Method and device for determining capacity risk based on dynamic bayesian network

Abstract

This disclosure relates to a method and apparatus for determining transport capacity risk based on dynamic Bayesian networks. The method identifies a rail transit network comprising at least one line, where each line includes at least two stations and operating sections between adjacent stations. The method determines the station risk attribute of each station within a target time interval, and the section risk attribute of each operating section. A dynamic Bayesian network is constructed based on the rail transit network, the risk attributes of each station, and the risk attributes of each section. A Gaussian mixture model is established based on the joint distribution of the dynamic Bayesian network to obtain network parameters. Furthermore, the dynamic Bayesian network is solved based on the risk attribute values ​​of each station and each section within the target time interval to obtain the network transport capacity risk. This disclosure improves the interpretability of the transport capacity risk prediction process by constructing a Bayesian network and improves the accuracy of the transport capacity risk prediction results by establishing a Gaussian mixture model.

Description

Technical Field

[0001] This disclosure relates to the field of rail transit, and in particular to a method and apparatus for determining transport capacity risk based on dynamic Bayesian networks. Background Technology

[0002] Rail transit plays an increasingly important role in modern urban transportation due to its advantages such as large capacity, good punctuality, high safety, environmental friendliness, and low cost, becoming the backbone and important support of modern urban transportation. Although rail transit offers higher safety than conventional road transportation, its large network scale, heavy transport load, and close coupling between lines mean that any malfunction or safety incident can have a significant impact on urban traffic. Therefore, effectively assessing and predicting the overall risk of the rail transit network and mitigating this overall risk, as well as its impact on the network's transport capacity, has always been a key research issue in the field of rail transit. Currently, related technologies lack interpretability in predicting rail transit network risks, and the accuracy of the prediction results is relatively low. Summary of the Invention

[0003] In view of this, this disclosure proposes a method and apparatus for determining transport capacity risk based on dynamic Bayesian networks, aiming to improve the interpretability of the prediction process and the accuracy of the prediction results.

[0004] According to a first aspect of this disclosure, a method for determining capacity risk based on dynamic Bayesian networks is provided, the method comprising:

[0005] Define a rail transit network comprising at least one line, each line comprising at least two stations, with an operating section between each pair of stations;

[0006] Determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval;

[0007] A dynamic Bayesian network for predicting network capacity risk is constructed based on the rail transit network, the risk attributes of each station, and the risk attributes of each section.

[0008] A Gaussian mixture model is established based on the joint distribution of the dynamic Bayesian network to determine the network parameters of the dynamic Bayesian network.

[0009] Based on the risk attribute values ​​of each station and each interval within the target time interval, the dynamic Bayesian network is solved to obtain the network capacity risk of the rail transit network.

[0010] In one possible implementation, the step of constructing a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each interval includes:

[0011] Determine the risk correspondence between the risk attribute of each station and the station capacity risk corresponding to each station in the rail transit network;

[0012] Determine the risk correspondence between each of the interval risk attributes and the interval capacity risk corresponding to each of the operating intervals in the rail transit network;

[0013] Determine the risk correspondence between the capacity risk of each station, the capacity risk of each section, and the line capacity risk of each line in the rail transit network;

[0014] Determine the risk correspondence between the capacity risk of each line and the network capacity risk of the corresponding rail transit network;

[0015] The station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are used as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0016] In one possible implementation, the method further includes:

[0017] Determine the station-specific risk for each of the aforementioned stations, and the section-specific risk for each of the aforementioned operating sections;

[0018] The step of constructing a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section further includes:

[0019] Determine the risk correspondence between each station's single-point risk and the station's risk attribute, as well as the risk correspondence between each interval's single-point risk and the interval's risk attribute;

[0020] The step of using the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the network capacity risk as network nodes, and then sequentially connecting each network node according to the risk correspondence to obtain a dynamic Bayesian network, includes:

[0021] The station single-point risk, the interval single-point risk, the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are taken as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0022] In one possible implementation, the step of establishing a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network to determine the network parameters of the dynamic Bayesian network includes:

[0023] Determine the historical station risk attributes, section risk attributes, station capacity risk, section capacity risk, line capacity risk, and network capacity risk of the rail transit network.

[0024] Based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, the Gaussian mixture model is solved to obtain the node parameters of each network node in the dynamic Bayesian network.

[0025] In one possible implementation, the step of solving the Gaussian mixture model based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk to obtain the node parameters of each network node in the dynamic Bayesian network includes:

[0026] The Gaussian mixture model is solved using the expectation-maximization algorithm based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, thereby obtaining the layer parameters of each node layer in the dynamic Bayesian network.

[0027] The node parameters of the network nodes included in each node layer are calculated based on the layer parameters of each node layer.

