A sampling method for extreme failure scenarios in road networks

Abstract

The application provides a road network extreme failure scenario sampling method. The method comprises the following steps: obtaining a same failure probability assumed for all road segments in advance, randomly sampling a specified number of failure scenarios of the road network according to the failure probability, selecting the first set proportion of failure scenarios with the most serious delay as risk scenarios, counting the frequency of each road segment appearing in the failure road segment set of the risk scenario to obtain the importance coefficient of each road segment, and resampling a specified number of failure scenarios of the road network according to the importance coefficient of the road segment. If the average travel time threshold of the new risk scenario is greater than the average travel time threshold of the extreme failure scenario, the sampling is performed according to the current importance coefficient and the real failure probability of the road segment to obtain the road network extreme failure scenario. The method of the application improves the failure probability of the road segment that has a key influence on the network performance, and provides an effective method for evaluating the network resilience of the road system under extreme conditions and taking improvement measures accordingly.

Description

Technical Field

[0001] This invention relates to the field of road network management technology, and in particular to a method for sampling extreme failure scenarios of road networks. Background Technology

[0002] Natural disasters and other emergencies can severely damage road networks, causing large-scale traffic paralysis and delays, and significantly reducing the operational efficiency and service level of the road network system. Therefore, generating extreme failure scenarios for road networks is of great significance. By simulating and analyzing these scenarios, we can guide corresponding preventative measures to protect the network, reduce the impact of extreme events, and thus improve network reliability and resilience in the face of emergencies. However, because extreme failure scenarios have a low probability of occurrence, using ordinary Monte Carlo methods for scenario sampling often requires collecting a large number of samples, resulting in a heavy computational burden. Therefore, a more efficient method is needed to collect extreme failure scenarios for road networks.

[0003] Currently, existing technologies for collecting extreme failure scenarios of road networks include: a method for extracting regular and extreme scenarios from historical time-series power output data of new energy sources using the DBSCAN clustering algorithm, identifying noise points not clustered as extreme scenarios; another method discloses a method for generating extreme wind power operation scenarios, using a conditional generative adversarial network (GAN) to generate wind power output scenarios, constructing a set of extreme wind power output scenarios using the Bootstrap interval generation method, and extracting extreme output scenarios based on extreme fluctuation intervals; yet another method discloses a method for generating extreme wind power scenarios based on a GAN, using VAE models for data augmentation to obtain extreme wind power generation data, which is then used as additional conditional information for the GAN to generate extreme scenarios. A further method discloses a distribution network extreme scenario modeling method based on uncertainty analysis, simulating typical natural disaster scenarios using typical meteorological characteristics of extreme disasters, calculating the failure probability of elements such as lines and power generation resources under these scenarios, obtaining multiple failure scenarios, and selecting the scenario corresponding to the system information entropy with the highest probability as the extreme failure scenario.

[0004] The drawbacks of existing technologies for collecting extreme failure scenarios of road networks include: Firstly, existing infrastructure extreme scenario sampling techniques rely on historical data. These data extract scenarios with low probability of occurrence as extreme scenarios from historical data or samples generated based on historical data. However, since historical data only reflects past events and scenarios, and when historical data is scarce or incomplete, the extracted extreme scenarios cannot comprehensively cover all possible extreme situations, resulting in insufficient representativeness. Secondly, some techniques rely on models to generate extreme scenarios. The accuracy of the generated scenarios is limited by the quality of the input data and the training effect of the model. Some complex models may require significant computational resources and time. Using ordinary Monte Carlo methods to generate random scenarios to obtain extreme failure scenarios with very low probability of occurrence also faces a heavy computational burden. Summary of the Invention

[0005] Embodiments of the present invention provide a method for sampling extreme failure scenarios of road networks, so as to effectively improve the management efficiency of road networks.

[0006] To achieve the above objectives, the present invention adopts the following technical solution.

