhead contact network fault tree model according to an embodiment of the present invention;

[0042] Figure 4 A schematic diagram of a causal Bayesian network evaluation model for the operation risk of a heavy-duty railway overhead contact network according to an embodiment of the present invention;

[0043] Figure 5 A schematic diagram of forward reasoning of heavy-load railway contact network operation risk according to an embodiment of the present invention;

[0044] Figure 6 A schematic diagram of reverse reasoning of the operating risk of a heavy-load railway contact network according to an embodiment of the present invention;

[0045] Figure 7 The figure is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the implementation of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] like Figures 1 to 7 As shown, this embodiment discloses a heavy-duty railway contact network risk assessment method based on a Bayesian network, comprising:

[0048] Based on the historical operation data of heavy-haul railway contact network, a causal Bayesian network assessment model for the operation risk of heavy-haul railway contact network is constructed;

[0049] Derivation of the conditional probability of each node in the causal Bayesian network evaluation model based on the statistical analysis of the heavy-haul railway contact network operation historical data;

[0050] Assuming that a system failure occurs in the heavy-haul railway contact network, the posterior probability of each node in the causal Bayesian network evaluation model is reversely derived to identify the key risk factors that have a greater impact on the operation risk of the heavy-haul railway contact network.

[0051] The above technical scheme is first based on the historical operation data of the heavy-duty railway contact network, and according to the characteristics of the heavy-duty railway contact network operation risk, a causal Bayesian network evaluation model of the heavy-duty railway contact network operation risk is constructed, and then the conditional probability of each node in the causal Bayesian network evaluation model is forward deduced, and finally the posterior probability of each node in the causal Bayesian network evaluation model is reversely deduced. Among them, the constructed causal Bayesian network evaluation model can more comprehensively reflect the causal relationship between various risk factors affecting the operation of the heavy-duty railway contact network system, and can well analyze the risks of the heavy-duty railway contact network system through forward deduction, and can well identify the key risk factors that have a greater impact on the risks of the heavy-duty railway contact network system through reverse deduction, which can provide good guidance for the subsequent formulation of risk prevention and control measures, help to strengthen safety risk prevention and control measures, and reduce operation risks.

[0052] The above-mentioned heavy-duty railway contact network risk assessment method based on Bayesian network will be further explained below.

[0053] According to the characteristics and actual working conditions of the heavy-load railway contact network operation risk, this embodiment selects climate conditions, vehicle load and equipment failure (i.e., equipment failure) as the main factors affecting the heavy-load railway contact network operation risk after analysis and summary. Therefore, the heavy-load railway contact network operation historical data used mainly include three types of historical data such as climate conditions, vehicle load, and equipment failure of the heavy-load railway contact network operation. In this embodiment, the climate conditions and vehicle load historical data mainly refer to the climate conditions (weather data), vehicle load (heavy-load railway load data), bow-net conditions and other data when the equipment fails (i.e., the heavy-load railway contact network components fail).

[0054] According to the climate conditions, vehicle loads, and equipment failure history data, risk factor analysis is performed. The risk factor analysis preferably determines the risk factors that affect the safe operation of the heavy-duty railway contact network through data statistical analysis and association rule mining. In some embodiments, the data statistical analysis mainly includes counting the number of safety incidents caused by various safety risk factors, and the association rule mining adopts the Apriori association rule method, which can better determine the risk factors that affect the safe operation of the heavy-duty railway contact network. Both the data statistical analysis and the Apriori association rule method can be further referred to the prior art, which will not be described in detail here.

[0055] The following will analyze the risk factors of climate conditions, vehicle load, and equipment failure in turn.

[0056] The impact of climatic conditions on the safe operation of heavy-duty railway contact networks is analyzed as follows:

[0057] (1) Lightning strikes are one of the most serious threats to the overhead contact system. Lightning strikes may directly hit the overhead contact system or nearby equipment, causing equipment failure, damage, or even fire.

[0058] (2) Strong winds may cause lateral loads on structures such as the overhead contact network pillars and suspension systems, which may affect the stability and safety of the overhead contact network. In addition, winds may also blow down trees or other objects, increasing the risk of damage to the overhead contact network.

[0059] (3) Haze weather may cause dust and dirt to accumulate on the surface of overhead contact lines and equipment, reducing the insulation performance of the lines and increasing the risk of short circuit or disconnection in the overhead contact system.