[0028] In one possible implementation, the layer parameters include first layer parameters, second layer parameters, and third layer parameters, and the node parameters include first node parameters, second node parameters, and third node parameters;

[0029] The first node parameters are based on The calculation yields β. l Let α be the parameter of the first node. l x is the first-level parameter of the node layer where the network node is located. S μ is the risk value of the underlying node corresponding to the network node. lS For the second node parameter of the underlying node corresponding to the network node, ∑ lSSM is the autocovariance of the third node parameter of the underlying node corresponding to the network node, and M is the number of first-level parameters of the node layer where the underlying node of the network node is located.

[0030] The second node parameter is based on the formula The calculation yields μ. lR|S μ is the conditional probability of the second node parameter corresponding to the network node. lR For the second node parameters corresponding to the network node, ∑ lRS The covariance is the third node parameter of the network node and the third node parameter of the corresponding underlying node of the network node.

[0031] The third node parameter is based on the formula. The calculation yields ∑ lR|S Let ∑ be the conditional probability of the third node parameter corresponding to the network node. lRR Let ∑ be the autocovariance of the third node parameters of the network node. lSR The covariance is the third node parameter of the underlying node corresponding to the network node and the third node parameter of the network node.

[0032] In one possible implementation, the step of solving the dynamic Bayesian network based on the risk attribute values ​​of each station and each interval within the target time interval to obtain the network capacity risk of the rail transit network includes:

[0033] Using the risk attribute value of each station and the risk attribute value of each interval within the target time interval as input, the mean distribution value corresponding to each network node in the dynamic Bayesian network is solved layer by layer.

[0034] The mean distribution value of the top-level network node of the Bayesian network is used as the network capacity risk of the rail transit network.

[0035] According to a second aspect of this disclosure, a capacity risk determination device based on dynamic Bayesian networks is provided, the device comprising:

[0036] A road network determination module is used to determine a rail transit road network including at least one line, each line including at least two stations, and an operating section between every two stations;

[0037] The attribute determination module is used to determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval.

[0038] A network construction module is used to construct a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section.

[0039] The model building module is used to establish a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network in order to determine the network parameters of the dynamic Bayesian network.

[0040] The risk prediction module is used to solve the dynamic Bayesian network based on the risk attribute value of each station and the risk attribute value of each interval within the target time interval, so as to obtain the network capacity risk of the rail transit network.

[0041] In one possible implementation, the network building module is further configured to:

[0042] Determine the risk correspondence between the risk attribute of each station and the station capacity risk corresponding to each station in the rail transit network;

[0043] Determine the risk correspondence between each of the interval risk attributes and the interval capacity risk corresponding to each of the operating intervals in the rail transit network;

[0044] Determine the risk correspondence between the capacity risk of each station, the capacity risk of each section, and the line capacity risk of each line in the rail transit network;

[0045] Determine the risk correspondence between the capacity risk of each line and the network capacity risk of the corresponding rail transit network;

[0046] The station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are used as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0047] In one possible implementation, the device further includes:

[0048] The single-point risk determination module is used to determine the single-point risk of each of the stations and the single-point risk of each of the operating sections.

[0049] The network construction module is further used for:

[0050] Determine the risk correspondence between each station's single-point risk and the station's risk attribute, as well as the risk correspondence between each interval's single-point risk and the interval's risk attribute;

[0051] The network construction module is further used for:

[0052] The station single-point risk, the interval single-point risk, the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are taken as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0053] In one possible implementation, the model building module is further configured to:

[0054] Determine the historical station risk attributes, section risk attributes, station capacity risk, section capacity risk, line capacity risk, and network capacity risk of the rail transit network.

[0055] Based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, the Gaussian mixture model is solved to obtain the node parameters of each network node in the dynamic Bayesian network.

[0056] In one possible implementation, the model building module is further configured to:

[0057] The Gaussian mixture model is solved using the expectation-maximization algorithm based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, thereby obtaining the layer parameters of each node layer in the dynamic Bayesian network.

[0058] The node parameters of the network nodes included in each node layer are calculated based on the layer parameters of each node layer.

[0059] In one possible implementation, the layer parameters include first layer parameters, second layer parameters, and third layer parameters, and the node parameters include first node parameters, second node parameters, and third node parameters;

[0060] The first node parameters are based on The calculation yields β. l Let α be the parameter of the first node. l x is the first-level parameter of the node layer where the network node is located. S μ is the risk value of the underlying node corresponding to the network node. lS For the second node parameter of the underlying node corresponding to the network node, ∑ lSS M is the autocovariance of the third node parameter of the underlying node corresponding to the network node, and M is the number of first-level parameters of the node layer where the underlying node of the network node is located.

[0061] The second node parameter is based on the formula The calculation yields μ. lR|S μ is the conditional probability of the second node parameter corresponding to the network node. lR For the second node parameters corresponding to the network node, ∑ lRS The covariance is the third node parameter of the network node and the third node parameter of the corresponding underlying node of the network node.

[0062] The third node parameter is based on the formula. The calculation yields ∑ lR|S Let ∑ be the conditional probability of the third node parameter corresponding to the network node. lRR Let ∑ be the autocovariance of the third node parameters of the network node. lSR The covariance is the third node parameter of the underlying node corresponding to the network node and the third node parameter of the network node.

