A port accident risk visualization analysis method based on network topology structure

Abstract

The application discloses a port accident risk visualization analysis method based on a network topology structure, and comprises the following steps: taking key facilities and important areas as network nodes of a port, determining a connection relationship to obtain a network topology structure, analyzing historical cases to determine a risk factor set, obtaining an uncertain factor set through edge computing, constructing an accident tree to quantitatively evaluate the network nodes and connection risks, constructing a risk evolution model, extracting an accident causation chain and calculating a risk influence range, visualizing the network topology structure, adding the accident causation chain and the risk influence range after color mapping, and obtaining a three-dimensional topology view. The method can realize the network topology structure visualization analysis of the port accident risk, comprehensively evaluate the risk, optimize subsequent port management decisions, and has good interpretability.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of port engineering management, and particularly relates to a port accident risk visualization analysis method based on network topology. BACKGROUND

[0002] With the continuous development and expansion of port business, the safety management of the port area becomes crucial. In modern port areas, various equipment, facilities, and personnel activities are intertwined, forming a complex network topology.

[0003] Traditional port accident risk analysis methods often rely on experience and simple data statistics, lacking in-depth understanding and visualization of the overall network structure of the port area. This makes it difficult to fully and accurately grasp the distribution and transmission path of risks in accident risk assessment and management. At the same time, with the rapid development of information technology, network topology analysis technology has been widely applied in various fields. Through the analysis of network structure, the relationship and mutual influence between nodes in the system can be revealed, providing strong support for risk assessment and decision-making.

[0004] A port accident risk visualization analysis method based on network topology uses network topology analysis to comprehensively analyze the accident risk in combination with the actual situation of the port area, and visually displays the distribution and transmission path of risks through visualization means, providing scientific basis and decision support for the safety management of the port area. SUMMARY

[0005] The purpose of the present application is to provide a port accident risk visualization analysis method based on network topology.

[0006] To achieve the above purpose, the present application is implemented according to the following technical solutions:

[0007] The present application provides a port accident risk visualization analysis method based on network topology, comprising:

[0008] S100 sets key facilities and important areas in the port area as network nodes, determines the connection relationship between network nodes according to actual physical connections and business processes, and obtains the network topology of the port area;

[0009] S200 analyzes historical accident cases of the network topology to obtain a set of determined risk factors; performs edge computing on the network topology to obtain a set of uncertain risk factors;

[0010] S300 constructs an accident tree based on the set of determined risk factors and the set of uncertain risk factors, and quantitatively evaluates the risk of the network topology based on the accident tree;

[0011] S400 constructs a risk evolution model, extracts an accident causation chain from the quantitative assessment result through the risk evolution model, and calculates a risk influence range according to the accident causation chain;

[0012] S500 visualizes the network topology structure, adds the accident causation chain and the risk influence range to the visualized result after color mapping, and obtains a three-dimensional topology view.

[0013] As a further method, the historical accident cases of the network topology structure are analyzed, and a method for obtaining a determined risk factor set comprises:

[0014] An accident record of a historical accident case is obtained, network topology data related to the accident is collected, including node information, connection relationship and edge flow data, and a key factor of the accident is identified;

[0015] A correlation coefficient of the key factor and the accident loss degree is calculated, and the expression is:

[0016]

[0017] Wherein, m is the number of historical accident cases, F ik represents the value of the key factor F i in the kth case, is the mean value of the key factor F i , L k is the accident loss degree of the kth case, is the mean value of the accident loss degree of all cases;

[0018] The key factor with a correlation coefficient of the key factor and the accident loss degree greater than 0.01 is determined as a risk factor, and a determined risk factor set is obtained.

[0019] As a further method, the network topology structure is edge calculated, and a method for obtaining an uncertain risk factor set comprises:

[0020] Edge calculation is performed on each network node to obtain a state variable of each network node;

[0021] The state variables between the network nodes are combined in any way, a correlation coefficient matrix is obtained by calculating a Pearson correlation coefficient, and the correlation coefficient matrix is obtained.

[0022] The factors corresponding to the state variables greater than a preset threshold in the correlation coefficient matrix are obtained, the repeated factors in the determined risk factor set are removed, and an uncertain risk factor set is obtained.

[0023] As a further method, the method of constructing an accident tree based on the determined risk factor set and the uncertain risk factor set comprises: taking the existence of a risk in a port area as a top event, taking each factor in the determined risk factor set and the uncertain risk factor set as a basic event, connecting the basic events using logical gates of AND gates and OR gates according to a logical relationship, obtaining an intermediate event, and forming an accident tree.