[0060] (4) In addition, in cold climates, ice and snow may accumulate on the contact network, increasing the chances of the contact network being affected by wind and possibly causing the insulation performance of the line to deteriorate, and even leading to the risk of short circuit or disconnection of the contact network system; high humidity and frequent precipitation may cause corrosion of contact network components, especially metal components. Corrosion may reduce the strength and durability of contact network components and increase the frequency of maintenance and replacement.

[0061] The impact of vehicle load on the safe operation of heavy-duty railway contact network is analyzed as follows:

[0062] The impact of vehicle load on heavy-duty railway contact network mainly includes load pressure, dynamic load effect, space limitation and electrical impact:

[0063] (1) Trains running on heavy-duty railways usually have large loads, which will increase the stress on the contact network. The weight and speed of trains have had a significant impact on the damage to the contact network. Long-term excessive load pressure may cause fatigue, deformation or wear of the contact network components, thereby affecting the normal operation of the contact network.

[0064] (2) When a train passes through the contact network, a dynamic load effect will be generated, that is, the contact network is affected by a transient load. This transient load may be greater than the static load, causing vibration and stress concentration in the contact network components. If the contact network is not designed properly or the components are severely fatigued, the dynamic load effect may accelerate the aging and damage of the contact network.

[0065] (3) Trains on heavy-haul railways are usually larger and heavier than ordinary trains, so they may require more space when passing through the overhead contact network. If the space around the overhead contact network is limited, the risk of collision between the train and the overhead contact network may increase, resulting in damage to the overhead contact network.

[0066] (4) Heavy-load railway trains usually require more power supply to ensure high-speed travel and load transportation. Therefore, when a large number of heavy-load trains pass through the overhead contact network at the same time, the electrical load may increase, causing impacts on the overhead contact network power supply system, such as voltage fluctuations and overloads, which in turn affect the stability and safety of the overhead contact network.

[0067] The impact of equipment failure on the safe operation of heavy-duty railway contact network is analyzed as follows:

[0068] (1) The overhead contact network is composed of many different components, such as pillars, beams, conductors, etc. The failure of these components may affect the stability of the overhead contact network. For example, a broken pillar or loose beam may cause the overhead contact network line to sag or deviate, thus affecting the normal operation of the train.

[0069] (2) Insulators are key components in the overhead contact system that play an insulating role and are used to isolate electrical conductors from the support or the ground. If the insulators are damaged, aged or contaminated, they may cause a short circuit or leakage in the overhead contact system, thus affecting the safety and operation of the train.

[0070] (3) The contact network power supply system includes power supply devices, contact wires, cables and other components. If any of them fails, the train may not be able to obtain power normally, affecting its running speed and capacity, and even causing safety problems such as stopping.

[0071] (4) Over time, contact network equipment will age, such as metal fatigue and plastic material aging, which may lead to a decrease in the strength and stability of contact network components and increase the risk of failure.

[0072] (5) In addition, if the overhead line monitoring system fails, it may not be possible to detect abnormal conditions of the overhead line equipment in a timely manner, thereby delaying the time for handling the fault and increasing the possibility of problems occurring during train operation.

[0073] In heavy-duty railway contact network systems, risk patterns may involve multiple factors such as equipment performance degradation, external environmental factors, and vehicle loads, and the correlation between these factors has an important impact on the safety of the system.

[0074] This embodiment uses historical data statistical analysis and association rule mining, combined with the heavy-duty railway contact network fault tree model and FMEA method to analyze the risk pattern and causal relationship of various risk factors, determine the various factors and scenarios that may lead to system failures or accidents, and then determine the risk pattern and causal relationship of the heavy-duty railway contact network operation risks caused by various risk factors, and establish a risk pattern and causal relationship analysis table for the heavy-duty railway contact network, as shown in Table 1 below.