[0007] A method for sampling extreme failure scenarios in road networks includes:

[0008] Obtain relevant attributes and travel demands of the road network to be sampled;

[0009] All road segments are assumed to have the same failure probability. A specified number of failure scenarios of the road network are randomly sampled according to the failure probability, and the performance index of each failure scenario is evaluated. The failure scenarios of the road network include the failure of multiple road segments in the road network.

[0010] All failure scenarios are sorted from high to low according to the network latency reflected by the performance indicators. The failure scenarios with the most severe latency are selected as risk scenarios. The frequency of each road segment appearing in the failure road segment cluster of risk scenarios is counted and weighted to obtain the importance coefficient of each road segment.

[0011] Based on the importance coefficient of the road segment, a specified number of failure scenarios of the road network are resampled, new risk scenarios are screened, and the average travel time threshold corresponding to the new risk scenarios is obtained.

[0012] The average travel time threshold for new risk scenarios is compared with the average travel time threshold for set extreme failure scenarios. If the average travel time threshold for new risk scenarios is greater than the average travel time threshold for extreme failure scenarios, sampling is performed based on the current importance coefficient and the actual failure probability of road segments to obtain extreme failure scenarios of the road network. Otherwise, the importance coefficient of road segments is re-obtained based on the current risk scenarios, a specified number of failure scenarios of the road network are resampled, and new risk scenarios are re-selected.

[0013] Preferably, obtaining the relevant attributes and travel demands of the road network to be sampled includes:

[0014] Obtain the relevant attributes and travel demand of the road network to be sampled, and represent the road network as a directed graph. The nodes in the directed graph represent intersections or traffic demand points, and the edges represent road segments. The relevant attributes include the free-flow time, traffic capacity, and actual failure probability of the road segments. The travel demand includes the origin and destination of the trip and the demand between each pair of origin and destination.

[0015] Preferably, the method involves pre-assuming the same failure probability for all road segments, randomly sampling a specified number of failure scenarios of the road network according to this failure probability, and evaluating the performance indicators of each failure scenario. The failure scenarios of the road network include the failure of multiple road segments within the road network, including:

[0016] All road segments are pre-assumed to have the same failure probability. Based on this failure probability, N segments are randomly sampled using the ordinary Monte Carlo method. t The failure scenarios of the road network are identified, and the performance index of each failure scenario is evaluated. The assumed failure probability of each road segment is taken as the expected value of the actual failure probability of all road segments. The failure scenario includes the failure of multiple road segments, and the traffic capacity of the failed road segments is reduced. The performance index of the failure scenario is the average travel time of the network. The average travel time of the network is calculated by dividing the sum of the products of the travel time of all road segments and the traffic flow of each road segment by the total travel demand of the network. The traffic flow of each road segment is obtained through a traffic assignment model based on user equilibrium.

[0017] Preferably, the process involves sorting all failure scenarios from highest to lowest according to the network latency reflected by performance indicators, selecting the failure scenarios with the most severe latency as risk scenarios, statistically analyzing the frequency of each road segment appearing in the failure segment cluster of risk scenarios, and performing weighted processing to obtain the importance coefficient of each road segment, including:

[0018] All failure scenarios are sorted from high to low according to the network delay level reflected by the average travel time of the network, and the failure scenarios with the most severe delays are selected as risk scenarios.

[0019] The frequency of each road segment occurring in the failure segment cluster of risk scenarios is statistically analyzed and weighted to obtain the road segment l. i Importance coefficient

[0020] The formula for weighted processing is as follows:

[0021]

[0022] Where, N t N represents the number of randomly sampled failure scenarios, and N is the total number of road segments in the network. fk Let be the number of failed road segments in the k-th failure scenario. It is a binary variable, when road segment l i It fails in the k-th failure scenario. Select 1 if the value is 1, otherwise select 0.

[0023] Preferably, the step of resampling a specified number of road network failure scenarios based on the importance coefficient of road segments, screening new risk scenarios, and obtaining the average travel time threshold corresponding to the new risk scenarios includes:

[0024] The importance coefficient of the road segment Multiply the assumed failure probability of the road segment by the assumed failure probability to obtain the new assumed failure probability, and sample N according to the new assumed failure probability. t Failure scenarios for each road network were identified, and performance metrics for each failure scenario were evaluated.