[0024] As a further method, the method of quantitatively evaluating the risk of a network topology based on the accident tree comprises:

[0025] S21 calculates the risk probability of a corresponding basic event in the determined risk factor set by using the Bayesian network, specifically:

[0026] The basic events in the accident tree are taken as nodes of the Bayesian network, and the Bayesian network is constructed.

[0027] A kernel function is established by using a hyperparameter method to correct the Bayesian network, and an expression of the kernel function is:

[0028]

[0029] wherein α is a global scaling factor, x and y are state variables and characteristic variables of events in the Bayesian network respectively, σ is a bandwidth of a Gaussian radial basis function, T represents a transposition operation, M is a normalization factor, d is a degree of a polynomial kernel, β and b are a slope and a bias of a Sigmoid kernel respectively, and p is a nonlinear index of the kernel function.

[0030] The probability distribution in the Bayesian network is updated by using an expectation maximization, and an expression of the update is:

[0031]

[0032] wherein θ is a parameter set in the Bayesian network, θ t and θ (t+1) are estimations of θ in the tth iteration and the t+1th iteration respectively, arg maxθ is a maximization operator, is an expectation operator, P(X hidden |D,θ (t) ) is a posterior probability distribution of a hidden node X hidden under a condition that observation data D and parameter estimation θ (t) are given, is a log-likelihood function of a joint probability distribution, n is a number of nodes in the Bayesian network, and P(X i |pa(X i );θ) represents an estimated node X i given its parent node pa(X i) under the condition of the basic event;

[0033] The probability distribution in the Bayesian network is taken as the risk probability of the basic event;

[0034] S22 calculates the risk probability of the corresponding basic event in the set of non-deterministic factors by using fuzzy logic, specifically:

[0035] Wherein the expression of the membership function is:

[0036]

[0037] Wherein, λ i is the i th non-deterministic risk factor, c i is the central value of x i , is the standard deviation of λ i ;

[0038] The membership function is de-fuzzified by the centroid method to obtain the risk probability of the basic event;

[0039] S23 calculates the risk probability of the network node and the top event according to the risk probability of the basic event by the fault tree, and the expression is:

[0040]

[0041] Wherein, E is a basic event or a complex event composed of basic events, P(ξ i ) is the risk probability of the basic event ξ i ;

[0042] S24 takes the risk probability of each network node and top event as the quantitative evaluation result.

[0043] As a further method, the method for constructing a risk evolution model and extracting an accident causation chain from the quantitative evaluation result through the risk evolution model, comprising:

[0044] A Markov chain algorithm is used to construct the risk evolution model, wherein the expression of the state transition function is:

[0045]

[0046] Wherein, P(X n+1 =j|X n =i) is the probability of transferring to the target state j at time step n+1 under the condition of being in state i at time step n, S is the set of all states, U is the set of deterministic risk factors, p u (n) is the probability distribution of the deterministic risk factor u at time step n, V is the set of non-deterministic risk factors, and q v(n) is the probability distribution of the uncertain risk factor v at time step n, p kj is the transition probability from the intermediate state k to the target state j.

[0047] The historical accident causation chain data is obtained from historical accident cases, and is divided into a training set, a test set and a verification set according to a ratio of 8:1:1. The risk evolution model is trained, and the risk probability of each network node is taken as input to obtain an accident causation chain.

[0048] The causation intensity between each network node and the risk source node is calculated, and the expression is:

[0049]

[0050] where ω1, ω2, ω3, ω4 and ω5 are weight coefficients of degree centrality, betweenness centrality, closeness centrality, eigenvector centrality and clustering coefficient, respectively, D(v), B(v), C(v), E(v) and T(v) are the measurement values of degree centrality, betweenness centrality, closeness centrality, eigenvector centrality and clustering coefficient, respectively, deg + (v) and deg - (v) are the in-degree and out-degree of node v, respectively, N is the total number of nodes in the topological network structure, σ st (v) and σ st are the number of shortest paths passing through node v and the total number of shortest paths from node s to node t, respectively, d(v, t) is the shortest path length from node v to node t, λ is a normalization constant, A vt is the connection weight between node v and node t, E(t) is the eigenvector centrality score of node t, T v is the actual number of edges between the neighbor nodes of node v, and deg(v) is the degree of node v.