[0075] Failure Mode, Effect and Critical Analysis (FMEA) is a set of analysis methods, including Failure Mode Analysis (FMA), Failure Effect Analysis FEA, Failure Critical Analysis FCA and various combinations thereof, often referred to as FMEA. It is a method for analyzing and evaluating potential failures, which was developed in reliability design. It originated from the recognition in the aviation industry that a large safety factor does not necessarily translate into product reliability, but leads to over-designed products with reliability problems. FMEA mainly conducts systematic analysis and evaluation of possible failure modes and their effects on products or production processes to determine the root causes that may lead to failures, and take preventive and corrective measures to reduce risks and improve quality. In practice, FMEA can help companies discover problems and avoid the same mistakes in the future, improve the reliability of product design, reduce costs and improve process quality. Therefore, this embodiment adopts the FMEA method, and combines the heavy-duty railway contact network fault tree model and the FMEA method to perform risk pattern and causal relationship analysis on the various risk factors, which can well determine the risk pattern and causal relationship of the heavy-duty railway contact network operation risk caused by the various risk factors, and then establish a risk pattern and causal relationship analysis table for the heavy-duty railway contact network.

[0076] Table 1 Risk model and causal relationship analysis of heavy-duty railway contact network

[0077]

[0078] To facilitate the subsequent construction of a causal Bayesian network assessment model for the operation risk of heavy-duty railway contact networks, this embodiment also encodes the various risk factors in the risk model and causal association analysis table, including 5 climate condition risk factors, coded C1-C5 in sequence, 4 vehicle load risk factors, coded T1-T4 in sequence, and 12 equipment failure risk factors, coded E1-E12 in sequence.

[0079] The heavy-duty railway contact network fault tree model of this embodiment is also based on the above-mentioned equipment failure risk factor codes E1-E12. In this model, the E node is a system failure, and the definitions of the remaining nodes are consistent with Table 1. The contact network system and its components have only two operating states, namely normal and failure, and the components are in series logic. If one or more components fail, the system will fail and maintenance activities need to be carried out until the system returns to normal working state. This embodiment takes "contact network system failure" as the basic event, and establishes a contact network system fault tree model (i.e., a heavy-duty railway contact network fault tree model) as shown in FIG. Figure 3 shown.

[0080] This embodiment constructs a causal Bayesian network evaluation model based on a heavy-duty railway contact network fault tree model, in which the nodes in the heavy-duty railway contact network fault tree model are used as key nodes of the causal Bayesian network evaluation model (i.e., subnodes of equipment failure risk factors). Specifically, through the key nodes (E1-E12) of the heavy-duty railway contact network fault tree model, these 12 nodes are used as key nodes of the causal Bayesian network evaluation model, and then through the risk model and causal association analysis table of the heavy-duty railway contact network, if lightning strikes the contact line (E4), it will cause the contact line to burn or be damaged, and then link the arrows between them, indicating that lightning strikes will cause equipment failure.

[0081] This embodiment constructs a causal Bayesian network assessment model for the operation risk of the heavy-haul railway contact network based on the heavy-haul railway contact network fault tree model and the risk pattern and causal association analysis table.

[0082] Specifically, after risk factor coding is performed on the various risk factors, the risk factor codes are used to represent the risk factors as sub-nodes, and the heavy-duty railway contact network system failure is used as the root node to construct a causal Bayesian network evaluation model for the operation risk of the heavy-duty railway contact network; wherein, for the sub-nodes of the equipment failure type, yes and no are used to represent the failure and normal operation states of the contact network components corresponding to the sub-nodes respectively; for the sub-nodes of the climate conditions and vehicle load types, yes and no are used to represent that the risk factors corresponding to the sub-nodes have an impact or no impact on the contact network operation; for the root node, yes and no are used to represent the failure and normal states of the heavy-duty railway contact network system respectively.

[0083] Based on the established risk model of heavy-duty railway contact network and causal correlation analysis table, the factor correlation between risk factors can be determined. The contact network system and components are represented by two operating states, "failure (yes)" and "normal (no)" (i.e. corresponding to equipment failure risk factors), and other climatic conditions and vehicle loads are represented by "yes" and "no" to indicate their impact and no impact on the contact network operation (i.e. corresponding to climatic conditions and vehicle load risk factors). These risk factors ultimately lead to heavy-duty railway contact network system failures (system nodes, "failure (yes)" and "normal (no)"). In specific implementation, it is preferred to use the Bayesian network analysis software Netica to build a causal Bayesian network assessment model for the operation risk of heavy-duty railway contact networks. After the construction is completed, Figure 4 In the figure, a line without an arrow indicates no causal relationship, and a line with an arrow indicates a causal relationship.