[0063] In one possible implementation, the risk prediction module is further configured to:

[0064] Using the risk attribute value of each station and the risk attribute value of each interval within the target time interval as input, the mean distribution value corresponding to each network node in the dynamic Bayesian network is solved layer by layer.

[0065] The mean distribution value of the top-level network node of the Bayesian network is used as the network capacity risk of the rail transit network.

[0066] According to a third aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing instructions stored in the memory.

[0067] According to a fourth aspect of this disclosure, a non-volatile computer-readable storage medium is provided that stores computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the above-described method.

[0068] According to a fifth aspect of this disclosure, a computer program product is provided, including computer-readable code or a non-volatile computer-readable storage medium carrying the computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0069] In this embodiment, a rail transit network comprising at least one line is identified, wherein each line includes at least two stations and operating sections between adjacent stations. The station risk attribute of each station within a target time interval and the section risk attribute of each operating section are determined. A dynamic Bayesian network is constructed based on the rail transit network, the risk attributes of each station, and the risk attributes of each section, and a Gaussian mixture model is established based on the joint distribution of the dynamic Bayesian network. Furthermore, the Gaussian mixture model is solved based on the risk attribute values ​​of each station and each section within the target time interval to obtain the network capacity risk of the rail transit network. This disclosure improves the interpretability of the capacity risk prediction process by constructing a Bayesian network and improves the accuracy of the capacity risk prediction results by establishing a Gaussian mixture model.

[0070] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0071] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0072] Figure 1 A flowchart illustrating a capacity risk determination method based on a dynamic Bayesian network according to an embodiment of the present disclosure is shown.

[0073] Figure 2 This diagram illustrates the structure of a dynamic Bayesian network according to an embodiment of the present disclosure.

[0074] Figure 3 This diagram illustrates a partial structure of a dynamic Bayesian network according to an embodiment of the present disclosure.

[0075] Figure 4 A schematic diagram of a capacity risk determination device based on a dynamic Bayesian network according to an embodiment of the present disclosure is shown.

[0076] Figure 5 A schematic diagram of an electronic device according to an embodiment of the present disclosure is shown;

[0077] Figure 6 A schematic diagram of another electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0078] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0079] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0080] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0081] The capacity risk determination method based on dynamic Bayesian networks in this disclosure can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal, such as user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, or wearable device. The server can be a single server or a server cluster consisting of multiple servers. Any electronic device can implement the capacity risk determination method based on dynamic Bayesian networks in this disclosure by having its processor call computer-readable instructions stored in its memory.

[0082] Figure 1 A flowchart illustrating a capacity risk determination method based on a dynamic Bayesian network according to an embodiment of this disclosure is shown. Figure 1 As shown, the capacity risk determination method based on dynamic Bayesian networks in this disclosure may include the following steps S10-S50.

[0083] Step S10: Determine the rail transit network that includes at least one line.

[0084] In one possible implementation, the rail transit network for which capacity risk needs to be predicted is determined using electronic equipment. The rail transit network can be a rail network or a subway network, where rail vehicles operate, including at least one line, each line including at least two stations, with corresponding operating sections between every two adjacent stations. For example, in the case of a subway network, it may include two lines: Metro Line 1 and Metro Line 2. Further, Metro Line 1 may include, in sequence, stations 1, 2, 3, and 4, with corresponding operating sections between stations 1 and 2, 2 and 3, and 3 and 4. Metro Line 2 may include, in sequence, stations 5, 3, and 6, with corresponding operating sections between stations 5 and 3, and between 3 and 6.

[0085] Step S20: Determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval.

[0086] In one possible implementation, electronic equipment determines the station risk attribute of each station and the section risk attribute of each operating section within a target time interval prior to the time when capacity risk needs to be predicted. The target time interval can be determined using a time window of preset length. For example, if the electronic equipment performs periodic risk attribute predictions at a frequency of once per hour, and the preset length of the time window is 3, and the risk attribute of the rail transit network at 10:00 AM needs to be predicted, the electronic equipment can determine the station risk attribute of each station and the section risk attribute of each operating section corresponding to the three time points of 10:00 AM, 9:00 AM, and 8:00 AM, respectively.

[0087] Optionally, the station risk attributes include at least one attribute information affecting the station's capacity risk, such as station passenger flow saturation and station passenger evacuation time. Station passenger flow saturation can be obtained by calculating the ratio of station passenger flow to station passenger capacity. Station passenger evacuation time can be the time it takes for passengers waiting in the station's waiting area to completely leave the waiting area. The section risk attributes include at least one attribute information affecting the section's capacity risk, such as section passenger flow saturation, which can be obtained by calculating the ratio of section passenger flow to passenger capacity.

[0088] Step S30: Construct a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section.