[0051] As a further method, the method of calculating the risk influence range according to the accident causation chain comprises:

[0052] The risk source influence range radius is calculated by quantifying the evaluation results with the risk source node as the center, and the expression is:

[0053]

[0054] where N i is the risk source node, P(T|N i ) is the risk probability of node N i , η is the attenuation coefficient of risk propagation, and τ is the minimum acceptable level of risk influence.

[0055] The influence range radius of the remaining nodes on the causation chain is calculated based on the causation intensity between each network node and the risk source node, and the expression is:

[0056] R i = ψ i × R0

[0057] wherein, R i is the influence range of the i-th network node, ψ i is the causation intensity of the i-th network node and the risk source node.

[0058] As a further method, the method of visualizing the network topology, adding the accident causation chain and the risk influence range to the visualization result after color mapping, comprises:

[0059] Taking the risk probability of each network node in the quantitative evaluation result as the attribute of the node, taking the risk probability of the top event as the overall attribute, grading the risk probability of each network node and the top event;

[0060] Creating a three-dimensional view of the port network topology, using different colors and depths to represent the risk probability of each network node and the top event, and mapping the accident causation chain and the risk influence range to the three-dimensional view through color mapping;

[0061] Adding interactive controls to allow users to filter and view data according to risk probability levels or information on the accident causation chain.

[0062] In a second aspect, the embodiments of the present application further provide an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the method steps of the first aspect.

[0063] In a third aspect, the embodiments of the present application further provide a computer readable storage medium storing one or more programs, which, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method steps of the first aspect.

[0064] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects:

[0065] (1) The present application quantitatively evaluates by constructing an accident tree based on a set of certain risk factors and a set of uncertain risk factors, and uses Bayesian networks and fuzzy logic methods to assign specific numerical values to risks and provide accurate basis for decision-making.

[0066] (2) The present application extracts the accident causation chain from the quantitative evaluation result by establishing a risk evolution model, which helps to dynamically analyze the development and transmission path of the risk, provides an early warning of the spread of potential risks, and provides a basis for formulating effective risk control strategies. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 A flow chart of steps of a port accident risk visualization analysis method based on network topology structure in an embodiment of the present application.

[0068] Figure 2 A structural schematic diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0069] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0070] Referring to Figure 1 The present application provides a port accident risk visualization analysis method based on network topology structure, which comprises:

[0071] S100 sets key facilities and important areas in the port as network nodes, determines the connection relationship between the network nodes according to actual physical connection and business process, and obtains the network topology structure of the port;

[0072] Taking a medium-sized comprehensive port as an example, the port mainly operates container transportation, bulk cargo loading and unloading, oil storage and other businesses;

[0073] In actual evaluation, the network nodes are determined: three container yards of the port are set as nodes A1, A2 and A3 respectively; two bulk cargo wharfs are set as nodes B1 and B2; four oil storage tank areas are set as nodes C1, C2, C3 and C4; a key position of the main channel of the port is set as node D1; according to actual transportation channels, pipelines and business processes, there is a commonly used cargo transportation channel from the container yard A1 to the bulk cargo wharf B1, so a connection is established between A1 and B1, the oil storage tank area C1 is connected with the oil loading and unloading wharf in the port through a pipeline, and a connection is established between C1 and the corresponding wharf, and finally the network topology structure of the port is constructed.

[0074] S200 analyzes historical accident cases of the network topology structure, obtains a certain risk factor set, and performs edge calculation on the network topology structure to obtain an uncertain risk factor set;

[0075] It needs to be explained that the uncertain risk factor set is those risk factors that are not found in historical accident cases or are difficult to predict in advance through edge computing. These factors may be caused by new technology applications, external environmental changes, or other unknown factors, such as sudden equipment failure, abnormal fluctuations in network traffic, etc. There may be no similar cases in historical accidents, but they can be detected at the moment they occur through edge computing. It reflects the risk level in the port network topology structure due to uncertain factors, and together with the first risk indicator, the second risk indicator provides more comprehensive information for evaluating the risk status of the port network topology structure.