[0084] The construction of the causal Bayesian network evaluation model of this embodiment is based on the Bayesian network, which is a probabilistic graphical model used to describe the dependencies between random variables. It is based on Bayes' theorem and uses a directed acyclic graph to represent the conditional dependencies between variables. The network consists of two parts: a graph structure and a conditional probability table. The graph structure is a directed acyclic graph consisting of nodes and directed edges, G = (X, E), where the nodes represent random variables, and the directed edges between the nodes represent the causal relationship between the nodes. The conditional probability table reflects the strength of the causal relationship between exogenous variables and endogenous variables, and the probability of occurrence of the variable is called the prior probability. The Bayesian network structure is established with the help of nodes and directed edges, and the Bayesian network is given mathematical meaning with the help of the conditional probability table, so that the Bayesian network model can solve practical problems qualitatively and quantitatively. Its expression is:

[0085]

[0086] Where Pa(X n ) is the node X n The parent node of , N is the number of random variables.

[0087] The basic probability law of Bayesian learning is

[0088] P(X i , X j )=P(X i |X j )P(X j )

[0089] In the formula, P(X i |X j ) is node X i At node X j Conditional probability of occurrence; P(X j ) is node X j Probability of occurrence.

[0090] Bayesian network technology has powerful and flexible complex system modeling capabilities and high computational efficiency. Most importantly, it can qualitatively and quantitatively describe the correlation between accident causes, and its probabilistic reasoning function has outstanding advantages in causal analysis, and performs very well in accident cause analysis. Therefore, this embodiment uses a causal Bayesian network evaluation model based on Bayesian network to perform heavy-duty railway contact network risk assessment.

[0091] When conducting risk assessment on the heavy-duty railway contact network, it is necessary to analyze and determine the conditional probabilities of all nodes in the causal Bayesian network assessment model (i.e., the probability of risk propagation), and the changes in all nodes will be reflected in the risk changes of the heavy-duty railway contact network system. This embodiment derives the conditional probability of each node in the causal Bayesian network assessment model (i.e., determining the possibility of risk propagation) based on the statistical analysis of the historical operation data of the heavy-duty railway contact network, and adopts the following expression:

[0092]

[0093] Where M is the number of risk factors related to climate conditions, P is the total number of overhead line equipment, J is the number of risk factors related to vehicle load, P(C M →E P ) is the probability of equipment failure caused by risk factors related to climate conditions; is the number of equipment failures caused by risk factors related to climate conditions; P(T J →E P ) is the probability of equipment failure caused by risk factors related to vehicle load; N is the number of equipment failures caused by risk factors related to vehicle load; E The equipment failure in this embodiment mainly includes various equipment failures in the risk mode and causal association analysis table.

[0094] According to IEC standards, the following formula is generally used for risk assessment calculations.

[0095] R=P(A)P(L|A)

[0096] Where R represents the system risk level; P(A) represents the probability of event A occurring; and P(L / A) represents the probability that event A leads to consequence or loss L.

[0097] For example, the probability of casing failure (self-cause) P(E 12 = yes | C5 = no) = 0.04152, the probability of casing failure under high humidity and frequent precipitation P (E 12 = yes | C5 = yes) = 0.5, the probability of high humidity and frequent precipitation is P(C5 = yes) = 0.0124, then

[0098] P(E 12 =yes) = P(E 12 =yes|C5=yes)*P(C5=yes)+P(E 12 =yes|C5=no)*P(C5=no)+P(E 12 =no|C5=yes)*P(C5=yes)+P(E12 =no|C5=no)*P(C5=no)

[0099] =0.0472

[0100] Combining the above-mentioned causal Bayesian network evaluation model and the derived risk propagation probability, the Bayesian network analysis software Netica is used to construct the overall operation risk of heavy-duty railway contact network. Figure 5 shown.

[0101] The overall operation risk diagram of heavy-haul railway contact network can be used to analyze the operation risk of heavy-haul railway contact network. Figure 5 It can be seen that when the combined effects of climatic conditions and vehicle loads are superimposed, the operating risks of positioning devices and insulators are extremely high, and the probability of risk is as high as 30%. In addition, the probability of risk in equipment such as pillar foundations, load-bearing cables, contact wires and suspension strings is also relatively high, about 15%. Since the contact network system is regarded as a series system, the failure of each device will lead to system failure, so the probability of system failure is relatively high, reaching about 45%. It can be seen that the forward reasoning of the causal Bayesian network evaluation model based on the operating risk of heavy-duty railway contact networks can well analyze the overall operating risk of heavy-duty railway contact network systems.