[0089] In one possible implementation, after determining the rail transit network, including at least one station risk attribute for each station within a target time interval, and at least one section risk attribute for each operating section within the target time interval, the electronic device constructs a dynamic Bayesian network based on the determined information to predict network capacity risk. The dynamic Bayesian network can be viewed as a directed acyclic graph (DAG), meaning the process of the electronic device constructing the dynamic Bayesian network can be described as drawing a DAG using at least one of the aforementioned pieces of information as nodes.

[0090] Optionally, when constructing a dynamic Bayesian network by drawing a directed acyclic graph, the electronic device needs to first determine the nodes included in the graph and the relationships between different nodes. Specifically, when constructing a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section, the following can be determined: first, the risk correspondence between each station's risk attribute and the station's capacity risk in the rail transit network; second, the risk correspondence between each section's risk attribute and the section's capacity risk in the rail transit network; third, the risk correspondence between each station's capacity risk, each section's capacity risk, and the line's capacity risk in the rail transit network; and fourth, the risk correspondence between each line's capacity risk and the network's capacity risk. Then, the station risk attribute, section risk attribute, station capacity risk, section capacity risk, line capacity risk, and network capacity risk are used as network nodes, and each network node is connected in an ordered manner according to the risk correspondence to obtain the dynamic Bayesian network.

[0091] Among them, station capacity risk characterizes the probability of a station experiencing a malfunction or accident; section capacity risk characterizes the probability of a malfunction or accident occurring within a corresponding operating section; line capacity risk characterizes the probability of a malfunction or accident occurring within an operating section; and network capacity risk characterizes the probability of a malfunction or accident occurring within the entire rail transit network. Each station capacity risk is determined by its corresponding station risk attribute, thus there is a risk correspondence between station capacity risk and station risk attribute. Each section capacity risk is determined by its corresponding section risk attribute, thus there is a risk correspondence between section capacity risk and section risk attribute. Each line capacity risk is determined by the station capacity risk of the stations included in that line and the section capacity risk of the operating sections included in that line, thus there is a risk correspondence between line capacity risk, station capacity risk, and section capacity risk. Network capacity risk is determined by the line capacity risk of all lines included in the network, thus there is a risk correspondence between network capacity risk and line risk.

[0092] Furthermore, after obtaining different risk correspondences, the electronic equipment treats station single-point risk, section single-point risk, station risk attribute, section risk attribute, station capacity risk, section capacity risk, line capacity risk, and network capacity risk as network nodes. It then connects each of these network nodes in an ordered manner according to the risk correspondences, drawing an ordered undirected graph to obtain a dynamic Bayesian network. The resulting dynamic Bayesian network includes multiple node layers, each containing at least one network node of the same type. Optionally, the different node layers can be ordered according to the types of nodes they include, from bottom to top: station risk attribute and section risk attribute layer, station capacity risk and section capacity risk attribute layer, line capacity risk attribute layer, and network capacity risk attribute layer.

[0093] In one possible implementation, the station risk attribute and section risk attribute in this embodiment can also be affected by the corresponding single-point risk. That is, the electronic device can also determine the single-point risk of each station and the single-point risk of each operating section. Furthermore, during the construction of the dynamic Bayesian network, network nodes corresponding to the single-point risks of stations and sections can be added. In other words, the electronic device can determine the risk correspondence between each single-point risk of a station and its risk attribute, as well as the risk correspondence between each single-point risk of a section and its risk attribute. Then, the single-point risk of a station, the single-point risk of a section, the station risk attribute, the section risk attribute, the station capacity risk, the section capacity risk, the line capacity risk, and the network capacity risk are used as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain the dynamic Bayesian network. Different node layers can be ordered according to the node types they include, from bottom to top: the station risk attribute and section risk attribute layer, the station capacity risk and section capacity risk attribute layer, the line capacity risk attribute layer, and the network capacity risk attribute layer.

[0094] Optionally, station-level single-point risk may include at least one piece of information affecting the station's risk attributes, such as single-point risk information, station risk probability, consequences of loss to station equipment and personnel, traffic impact level of station risk, and duration of station risk. Section-level single-point risk may include at least one piece of information affecting section-level risk attributes, such as single-point risk information, section risk probability, consequences of loss to section equipment and personnel, traffic impact level of section risk, and duration of section risk.

[0095] Figure 2 A schematic diagram of the structure of a dynamic Bayesian network according to an embodiment of the present disclosure is shown. Figure 2As shown, considering the single-point risk impact of stations and operating sections, the constructed dynamic Bayesian network includes five node layers from bottom to top: station risk attribute and section risk attribute layer, station capacity risk and section capacity risk attribute layer, line capacity risk attribute layer, and network capacity risk attribute layer. At least one network node in each node layer is connected to the network node in the previous node layer according to the risk correspondence relationship.