[0076] In actual evaluation, 15 representative accidents were found by sorting out the accidents in the past 5 years, including container handling equipment failure causing goods to fall, dust explosion during bulk cargo loading and unloading, oil leakage and other accidents; risk factors were extracted, including: container stacking too high, dust concentration monitoring of bulk cargo not in place, oil storage tank aging, inaccurate water depth measurement of the channel; taking an oil leakage accident as an example, collect the network topology data related to the accident, determine the node information as the accident occurred in the oil storage tank area C2, the connection relationship involves the pipeline connection with the surrounding oil loading and unloading wharf and other storage tank areas, the flow data of the edge shows that the oil flow in the pipeline is abnormal when the leakage occurs, at the same time, check the safety equipment status of the C2 area at the time of the accident, the operation records of the staff, etc.; for the oil leakage accident, the key factors are determined as the aging degree of the storage tank, the reliability of the safety equipment, and the operation standardization of the staff; after calculation, the correlation coefficient between the aging degree of the storage tank and the degree of accident loss is 0.024; the correlation coefficient of safety equipment reliability is 0.021, and the correlation coefficient of staff operation standardization is 0.026; the aging degree of the storage tank, the reliability of the safety equipment, and the operation standardization of the staff are obtained as the determined risk factor set;

[0077] In the actual evaluation, edge computing is performed on each network node, at the container yard A1, the calculated state variables include the container stacking height, the handling equipment running time, the traffic flow around the yard, at the bulk cargo terminal B1, the state variables include the bulk cargo stacking height, the handling equipment maintenance state, at the oil product storage tank area C3, the state variables include the storage tank pressure value, the temperature sensor reading, any combination of the state variables between the network nodes is performed, the Pearson correlation coefficient is calculated, the factors corresponding to the state variables greater than the preset threshold (0.7) in the correlation coefficient matrix are obtained, the factors repeated in the determined risk factor set are removed, it is found that the temperature fluctuation anomaly (corresponding to the temperature sensor reading related factor) of the oil product storage tank area meets the condition in the correlation coefficient matrix and is not repeated in the determined risk factor set, and the temperature fluctuation anomaly is taken as a factor in the uncertain risk factor set, and finally the uncertain risk factor set of the port is determined to include the temperature fluctuation anomaly of the oil product storage tank area, the liquid level change rate anomaly, the valve sealing property, the loading and unloading efficiency of the handling equipment and the like.

[0078] S300 constructs an accident tree based on the determined risk factor set and the uncertain risk factor set, and quantitatively evaluates the risk of the network topology based on the accident tree.

[0079] It should be explained that by using the logical structure of the accident tree and the occurrence probability of each event, the probability of the top event (accident occurrence) can be calculated by a probability calculation method (such as Boolean algebra method, minimal cut set method, etc.), and this probability reflects the possibility of accident occurrence under the current risk factor combination; according to the probability of the top event and the contribution degree of each basic event and intermediate event to the top event, the risk of each network node and connection can be quantitatively evaluated, which can help us to determine which nodes and connections are high-risk areas and need to be focused on and take risk control measures;

[0080] It should be understood that if the top event is composed of a variety of complex event combinations, including AND gate and OR gate, then a recursive expression is needed until all the child events participating in the calculation are decomposed into basic events.

[0081] In actual evaluation, the risk of the medium-scale comprehensive port is taken as the top event, the aging degree of the storage tank and the operation specification of the staff in the risk factor set and the abnormal temperature fluctuation and the abnormal liquid level change rate in the uncertain risk factor set are taken as basic events, wherein the logical relationship construction includes: if the storage tank is seriously aged and the staff's operation is not standardized, oil leakage and other major accidents occur, and the two basic events are connected by an AND gate; for the uncertain risk factors such as abnormal temperature fluctuation and abnormal liquid level change rate, if they occur alone or in combination with the determined risk factors, the probability of accident occurrence will increase, and an OR gate is used according to the actual situation, wherein the abnormal temperature fluctuation and the abnormal liquid level change rate, and the high aging degree of the storage tank, greatly increase the risk of oil leakage, and the three basic events are connected by an AND gate, and the fault tree is gradually constructed;

[0082] In actual evaluation, the risk probability of the risk factor set is determined by using the Bayesian network calculation, the risk probability of the uncertain risk factor set is determined by using the fuzzy logic calculation, the risk probability of the network nodes and the top event is calculated by the fault tree, and the risk probability of each network node and the top event is taken as the quantitative evaluation result, specifically, the probability of serious aging of the storage tank is 0.22, the probability of non-standard operation of the staff is 0.15, the risk probability of abnormal temperature fluctuation is 0.18, the risk probability of abnormal liquid level change rate is 0.16, the risk probability of the oil storage tank area node is 0.28, and the risk probability of the top event (port risk) is 0.35.

[0083] S400 constructs a risk evolution model, extracts an accident causation chain from the quantitative evaluation result through the risk evolution model, and calculates a risk influence range according to the accident causation chain.