[0025] Newly sampled N t The failure scenarios are sorted from highest to lowest according to the network latency reflected by the performance indicators. The failure scenarios with the most severe latency, representing a predetermined proportion, are selected as new risk scenarios. The average travel time threshold θ′ corresponding to these new risk scenarios is then obtained. r .

[0026] Preferably, the step of comparing the average travel time threshold of the new risk scenario with the set average travel time threshold of the extreme failure scenario, if the average travel time threshold of the new risk scenario is greater than the average travel time threshold of the extreme failure scenario, sampling is performed based on the current importance coefficient and the actual failure probability of the road segment to obtain the extreme failure scenario of the road network; otherwise, the importance coefficient of the road segment is re-obtained based on the current risk scenario, and N is resampled. t Failure scenarios for the road network were analyzed, and new risk scenarios were re-screened, including:

[0027] Compare the average travel time threshold θ′ for new risk scenarios r With the set average travel time threshold θ for extreme failure scenarios e When θ′ r Greater than θe The current importance coefficient is multiplied by the actual failure of the road segment to obtain the sampling probability of the road segment failure. Extreme failure scenarios of the road network are sampled and obtained. Otherwise, the importance coefficient of each road segment is recalculated based on the current risk scenario, a specified number of failure scenarios are sampled, and new risk scenarios are re-selected until the average travel time threshold of the new risk scenario is greater than the average travel time threshold of the extreme failure scenario.

[0028] As can be seen from the technical solutions provided by the embodiments of the present invention above, the present invention proposes a sampling method for extreme failure scenarios of road networks based on the cross-entropy method in importance sampling theory. By identifying road segments that have a critical impact on network performance and increasing their failure probability, while reducing the failure probability of non-critical road segments, the sampling efficiency for extreme failure scenarios is improved. The proposed method does not rely on historical data, enabling more comprehensive collection of possible extreme failure scenarios. Furthermore, this method is model-free, easy to implement, has low computational complexity, and strong scalability, providing a more efficient and comprehensive scenario generation method for studying the resilience of infrastructure networks under extreme scenarios.

[0029] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of the invention. Attached Figure Description

[0030] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating a sampling method for extreme failure scenarios of a road network provided in an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram of a SiouxFalls road network topology provided in an embodiment of the present invention. Detailed Implementation

[0033] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0034] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or couplings. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0035] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0036] To facilitate understanding of the embodiments of the present invention, the following will provide further explanation and description with reference to the accompanying drawings and several specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0037] The processing flow of a road network extreme failure scenario sampling method provided in this embodiment of the invention is as follows: Figure 1 As shown, the processing steps include the following:

[0038] Step S1: Obtain the relevant attributes and travel demand of the road network to be sampled. Represent the road network as a directed graph. The nodes in the directed graph represent intersections or traffic demand points, and the edges represent road segments. The relevant attributes include the free-flow time, capacity, and actual failure probability of the road segments. The travel demand includes the origin and destination of the trip and the demand between each pair of origin and destination.

[0039] Step S2: Assume the same failure probability for all road segments, and randomly sample N segments using the ordinary Monte Carlo method according to this failure probability. tThis study identifies failure scenarios for a road network and evaluates the performance metrics for each failure scenario. The standard Monte Carlo method for sampling a failure scenario involves: for each road segment, generating a uniformly distributed random number between 0 and 1; comparing this random number with the road segment's failure probability; if the random number is less than the failure probability, the road segment is considered failed; otherwise, the road segment is considered alive. The assumed failure probability for each road segment is the expected value of the actual failure probabilities of all road segments. Failure scenarios include multiple road segment failures, resulting in reduced traffic capacity for the failed road segments.