[0102] This embodiment also performs reverse reasoning on the operation risk of the heavy-duty railway contact network. Specifically, combined with the reverse reasoning ability of the causal Bayesian network, it sets the probability value of the final node "yes" to 100%, that is, assuming that a risk event occurs in the contact network system, the probability of each basic event corresponding to the node is converted into a posterior probability through the risk propagation probability, which represents the probability of each node occurring under the condition that the heavy-duty railway contact network system has generated risks. Through reverse deduction, the risk factors that pose a greater risk to the operation of the heavy-duty railway contact network can be analyzed. Set the node system to occur, and then perform reverse reasoning operations to obtain the following: Figure 6 The reverse reasoning diagram of the heavy-haul railway contact network operation risk is shown in Figure 1. In specific implementation, the reverse reasoning diagram of the heavy-haul railway contact network operation risk is also constructed using the Bayesian network analysis software Netica.

[0103] Depend on Figure 6It can be seen that when the heavy-duty railway contact network encounters risks, the reverse reasoning results show that the posterior probabilities of positioning device failure (E2) and insulator failure (E11) are relatively high, at 0.345 and 0.356 respectively, which is basically consistent with the actual probability of the occurrence of the Jingshen Railway contact network operation risk. In addition, risks in equipment such as support foundations, load-bearing cables, contact wires and suspension strings can also easily lead to system failures. It can be seen that the reverse reasoning based on the causal Bayesian network evaluation model of the heavy-duty railway contact network operation risk can well identify the key risk factors that have a greater impact on the system risk of the heavy-duty railway contact network (the key risk factors identified in this embodiment are mainly positioning device failure and insulator failure), which can provide good guidance for the subsequent formulation of risk prevention and control measures, help strengthen safety risk prevention and control measures, and reduce operational risks.

[0104] The present invention identifies various risk factors (climate conditions, vehicle loads, equipment failures) of the heavy-duty railway contact network, and combines FMEA and other methods to mine and explore the causal relationship between risk factors, thereby establishing a risk propagation network diagram of the heavy-duty railway contact network (i.e., a causal Bayesian network evaluation model). By forward reasoning and reverse reasoning, the causal chain of contact network operation risks and key risk nodes are analyzed and extracted, which can help strengthen safety risk prevention and control measures and reduce the level of operational risk. The present invention first analyzes and ascertains the operational risk factors that affect the heavy-duty railway contact network, constructs a risk assessment model (i.e., a causal Bayesian network evaluation model) to mine key risk nodes, and proposes targeted risk prevention and control measures, which is of great significance to improving the safety level of system operation.

[0105] After identifying the key risk factors that have a significant impact on the operating risk of the heavy-load railway contact network, this embodiment proposes targeted risk prevention and control measures based on the key risk factors.

[0106] In the specific implementation, regular maintenance can be strengthened for the vulnerable equipment corresponding to the key risk factors (this embodiment mainly takes the positioning device, insulator and pillar foundation as examples for explanation), and technical upgrades and improvements are given priority. In addition, data monitoring and analysis are strengthened, as follows:

[0107] (1) Regular maintenance: Establish a comprehensive regular inspection and maintenance system to regularly inspect, clean and maintain overhead contact network equipment (especially equipment that is prone to failure and corresponds to key risk factors) to ensure that the equipment is in good condition.

[0108] For the positioning device (E2), regularly check the accuracy of the positioning device to ensure that it can accurately maintain the positioning of the contact network; calibrate and adjust when necessary; lubricate the moving parts of the positioning device to reduce wear; and replace severely worn parts in a timely manner. For insulators (E11), regularly clean the insulators to remove dust and pollutants to maintain their insulation performance. Use an insulation tester to test the electrical performance of the insulators, and replace insulators with degraded performance to avoid affecting the normal operation of the system. For the pillar foundation (E1), regularly check whether the foundation has settlement, cracks or corrosion. Clean up the debris and accumulated water around the foundation to prevent further damage to the foundation. For foundations with problems, promptly reinforce or repair the structure. Consider using high-strength materials for reinforcement to extend the service life of the foundation.