[0084] It should be explained that by constructing the risk evolution model, the change of the risk with time and various factors can be simulated, the development trend of the potential risk can be predicted in advance, the extraction of the accident causation chain helps to determine the root cause and path of the accident, provides a direction for taking targeted prevention measures, and the calculation of the risk influence range enables the manager to clearly understand the area and degree that may be affected by the accident, so that resources can be reasonably allocated for emergency response and risk control.

[0085] In actual evaluation, a Markov chain is used to build a risk evolution model, the state set is {safe, low risk, medium risk, high risk}, the probability distribution of the risk factor set is {0.45, 0.35, 0.15, 0.05}, the uncertain risk factor set is {0.42, 0.32, 0.2, 0.06}, data is obtained from historical cases to divide a training set to train a model, the node risk probability is input into the model, the node probability of the oil product storage tank area is 0.25, the node probability of the container yard is 0.3, an accident causation chain is obtained, the risk influence range is calculated, the node probability of the risk source is 0.35, the attenuation coefficient is 0.8, the minimum acceptable level is 0.1, and the radius is 0.65, wherein the influence range radius of a certain node is 0.26.

[0086] S500 visualizes the network topology structure, adds the accident causation chain and the risk influence range to the visualization result after color mapping, and obtains a three-dimensional topology view.

[0087] In actual evaluation, the network node risk probability is classified as low (0-0.2), medium (0.2-0.4) and high (0.4 or more) risk, a three-dimensional view is created, the low-risk node is light blue and shallow, the medium-risk node is yellow and moderate, and the high-risk node is red and prominent, the accident causation chain is displayed by a blue luminous line, the risk influence range is represented by a semi-transparent circle with different radii, interactive controls are added, the user can click a node to view the risk probability, select a causation chain node to focus on and view the related area risk, and the port risk condition is intuitively presented.

[0088] In the embodiment, the method for determining the risk factor set by analyzing historical accident cases of the network topology structure includes the following steps.

[0089] An accident record of a historical accident case is obtained, network topology data related to the accident is collected, including node information, connection relationship and edge flow data, and a key factor of the accident is identified;

[0090] The correlation coefficient of the key factor and the accident loss degree is calculated, and the expression is as follows:

[0091]

[0092] wherein m is the number of historical accident cases, F ik represents the value of the key factor F i in the kth case, is the mean value of the key factor F i , L k is the accident loss degree of the kth case, is the mean value of the accident loss degree of all cases;

[0093] The key factor with a correlation coefficient greater than 0.01 with the accident loss degree is determined as a risk factor, and a risk factor set is obtained.

[0094] In the embodiment, the method for performing edge computing on the network topology to obtain the uncertain risk factor set comprises:

[0095] The edge computing is performed on each network node to obtain a state variable of each network node.

[0096] The state variables between the network nodes are combined in an arbitrary manner, and a correlation coefficient matrix is obtained by calculating a Pearson correlation coefficient.

[0097] The factors corresponding to the state variables greater than a preset threshold in the correlation coefficient matrix are obtained, the factors repeated in the determined risk factor set are removed, and an uncertain risk factor set is obtained.

[0098] In the embodiment, the method for constructing the accident tree based on the determined risk factor set and the uncertain risk factor set comprises: taking the existence of a risk in a port area as a top event, taking each factor in the determined risk factor set and the uncertain risk factor set as a basic event, connecting the basic events by using logical gates of AND gates and OR gates according to a logical relationship, obtaining an intermediate event, and forming an accident tree.

[0099] In the embodiment, the method for quantitatively evaluating the risk of the network topology based on the accident tree comprises:

[0100] S21 calculates the risk probability of the corresponding basic event in the determined risk factor set by using the Bayesian network, and the calculation specifically comprises:

[0101] The basic event in the accident tree is taken as a node of the Bayesian network, and the Bayesian network is constructed.

[0102] The Bayesian network is corrected by using a hyperparameter method to establish a kernel function, and an expression of the kernel function is:

[0103]

[0104] wherein α is a global scaling factor, x and y are respectively a state variable and a characteristic variable of an event in the Bayesian network, σ is a bandwidth of a Gaussian radial basis function, T represents a transposition operation, M is a normalization factor, d is a degree of a polynomial kernel, β and b are respectively a slope and a bias of a Sigmoid kernel, and p is a nonlinear index of the kernel function.