[0096] Figure 3 A schematic diagram of a local structure in a dynamic Bayesian network according to an embodiment of the present disclosure is shown. Figure 3 As shown, when determining the station capacity risk and interval capacity risk at the current moment, the electronic equipment not only considers the station risk attribute and interval risk attribute corresponding to the station or interval at the current moment, but also considers all station risk attributes and interval risk attributes at at least one previous moment, i.e., within the target time interval.

[0097] Step S40: Establish a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network to determine the network parameters of the dynamic Bayesian network.

[0098] In one possible implementation, after constructing a dynamic Bayesian network, the electronic device uses a linear Gaussian model to model the conditional probabilities of the Bayesian network. Gaussian Bayesian networks are among the most classic continuous Bayesian networks, modeling the joint distribution of all node variables as a multivariate normal distribution, which is convenient for calculation and derivation. However, there is a certain gap between the normality assumption of the conditional probability distribution and the distribution of the collected data. Therefore, in order to further improve prediction performance, without changing the network topology, this embodiment of the disclosure enhances the fitting effect on the nonlinear relationships between network nodes of the dynamic Bayesian network by establishing a Gaussian mixture model, modeling the joint distribution to facilitate the calculation of network parameters in the dynamic Bayesian network. The network parameters include at least one node parameter for each network node in the dynamic Bayesian network.

[0099] Optionally, the electronic device can predetermine the network parameters of the dynamic Bayesian network, or determine the network parameters in real time when predicting the network capacity risk at the current moment. Specifically, when determining the network parameters, the electronic device can first determine the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk of the rail transit network. Then, based on these historical attributes, it solves the Gaussian mixture model to obtain the node parameters of each network node in the dynamic Bayesian network. The historical information such as station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk can include historical information corresponding to multiple time points.

[0100] Furthermore, the electronic device can first calculate the layer parameters of each node layer in the Bayesian network, and then further calculate the node parameters of the network nodes based on the layer parameters. That is, the electronic device can use the expectation-maximization algorithm to solve the Gaussian mixture model based on historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk to obtain the layer parameters of each node layer in the dynamic Bayesian network. Then, based on the layer parameters of each node layer, the node parameters of the network nodes included in the node layer are calculated. Among them, the layer parameters of each node layer can also include three parameters: first-layer parameters, second-layer parameters, and third-layer parameters. The node parameters corresponding to each network node also include first-node parameters, second-node parameters, and third-node parameters.

[0101] According to existing technology, a Gaussian mixture model (GMM) is a weighted average sum of finite Gaussian distributions, which can approximate any distribution when the number of components is sufficient. Satisfy ∑ k π k =1, where each Gaussian distribution is called a component, K is the number of mixture components, and π k The weights of the corresponding components, and the probability density function for each component is: In solving for layer parameters using the expectation-maximization algorithm, the parameters of the first, second, and third layers to be estimated are first initialized, and then parameter estimation is performed iteratively multiple times. Each iteration executes the following two steps: Step E calculates the complete data log-likelihood function based on the existing observed data and the current estimated values ​​of the parameters. Assume there is a dataset Z, Z = {X, Y}, where X is called the observed dataset, i.e., the historical information of the adjacent lower layers of the node layer whose layer parameters need to be solved; Y is called the unobserved dataset, i.e., the historical information of the node layer whose layer parameters need to be solved; and Z is called the complete dataset. The complete data expected log-likelihood function is also called the Q function, i.e. The M-step selects the parameter that maximizes this expectation as the new estimate, i.e. The first layer parameters obtained when the algorithm runs to the kth iteration are: The weights of the estimated mixture components of the joint distribution are given, and the second-level parameters are... The mean of the estimated mixture components of the joint distribution is given, and the third-level parameters are... Let be the covariance matrix of the estimated mixture components of the joint distribution, where That is, the probability that the i-th sample belongs to each of the mixture components. Furthermore, to avoid overflow and singular values ​​when solving for the third-layer parameters, the third-layer parameters can be obtained by modifying the estimation of the covariance matrix. λ is a preset parameter, I d It is a d-dimensional identity matrix, and using it for estimation ensures that the covariance matrix is ​​non-zero and avoids singularities.

[0102] Furthermore, regarding the Gaussian mixture model constructed in the embodiments of this application... x R For the current network node, x S These are the corresponding bottom-level nodes, representing the first-level parameters, second-level parameters, and third-level parameters, respectively. S The marginal distribution can be written as p(x) S )=∑ xR p(x S ,x R According to Bayesian theory, the conditional probability p(x) R |x S ) can be represented as For a multivariate Gaussian distribution with density function G(x; μ, ∑), and in That is, the Gaussian mixture model can be further expressed as Based on Bayes' theorem, the conditional probability can be further expressed as: This represents the conditional probability of the current network node risk given the underlying node information. Since the first, second, and third layer parameters have already been obtained using the aforementioned expectation-maximization algorithm, the conditional probability representation can be further derived, yielding the first node parameter β. l Second node parameter μ lR|S and the third node parameter ∑ lR|S .