[0040] The performance metric for failure scenarios is the average travel time of the network. The average travel time of the network is calculated by summing the products of the travel time of all road segments and the traffic flow of each road segment, and dividing the total network travel demand. The traffic flow of each road segment is obtained through a traffic assignment model based on user equilibrium.

[0041] Step S3: Sort all failure scenarios from high to low according to the network latency reflected by the performance indicators, and select the failure scenario with the most severe latency at a predetermined percentage (e.g., 10%) as the risk scenario.

[0042] Step S4: Count the frequency of each road segment in the failure road segment cluster of the risk scenario, and perform weighted processing to obtain road segment l. i Importance coefficient

[0043] The formula for weighted processing is as follows:

[0044]

[0045] Where, N t N represents the number of randomly sampled failure scenarios, and N is the total number of road segments in the network. fk Let be the number of failed road segments in the k-th failure scenario. It is a binary variable, when road segment l i It fails in the k-th failure scenario. Select 1 if the value is 1, otherwise select 0.

[0046] Step S5: Assign importance coefficients to road segments Multiply the assumed failure probability of the road segment by the assumed failure probability to obtain the new assumed failure probability, and sample N according to the new assumed failure probability. t The failure scenarios of the road network are identified, and the average travel time for each failure scenario is evaluated.

[0047] Step S6: Transfer the newly sampled N t The failure scenarios are sorted from highest to lowest based on the network delay level reflected by the average travel time. The failure scenario with the most severe delay, representing a predetermined proportion (e.g., 10%), is selected as a new risk scenario, and the average travel time threshold θ′ corresponding to these new risk scenarios is obtained. rThat is, in these new risk scenarios, the minimum of the average travel times is taken as the average travel time threshold.

[0048] Step S7: Compare θ′ r The average travel time threshold θ for extreme failure scenarios e When θ′ r Greater than θ e Proceed to step S8; otherwise, proceed to step S4. The average travel time threshold for extreme failure scenarios can be set according to actual conditions and requirements.

[0049] Step S8: Assign importance coefficients to road segments Multiplying this by the actual failure probability of the road segment yields the new actual failure probability of the road segment. Failure scenarios of the road network are then sampled according to this new actual failure probability, with the average travel time threshold being greater than θ. e These failure scenarios serve as extreme failure scenarios for road networks, and can be further applied to the resilience and reliability assessment and optimization of the system.

[0050] Taking the SiouxFalls road network as an example network, according to step S1, the relevant network attributes are obtained as shown in Table 1. Assuming that the actual failure probability of each road segment is 0.1, the travel demand is shown in Table 2. Figure 2 This is a schematic diagram of a SiouxFalls road network topology provided in an embodiment of the present invention.

[0051] Table 1 SiouxFalls Network Attributes

[0052]

[0053]

[0054] Table 2 SiouxFalls Network Travel Demand (×10) 2 )