[0105] The probability distribution in the Bayesian network is updated by using an expectation maximization, and an expression of the update is:

[0106]

[0107] where θ is a set of parameters in the Bayesian network, θ t and θ (t+1) are the estimates of θ in the tth iteration and the t+1th iteration, respectively, arg maxθ is a maximization operator, is an expectation operator, P(X hidden |D, θ (t) ) is a posterior probability distribution of hidden nodes X (t) given the observed data D and the parameter estimate θ hidden , is a log-likelihood function of the joint probability distribution, n is the number of nodes in the Bayesian network, P(X i |pa(X i ); θ) represents the conditional probability of an estimated node X i given its parent nodes pa(X i ) depending on the parameter estimate θ;

[0108] taking the probability distribution in the Bayesian network as the risk probability of the basic event;

[0109] S22 calculates the risk probability of the basic event corresponding to the non-deterministic factor set by using fuzzy logic, specifically:

[0110] where the expression of the membership function is:

[0111]

[0112] where λ i is the ith non-deterministic risk factor, c i is the central value of x i , is the standard deviation of λ i ;

[0113] The membership function is de-fuzzified by the centroid method to obtain the risk probability of the basic event;

[0114] S23 calculates the risk probability of the network nodes and the top event according to the risk probability of the basic event by the fault tree, and the expression is:

[0115]

[0116] where E is a basic event or a complex event composed of basic events, P(ξ i ) is the risk probability of the basic event ξ i ;

[0117] S24 takes the risk probability of each network node and the top event as the quantitative evaluation result.

[0118] In the embodiment, the method for constructing the risk evolution model and extracting the accident causation chain from the quantitative evaluation result comprises:

[0119] The risk evolution model is constructed using a Markov chain algorithm, wherein the expression of the state transition function is:

[0120]

[0121] wherein P(X n+1 j|X n i) is the probability of transitioning to the target state j at time step n+1 under the condition of being in state i at time step n, S is the set of all states, U is the set of certain risk factors, p u (n) is the probability distribution of the certain risk factor u at time step n, V is the set of uncertain risk factors, q v (n) is the probability distribution of the uncertain risk factor v at time step n, p kj is the transition probability from the intermediate state k to the target state j;

[0122] The historical accident causation chain data is obtained from historical accident cases, and is divided into a training set, a test set and a validation set in a ratio of 8:1:1 to train the risk evolution model, wherein the risk probability of each network node is taken as an input to obtain the accident causation chain.

[0123] The causation intensity between each network node and the risk source node is calculated, and the expression is:

[0124]

[0125] wherein ω1, ω2, ω3, ω4 and ω5 are weight coefficients of the degree centrality, the betweenness centrality, the closeness centrality, the eigenvector centrality and the clustering coefficient, respectively, D(v), B(v), C(v), E(v) and T(v) are the measurement values of the degree centrality, the betweenness centrality, the closeness centrality, the eigenvector centrality and the clustering coefficient, respectively, deg + (v) and deg - (v) are the in-degree and out-degree of node v, respectively, N is the total number of nodes in the topological network structure, σ st (v) and σ st are the number of shortest paths passing through node v and the total number of shortest paths from node s to node t, respectively, d(v, t) is the shortest path length from node v to node t, λ is a normalization constant, A vt is the connection weight between node v and node t, E(t) is the eigenvector centrality score of node t, T v is the actual number of edges between the neighbor nodes of node v, and deg(v) is the degree of node v.

[0126] In the embodiment, the method for calculating the risk influence range according to the accident causation chain comprises:

[0127] The influence range radius of the risk source is calculated by quantifying the evaluation result, and the expression is:

[0128]

[0129] Wherein, N i is the risk source node, P(T|N i ) is the risk probability of node N i , η is the attenuation coefficient of risk propagation, and τ is the minimum acceptable level of risk influence.

[0130] The influence range radius of the remaining nodes on the causation chain is calculated based on the causation strength between each network node and the risk source node, and the expression is:

[0131] R i = ψ i × R0

[0132] Wherein, R i is the influence range of the i th network node, and ψ i is the causation strength between the i th network node and the risk source node.

[0133] In the embodiment, the method for visualizing the network topology, adding the accident causation chain and the risk influence range to the visualization result after color mapping comprises:

[0134] The risk probability of each network node in the quantitative evaluation result is taken as the attribute of the node, the risk probability of the top event is taken as the overall attribute, and the risk probability of each network node and the top event is graded.

[0135] A three-dimensional view of the port network topology is created, different colors and depths are used to represent the risk probability of each network node and the top event, the accident causation chain and the risk influence range are color-mapped into the three-dimensional view, and interactive controls are added to allow users to filter and view data according to the risk probability level or information on the accident causation chain.