[0103] Furthermore, the parameters of the first node are based on The calculation yields β. l Let α be the parameter of the first node. lx is the first-level parameter of the node layer where the network node resides. S μ represents the risk value of the underlying node corresponding to the network node. lS For the second node parameter of the underlying node corresponding to the network node, ∑ lSS Let M be the autocovariance of the third node parameter corresponding to the underlying node of the network node, and M be the number of first-level parameters in the node layer where the underlying node of the network node resides. The second node parameter is calculated using the formula... The calculation yields μ. lR|S μ represents the conditional probability of the second node parameter corresponding to the network node. lR For the second node parameter corresponding to the network node, ∑ lRS This is the covariance between the third-node parameters of a network node and the third-node parameters of the corresponding underlying nodes. The third-node parameters are calculated using the formula... The calculation yields ∑ lR|S Let ∑ be the conditional probability of the parameters of the third node corresponding to the network node. lRR Let ∑ be the autocovariance of the third node parameters of the network nodes. lSR This is the covariance between the third node parameter of the underlying node corresponding to the network node and the third node parameter of the network node.

[0104] Step S50: Based on the risk attribute value of each station and the risk attribute value of each interval within the target time interval, solve the dynamic Bayesian network to obtain the network capacity risk of the rail transit network.

[0105] In one possible implementation, after the electronic device determines the node parameters of each network node in the dynamic Bayesian network, it can solve the dynamic Bayesian network based on the risk attribute values ​​of each station and each interval within the target time interval to obtain the network capacity risk of the rail transit network. Optionally, the electronic device can use the risk attribute values ​​of each station and each interval within the target time interval as input to solve the mean of the distribution value corresponding to each network node in the dynamic Bayesian network layer by layer. Then, the mean of the distribution value corresponding to the top-level network node of the Bayesian network is used as the network capacity risk of the rail transit network.

[0106] Optionally, the electronic equipment can determine the observed value x based on the pre-determined risk attribute values ​​for each station and the risk attribute values ​​for the interval. S To predict the corresponding station capacity risk x R Further, using station capacity risk as an observation, the corresponding interval capacity risk is predicted. Then, using interval capacity risk as an observation, the network capacity risk is predicted. Finally, the mean of the distribution value corresponding to the network capacity risk is calculated to obtain the network capacity risk of the rail transit network. In each prediction process, the optimal estimate of the capacity risk, i.e., the mean of the distribution value, is obtained. It can be done through formula Calculated.

[0107] Furthermore, in order to grasp the uncertainty of the final calculated mean of the distribution value and improve the interpretability of the prediction model, the electronic device can simultaneously calculate the conditional variance corresponding to the mean of the distribution value.

[0108] Based on the aforementioned technical features, this disclosure proposes a method for predicting the transportation safety situation of rail transit networks based on dynamic Bayesian networks and Gaussian mixture models. A dynamic Bayesian network structure is constructed based on the network topology and the interaction between capacity risks. A Gaussian mixture model is used to model the joint probability distribution of the local network, and the EM algorithm is used for parameter learning, demonstrating the superior performance of the Gaussian mixture model and dynamic network compared to linear Gaussian Bayesian networks. This method for predicting network capacity risks ensures interpretability and further improves prediction accuracy, and can be transferred to abnormal scenarios, further expanding its applicability.

[0109] Figure 4 A schematic diagram of a capacity risk determination device based on a dynamic Bayesian network according to an embodiment of the present disclosure is shown. Figure 4 As shown, the capacity risk determination device based on dynamic Bayesian networks in this embodiment of the present disclosure may include:

[0110] The network determination module 40 is used to determine a rail transit network including at least one line, each line including at least two stations, and an operating section between each two stations;

[0111] The attribute determination module 41 is used to determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval.

[0112] Network construction module 42 is used to construct a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section.

[0113] Model building module 43 is used to build a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network in order to determine the network parameters of the dynamic Bayesian network.

[0114] The risk prediction module 44 is used to solve the dynamic Bayesian network based on the risk attribute value of each station and the risk attribute value of each interval within the target time interval, so as to obtain the network capacity risk of the rail transit network.

[0115] In one possible implementation, the network building module 42 is further configured to:

[0116] Determine the risk correspondence between the risk attribute of each station and the station capacity risk corresponding to each station in the rail transit network;

[0117] Determine the risk correspondence between each of the interval risk attributes and the interval capacity risk corresponding to each of the operating intervals in the rail transit network;

[0118] Determine the risk correspondence between the capacity risk of each station, the capacity risk of each section, and the line capacity risk of each line in the rail transit network;

[0119] Determine the risk correspondence between the capacity risk of each line and the network capacity risk of the corresponding rail transit network;

[0120] The station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are used as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0121] In one possible implementation, the device further includes:

[0122] The single-point risk determination module is used to determine the single-point risk of each of the stations and the single-point risk of each of the operating sections.