[0109] (2) Technology upgrade and improvement: Actively adopt advanced technologies to carry out necessary upgrades and improvements to overhead line equipment (especially equipment prone to failure corresponding to key risk factors) to improve equipment performance and reliability.

[0110] For the positioning device (E2), optimize the structural design of the positioning device to reduce friction and wear and improve durability; for the insulator (E11), use new composite materials or nanomaterials to manufacture insulators to improve their resistance to pollution and aging; for the pillar foundation (E1), consider using new high-performance materials to replace traditional materials to improve the corrosion resistance and structural strength of the pillar foundation.

[0111] (3) Data monitoring and analysis: Establish a comprehensive data monitoring and analysis system to monitor and analyze the operating data of overhead line equipment (especially equipment prone to failure corresponding to key risk factors) in real time, so as to promptly identify and resolve potential problems.

[0112] For the positioning device (E2), a real-time monitoring system is deployed to record the working status, operating deviation and other data of the positioning device, and use data analysis technology to optimize the working performance of the positioning device. For the insulator (E11), a comprehensive analysis of the monitoring data is carried out to identify the aging trend or abnormal conditions that may occur in the insulator, so as to take preventive measures. For the pillar foundation (E1), sensors are installed at the foundation to monitor the settlement, vibration and other data of the foundation in real time, and combined with data analysis, predict potential problems of the foundation. In addition, historical data is used to predict faults, provide early warnings, and reduce the incidence of equipment failures.

[0113] Risk prevention and control measures for heavy-duty railway contact networks are an important means to ensure the safe and stable operation of the system. When formulating these measures, it is necessary to comprehensively consider various potential risk factors of the system and take effective measures to reduce the possibility and impact of these risks. Therefore, in addition to the above three risk prevention and control measures, other prevention and control measures are also taken in detail during implementation, including but not limited to:

[0114] (1) Climate monitoring and early warning: Establish a meteorological monitoring system to timely monitor and warn of severe weather that may affect overhead contact network equipment, and take corresponding preventive and protective measures.

[0115] (2) Load management: Rationally manage and control vehicle loads to avoid overloading and reduce additional pressure on contact network equipment.

[0116] (3) Emergency response mechanism: A sound emergency response mechanism should be established so that effective measures can be taken promptly to deal with and repair equipment failures or accidents, thereby minimizing losses.

[0117] (4) Personnel training and awareness raising: Strengthen the training and awareness raising of relevant personnel, increase their attention to the safe operation of contact network equipment, and reduce the impact of human errors on system safety.

[0118] (5) Cooperation and coordination: Strengthen cooperation and coordination with relevant departments and institutions, jointly study and solve various challenges and problems in the operation of the overhead line, and form a joint force to ensure the safe and stable operation of the system.

[0119] Some possible risk prevention and control measures are given above. By strengthening these safety risk prevention and control measures, the operating risks of heavy-load railway contact networks can be effectively reduced.

[0120] In summary, the present invention analyzes and summarizes the main factors affecting the operating risks of heavy-duty railway contact networks, including climatic conditions, vehicle loads and equipment failure (i.e., equipment failure), and analyzes the causal relationship between them through a causal Bayesian network evaluation model based on a Bayesian network, identifies and analyzes risk patterns and causal associations, verifies and evaluates the evaluation model through actual data, and provides risk prevention and control measures for the operation of heavy-duty railway contact networks, aiming to improve the reliability and safety of heavy-duty railway contact networks, thereby ensuring the smooth progress of railway transportation.

[0121] Based on the same inventive concept, the present invention also provides a computer-readable storage medium storing one or more programs, which, when executed, can implement the aforementioned Bayesian network-based heavy-duty railway contact network risk assessment method.

[0122] Based on the same inventive concept, the present invention also provides an electronic device, comprising a processor, a communication interface, a computer-readable storage medium as described above, and a communication bus; wherein the processor, the communication interface, and the computer-readable storage medium communicate with each other via the communication bus; and the processor is used to execute a program stored in the computer-readable storage medium.

[0123] Parts not involved in this embodiment are the same as the prior art or can be implemented by using the prior art, and will not be further described here.