[0136]

[0137] Figure 2 is a structural schematic diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 2 ​At the hardware level, the electronic device comprises a processor, and optionally further comprises an internal bus, a network interface, a memory. The memory can include a memory such as a random-access memory (RAM), and can also include a non-volatile memory such as at least one disk memory. Of course, the electronic device can also include other hardware required by the business.

[0138] The processor, the network interface and the memory can be connected to each other through the internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0139] The memory is used to store programs. Specifically, the program can include program code, and the program code includes computer operation instructions. The memory can include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0140] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs, and forms a port accident risk visualization analysis device based on network topology at the logical level. The processor executes the program stored in the memory, and is specifically used to execute any one of the preceding port accident risk visualization analysis methods based on network topology.

[0141] The present application can be applied to a processor or implemented by the processor. The processor can be an integrated circuit chip with processing capability. In the implementation process, each step of the above method can be completed by integrated logic circuits in hardware or instructions in software form in the processor. The processor described above can be a general processor, including a central processing unit, a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in 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 in the memory is read by the processor, and the hardware thereof is combined to complete the steps of the above method.

[0142] The embodiments of the present application also propose a computer readable storage medium, which stores one or more programs, the one or more programs including instructions, which when executed by an electronic device including a plurality of application programs, execute any one of the above-mentioned port accident risk visualization analysis methods based on network topology.

[0143] The above is only an example and description of the structure of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace, as long as the modifications or supplements do not deviate from the structure of the present application or exceed the scope defined by the present claims, and should belong to the protection scope of the present application.

Claims

1. A port area accident risk visualization analysis method based on network topology, characterized in that, Includes the following steps: Key facilities and important areas in the port area are designated as network nodes. The connection relationships between network nodes are determined based on actual physical connections and business processes to obtain the network topology of the port area. Analyze historical incident cases of network topology to obtain a set of identifiable risk factors; Edge computing is performed on the network topology to obtain a set of uncertain risk factors; Based on the set of determined risk factors and the set of uncertain risk factors, an accident tree is constructed, and based on the accident tree, the risk of the network topology is quantitatively assessed, including: S21 uses a Bayesian network to calculate the risk probability of basic events corresponding to a given set of risk factors: S22 uses fuzzy logic to calculate the risk probability of basic events corresponding to a set of uncertain factors; S23 calculates the risk probabilities of network nodes and top events using a fault tree based on the risk probabilities of basic events; S24 uses the risk probability of each network node and top event as a quantitative assessment result; A risk evolution model is constructed, and the accident causal chain is extracted from the quantitative assessment results through the risk evolution model. The scope of risk impact is calculated based on the accident causal chain. The network topology is visualized by color mapping the accident causation chain and the risk impact range, which is then added to the visualization result to obtain a three-dimensional topology view. The method for performing edge computing on the network topology to obtain a set of uncertain risk factors includes: Edge computing is performed on each network node to obtain the state variables of each network node; By arbitrarily combining the state variables between network nodes, the correlation coefficient matrix is ​​obtained by calculating the Pearson correlation coefficient. Obtain the factors corresponding to state variables with correlation coefficients greater than a preset threshold in the correlation coefficient matrix, remove factors that overlap with the set of known risk factors, and obtain the set of uncertain risk factors; The method for constructing a risk evolution model and extracting accident causal chains from quantitative assessment results using the risk evolution model includes: A risk evolution model is constructed using the Markov chain algorithm, where the expression for the state transition function is: in, In time step In state Under the conditions of time step Transition to the target state The probability, For the set of all states, To determine the set of risk factors, To identify risk factors At time step The probability distribution, It is a set of uncertain risk factors. Uncertain risk factors At time step The probability distribution, To start from the intermediate state Transition to the target state The transition probability; Historical accident causation chain data is obtained from historical accident cases and divided into training set, test set and validation set in a ratio of 8:1:1 to train the risk evolution model. The risk probability of each network node is used as input to obtain the accident causation chain. The causation strength between each network node and the risk source node is calculated using the following expression: in, , , , and These are the weight coefficients for degree centrality, betweenness centrality, proximity centrality, eigenvector centrality, and clustering coefficient, respectively. , , , and These are measures of degree centrality, betweenness centrality, proximity centrality, eigenvector centrality, and clustering coefficient, respectively. and Represented as nodes in-degree and out-degree, This represents the total number of nodes in the network topology. and respectively through nodes Shortest path number and nodes To the node The total number of shortest paths, For nodes To the node The shortest path length, The normalization constant is For nodes and nodes Connection weights between them For nodes Feature vector center score, For nodes The actual number of edges between neighboring nodes. For nodes The degree; The method for calculating the scope of risk impact based on the accident causation chain includes: Using the risk source node as the center, the radius of the risk source's influence range is calculated based on the quantitative assessment results. The expression is: in, As a risk source node, For nodes The probability of risk, This represents the attenuation coefficient of risk propagation. This represents the minimum acceptable level of risk impact. The radius of influence of other nodes in the causation chain is calculated based on the causation strength between each network node and the risk source node, expressed as follows: in, For the first The scope of influence of each network node For the first The causal strength of each network node and the risk source node.