[0123] The network construction module 42 is further used for:

[0124] Determine the risk correspondence between each station's single-point risk and the station's risk attribute, as well as the risk correspondence between each interval's single-point risk and the interval's risk attribute;

[0125] The network construction module 42 is further used for:

[0126] The station single-point risk, the interval single-point risk, the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are taken as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

[0127] In one possible implementation, the model building module 43 is further configured to:

[0128] Determine the historical station risk attributes, section risk attributes, station capacity risk, section capacity risk, line capacity risk, and network capacity risk of the rail transit network.

[0129] Based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, the Gaussian mixture model is solved to obtain the node parameters of each network node in the dynamic Bayesian network.

[0130] In one possible implementation, the model building module 43 is further configured to:

[0131] The Gaussian mixture model is solved using the expectation-maximization algorithm based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, thereby obtaining the layer parameters of each node layer in the dynamic Bayesian network.

[0132] The node parameters of the network nodes included in each node layer are calculated based on the layer parameters of each node layer.

[0133] In one possible implementation, the layer parameters include first layer parameters, second layer parameters, and third layer parameters, and the node parameters include first node parameters, second node parameters, and third node parameters;

[0134] The first node parameters are based on The calculation yields β. l Let α be the parameter of the first node. l x is the first-level parameter of the node layer where the network node is located. S μ is the risk value of the underlying node corresponding to the network node. lS For the second node parameter of the underlying node corresponding to the network node, ∑ lSS M is the autocovariance of the third node parameter of the underlying node corresponding to the network node, and M is the number of first-level parameters of the node layer where the underlying node of the network node is located.

[0135] The second node parameter is based on the formula The calculation yields μ. lR|S μ is the conditional probability of the second node parameter corresponding to the network node. lR For the second node parameters corresponding to the network node, ∑ lRS The covariance is the third node parameter of the network node and the third node parameter of the corresponding underlying node of the network node.

[0136] The third node parameter is based on the formula. The calculation yields ∑ lR|SLet ∑ be the conditional probability of the third node parameter corresponding to the network node. lRR Let ∑ be the autocovariance of the third node parameters of the network node. lSR The covariance is the third node parameter of the underlying node corresponding to the network node and the third node parameter of the network node.

[0137] In one possible implementation, the risk prediction module 44 is further configured to:

[0138] Using the risk attribute value of each station and the risk attribute value of each interval within the target time interval as input, the mean distribution value corresponding to each network node in the dynamic Bayesian network is solved layer by layer.

[0139] The mean distribution value of the top-level network node of the Bayesian network is used as the network capacity risk of the rail transit network.

[0140] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0141] This disclosure also proposes a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the above-described method. The computer-readable storage medium can be volatile or non-volatile.

[0142] This disclosure also proposes an electronic device, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.

[0143] This disclosure also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0144] Figure 5 A schematic diagram of an electronic device 800 according to an embodiment of the present disclosure is shown. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0145] Reference Figure 5The electronic device 800 may include one or more of the following components: processing component 802, memory 804, power supply component 806, multimedia component 808, audio component 810, input / output interface 812 (I / O interface), sensor component 814, and communication component 816.

[0146] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0147] Memory 804 is configured to store various types of data to support the operation of electronic device 800. Examples of this data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0148] Power supply component 806 provides power to various components of electronic device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.

[0149] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0150] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0151] Input / output interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0152] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 can detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0153] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0154] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0155] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions that can be executed by a processor 820 of an electronic device 800 to perform the above-described method.

[0156] Figure 6 A schematic diagram of another electronic device 1900 according to an embodiment of the present disclosure is shown. For example, the electronic device 1900 may be provided as a server or a terminal device. (Refer to...) Figure 6 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0157] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). Electronic device 1900 can operate on an operating system, such as Windows Server, stored in memory 1932. TM Mac OS X TM Unix TMLinux TM FreeBSD TM Or similar.

[0158] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of an electronic device 1900 to perform the above-described method.

[0159] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0160] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0161] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0162] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0163] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0164] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0165] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0166] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0167] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for determining transport capacity risk based on dynamic Bayesian networks, characterized in that, The method includes: Define a rail transit network comprising at least one line, each line comprising at least two stations, with an operating section between each pair of stations; Determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval; A dynamic Bayesian network for predicting network capacity risk is constructed based on the rail transit network, the risk attributes of each station, and the risk attributes of each section. A Gaussian mixture model is established based on the joint distribution of the dynamic Bayesian network to determine the network parameters of the dynamic Bayesian network. Based on the risk attribute value of each station and the risk attribute value of each interval within the target time interval, the dynamic Bayesian network is solved to obtain the network capacity risk of the rail transit network; The construction of a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section includes: Determine the risk correspondence between the risk attribute of each station and the station capacity risk corresponding to each station in the rail transit network; Determine the risk correspondence between each of the interval risk attributes and the interval capacity risk corresponding to each of the operating intervals in the rail transit network; Determine the risk correspondence between the capacity risk of each station, the capacity risk of each section, and the line capacity risk of each line in the rail transit network; Determine the risk correspondence between the capacity risk of each line and the network capacity risk of the corresponding rail transit network.