[0124] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A heavy-duty railway contact network risk assessment method based on Bayesian network, characterized in that: include: Based on the historical operation data of heavy-haul railway contact network, a causal Bayesian network assessment model for the operation risk of heavy-haul railway contact network is constructed; Derivation of the conditional probability of each node in the causal Bayesian network evaluation model based on the statistical analysis of the heavy-haul railway contact network operation historical data; Assuming that a system failure occurs in the heavy-haul railway contact network, the posterior probability of each node in the causal Bayesian network evaluation model is reversely derived to identify the key risk factors that affect the operation risk of the heavy-haul railway contact network.

2. The heavy-duty railway contact network risk assessment method based on Bayesian network according to claim 1 is characterized in that: The heavy-duty railway contact network operation historical data includes at least three types of historical data including climatic conditions, vehicle loads, and equipment failures of the heavy-duty railway contact network operation.

3. The heavy-duty railway contact network risk assessment method based on Bayesian network according to claim 2 is characterized in that: The causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed based on the historical operation data of the heavy-haul railway contact network, including: Conduct risk factor analysis based on the climate conditions, vehicle loads, and equipment failure history data to determine the risk factors that affect the safe operation of heavy-duty railway contact networks; Based on the risk factor analysis, the risk model and causal relationship analysis of various risk factors are carried out in combination with the heavy-duty railway contact network fault tree model and FMEA method, the risk model and causal relationship of the various risk factors leading to the operation risk of the heavy-duty railway contact network are determined, and a risk model and causal relationship analysis table of the heavy-duty railway contact network is established; According to the heavy-haul railway contact network fault tree model and the risk pattern and causal association analysis table, a causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed.

4. The heavy-duty railway contact network risk assessment method based on Bayesian network according to claim 3 is characterized in that: The risk factor analysis includes data statistical analysis and association rule mining.

5. The heavy-duty railway contact network risk assessment method based on Bayesian network according to claim 3 is characterized in that: The causal Bayesian network evaluation model for the operation risk of the heavy-haul railway contact network is constructed according to the heavy-haul railway contact network fault tree model and the risk pattern and causal association analysis table, including: Encoding the various risk factors, using the risk factor codes to represent the risk factors as child nodes, and using the heavy-duty railway contact network system failure as the root node to construct a causal Bayesian network assessment model for the heavy-duty railway contact network operation risk; Among them, for the sub-nodes of equipment failure type, yes and no are used to represent the failure and normal operating states of the contact network components corresponding to the sub-nodes respectively; for the sub-nodes of climate conditions and vehicle load types, yes and no are used to represent whether the risk factors corresponding to the sub-nodes have an impact or no impact on the contact network operation; for the root node, yes and no are used to represent the failure and normal states of the heavy-load railway contact network system respectively.

6. A heavy-duty railway contact network risk assessment method based on Bayesian network according to any one of claims 3 to 5, characterized in that: The conditional probability of each node in the causal Bayesian network evaluation model is derived based on the statistical analysis of the heavy-haul railway contact network operation history data, using the following expression: Where M is the number of risk factors related to climate conditions, P is the total number of overhead line equipment, J is the number of risk factors related to vehicle load, P(C M →E P ) is the probability of equipment failure caused by risk factors related to climate conditions; is the number of equipment failures caused by risk factors related to climate conditions; P(T J →E P ) is the probability of equipment failure caused by risk factors related to vehicle load; N is the number of equipment failures caused by risk factors related to vehicle load; E is the total number of equipment failure events.

7. The heavy-duty railway contact network risk assessment method based on Bayesian network according to claim 1 is characterized in that: The construction of the causal Bayesian network evaluation model and the calculation of the probability of each node in the model are based on the Bayesian network analysis software Netica.

8. The heavy-haul railway overhead contact network risk assessment method based on Bayesian network according to claim 1, characterized in that: After identifying the key risk factors that have an impact on the operational risks of heavy-duty railway contact networks, targeted risk prevention and control measures are proposed based on the key risk factors.

9. A computer-readable storage medium storing one or more programs, characterized in that: When the one or more programs are executed, the heavy-duty railway contact network risk assessment method based on Bayesian network as described in any one of claims 1 to 8 can be implemented.

10. An electronic device comprising a processor, a communication interface, the computer-readable storage medium of claim 9, and a communication bus; wherein: The processor, the communication interface, and the computer-readable storage medium communicate with each other via a communication bus; It is characterized in that The processor is configured to execute a program stored in a computer-readable storage medium.

Images