2. The port area accident risk visualization analysis method based on network topology according to claim 1, characterized in that, The method for analyzing historical incident cases of network topology to obtain a set of risk factors includes: Obtain accident records from historical accident cases, collect network topology data related to the accidents, including node information, connection relationships and edge flow data, and identify the key factors that caused the accidents; The correlation coefficient between key factors and the degree of accident loss is calculated using the following expression: in, The number of historical accident cases, Represented as the first Key factors in each case The value, Key factors The mean, For the first The extent of the accident damage in each case This represents the average extent of damage from all accidents. Key factors with a correlation coefficient greater than 0.01 with the degree of accident loss are identified as risk factors, thus obtaining a set of risk factors.

3. The port area accident risk visualization analysis method based on network topology according to claim 1, characterized in that, The method for constructing a fault tree based on the set of determined risk factors and the set of uncertain risk factors includes: taking the existence of risk in the port area as the top event, taking each factor in the set of determined risk factors and the set of uncertain risk factors as basic events, connecting the basic events using AND gates and OR gates according to logical relationships, obtaining intermediate events, and forming a fault tree.

4. The port area accident risk visualization analysis method based on network topology according to claim 1, characterized in that, The method for quantitatively assessing the risk of network topology based on the fault tree includes: S21 uses a Bayesian network to calculate the risk probability of the basic events corresponding to a given set of risk factors, specifically: Construct a Bayesian network using the basic events in the fault tree as nodes; The Bayesian network is modified by establishing a kernel function using the hyperparameter method. The expression for the kernel function is as follows: in, This is the global scaling factor. and These represent the state variables and feature variables of the event in the Bayesian network, respectively. Let be the bandwidth of the Gaussian radial basis function. This indicates that a transpose operation is being performed. As the normalization factor, Let be the degree of the polynomial kernel. and These represent the slope and bias of the sigmoid kernel, respectively. is the nonlinear exponent of the kernel function; By maximizing the update of the probability distribution in the Bayesian network, the update expression is: in, The parameter set in a Bayesian network. and They were respectively in the second The second iteration and the first In the next iteration, The estimate, To maximize the operator, For expectation operator, In order to provide the observation data and parameter estimation Under the condition of hidden nodes The posterior probability distribution, Let be the log-likelihood function of the joint probability distribution. The number of nodes in the Bayesian network. Represented as dependent on parameters Estimated nodes Given its parent node The conditional probability; The probability distribution in a Bayesian network is used as the risk probability of basic events; S22 uses fuzzy logic to calculate the risk probability of basic events corresponding to a set of uncertain factors, specifically: The expression for the membership function is: in, For the first One uncertain risk factor, for The central value, for Standard deviation; By defuzzifying the membership function using the centroid method, the risk probability of the basic event can be obtained. S23 calculates the risk probabilities of network nodes and the top event using the fault tree based on the risk probabilities of basic events, with the following expression: in, These are basic events or complex events composed of basic events. For basic events The probability of risk; S24 uses the risk probability of each network node and top event as a quantitative assessment result.

5. The port area accident risk visualization analysis method based on network topology according to claim 1, characterized in that, The method for visualizing the network topology and adding the accident causal chain and the risk impact range to the visualization result after color mapping includes: The risk probability of each network node in the quantitative evaluation results is used as the node's attribute, and the risk probability of the top event is used as the overall attribute. The risk probabilities of each network node and the top event are then classified. Create a 3D view of the port area network topology, using different colors and depths to represent the risk probability of each network node and top event, and map the accident causal chain and the scope of risk impact to the 3D view using color; Add interactive controls to allow users to filter and view data based on risk probability levels or information in the accident causation chain.

6. An electronic device, comprising: processor; as well as A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 5.

7. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the method of any one of claims 1 to 5.

Images