2. The method according to claim 1, characterized in that, The construction of a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section includes: The station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are used as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

3. The method according to claim 2, characterized in that, The method further includes: Determine the station-specific risk for each of the aforementioned stations, and the section-specific risk for each of the aforementioned operating sections; The step of constructing a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section further includes: Determine the risk correspondence between each station's single-point risk and the station's risk attribute, as well as the risk correspondence between each interval's single-point risk and the interval's risk attribute; The step of using the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the network capacity risk as network nodes, and then sequentially connecting each network node according to the risk correspondence to obtain a dynamic Bayesian network, includes: The station single-point risk, the interval single-point risk, the station risk attribute, the interval risk attribute, the station capacity risk, the interval capacity risk, the line capacity risk, and the road network capacity risk are taken as network nodes, and each network node is connected in an orderly manner according to the risk correspondence to obtain a dynamic Bayesian network.

4. The method according to claim 2 or 3, characterized in that, The step of establishing a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network to determine the network parameters of the dynamic Bayesian network includes: Determine the station risk attributes, section risk attributes, station capacity risk, section capacity risk, line capacity risk, and network capacity risk of the rail transit network in history; Based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, the Gaussian mixture model is solved to obtain the node parameters of each network node in the dynamic Bayesian network.

5. The method according to claim 4, characterized in that, The Gaussian mixture model is solved based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk to obtain the node parameters of each network node in the dynamic Bayesian network, including: The Gaussian mixture model is solved using the expectation-maximization algorithm based on the historical station risk attributes, interval risk attributes, station capacity risk, interval capacity risk, line capacity risk, and network capacity risk, thereby obtaining the layer parameters of each node layer in the dynamic Bayesian network. The node parameters of the network nodes included in each node layer are calculated based on the layer parameters of each node layer.

6. The method according to claim 5, characterized in that, The layer parameters include first layer parameters, second layer parameters, and third layer parameters, and the node parameters include first node parameters, second node parameters, and third node parameters; The first node parameters are based on The calculation yields β. l Let α be the parameter of the first node. l x is the first-level parameter of the node layer where the network node is located. S μ is the risk value of the underlying node corresponding to the network node. lS For the second node parameter of the underlying node corresponding to the network node, ∑ lSS M is the autocovariance of the third node parameter of the underlying node corresponding to the network node, and M is the number of first-level parameters of the node layer where the underlying node of the network node is located. The second node parameter is based on the formula The calculation yields μ. lR|S μ is the conditional probability of the second node parameter corresponding to the network node. lR For the second node parameters corresponding to the network node, ∑ lRS The covariance is the third node parameter of the network node and the third node parameter of the corresponding underlying node of the network node. The third node parameter is based on the formula. The calculation yields ∑ lR|S Let ∑ be the conditional probability of the third node parameter corresponding to the network node. lRR Let ∑ be the autocovariance of the third node parameters of the network node. lSR The covariance is the third node parameter of the underlying node corresponding to the network node and the third node parameter of the network node.

7. The method according to claim 2, characterized in that, The process of solving the dynamic Bayesian network based on the risk attribute values ​​of each station and each interval within the target time interval to obtain the network capacity risk of the rail transit network includes: Using the risk attribute value of each station and the risk attribute value of each interval within the target time interval as input, the mean distribution value corresponding to each network node in the dynamic Bayesian network is solved layer by layer. The mean distribution value of the top-level network node of the Bayesian network is used as the network capacity risk of the rail transit network.

8. A capacity risk determination device based on dynamic Bayesian networks, characterized in that, The device includes: A road network determination module is used to determine a rail transit road network including at least one line, each line including at least two stations, and an operating section between every two stations; The attribute determination module is used to determine the station risk attribute of each station in the rail transit network within the target time interval, and the interval risk attribute of each operating interval. A network construction module is used to construct a dynamic Bayesian network for predicting network capacity risk based on the rail transit network, the risk attributes of each station, and the risk attributes of each section. The model building module is used to establish a Gaussian mixture model based on the joint distribution of the dynamic Bayesian network in order to determine the network parameters of the dynamic Bayesian network. The risk prediction module is used to solve the dynamic Bayesian network based on the risk attribute value of each station and the risk attribute value of each interval within the target time interval, so as to obtain the network capacity risk of the rail transit network. The network construction module is further used for: Determine the risk correspondence between the risk attribute of each station and the station capacity risk corresponding to each station in the rail transit network; Determine the risk correspondence between each of the interval risk attributes and the interval capacity risk corresponding to each of the operating intervals in the rail transit network; Determine the risk correspondence between the capacity risk of each station, the capacity risk of each section, and the line capacity risk of each line in the rail transit network; Determine the risk correspondence between the capacity risk of each line and the network capacity risk of the corresponding rail transit network.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1 to 7 when executing instructions stored in the memory.

10. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images