[0055] O\D 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 0 1 1 5 2 3 5 8 5 13 5 2 5 3 5 5 4 1 3 3 1 4 3 1 2 1 0 1 2 1 4 2 4 2 6 2 1 3 1 1 4 2 0 1 1 0 1 0 0 3 1 1 0 2 1 3 1 2 1 3 3 2 1 1 1 2 1 0 0 0 0 1 1 0 4 5 2 2 0 5 4 4 7 7 12 14 6 6 5 5 8 5 1 2 3 2 4 5 2 5 2 1 1 5 0 2 2 5 8 10 5 2 2 1 2 5 2 0 1 1 1 2 1 0 6 3 4 3 4 2 0 4 8 4 8 4 2 2 1 2 9 5 1 2 3 1 2 1 1 7 5 2 1 4 2 4 0 10 6 19 5 7 4 2 5 14 10 2 4 5 2 5 2 1 8 8 4 2 7 5 8 10 0 8 16 8 6 6 4 6 22 14 3 7 9 4 5 3 2 9 5 2 1 7 8 4 6 8 0 28 14 6 6 6 9 14 9 2 4 6 3 7 5 2 10 13 6 3 12 10 8 19 16 28 0 40 20 19 21 40 44 39 7 18 25 12 26 18 8 11 5 2 3 15 5 4 5 8 14 39 0 14 10 16 14 14 10 1 4 6 4 11 13 6 12 2 1 2 6 2 2 7 6 6 20 14 0 13 7 7 7 6 2 3 4 3 7 7 5 13 5 3 1 6 2 2 4 6 6 19 10 13 0 6 7 6 5 1 3 6 6 13 8 8 14 3 1 1 5 1 1 2 4 6 21 16 7 6 0 13 7 7 1 3 5 4 12 11 4 15 5 1 1 5 2 2 5 6 10 40 14 7 7 13 0 12 15 2 8 11 8 26 10 4 16 5 4 2 8 5 9 14 22 14 44 14 7 6 7 12 0 28 5 13 16 6 12 5 3 17 4 2 1 5 2 5 10 14 9 39 10 6 5 7 15 28 0 6 17 17 6 17 6 3 18 1 0 0 1 0 1 2 3 2 7 2 2 1 1 2 5 6 0 3 4 1 3 1 0 19 3 1 0 2 1 2 4 7 4 18 4 3 3 3 8 13 17 3 0 12 4 12 3 1 20 3 1 0 3 1 3 5 9 6 25 6 5 6 5 11 16 17 4 12 0 12 24 7 4 21 1 0 0 2 1 1 2 4 3 12 4 3 6 4 8 6 6 1 4 12 0 18 7 5 22 4 1 1 4 2 2 5 5 7 26 11 7 13 12 26 12 17 3 12 24 18 0 21 11 23 3 0 1 5 1 1 2 3 5 18 13 7 8 11 10 5 6 1 3 7 7 21 0 7 24 1 0 0 2 0 1 1 2 2 8 6 5 7 4 4 3 3 0 1 4 5 11 7 0

[0056] According to step S2, a failure probability of 0.1 is assumed for all road segments. Based on this failure probability, 1000 failure scenarios of the road network are randomly sampled using the ordinary Monte Carlo method, where the traffic capacity of the failed road segments is reduced by 50%. The average travel time for each failure scenario is evaluated.

[0057] According to step S3, all failure scenarios are sorted from high to low according to the degree of network delay, and the top 10% of failure scenarios with the most severe delays are selected as risk scenarios.

[0058] According to step S4, the frequency of each road segment appearing in the failure road segment cluster of the risk scenario is counted and weighted to obtain the importance coefficient of road segment li. The results are shown in Table 3.

[0059] Table 3. Importance coefficients of SiouxFalls network segments

[0060]

[0061]

[0062] According to step S5, the importance coefficient of the road segment is multiplied by the assumed failure probability of the road segment to obtain the new assumed failure probability. 1000 failure scenarios of the road network are sampled according to the new assumed failure probability, and the average travel time of each failure scenario is evaluated.

[0063] According to step S6, the 1000 newly sampled failure scenarios are sorted from high to low according to the degree of network delay, and the top 10% of failure scenarios with the most severe delays are selected as new risk scenarios. The average travel time threshold θ′ corresponding to these new risk scenarios is determined. r It lasted 31.23 minutes.

[0064] According to step S7, if the average travel time threshold θ for extreme failure scenarios... e Set to 30 minutes, compare the average travel time threshold for new risk scenarios with the average travel time threshold for extreme failure scenarios. At this point, θ′ r Greater than θ e Proceed to step S8.

[0065] According to step S8, the importance coefficient of the road segment is... Multiplying this by the actual failure probability of the road segment yields the new actual failure probability of the road segment. Failure scenarios of the road network are then sampled according to this new actual failure probability, with the average travel time threshold being greater than θ. e The failure scenarios were taken as extreme failure scenarios of the road network. In this case, 149 extreme failure scenarios were sampled out of 1000 samples, while the ordinary Monte Carlo method sampled 34 extreme failure scenarios with the same number of samples.

[0066] In summary, this invention proposes a specific sampling method for extreme failure scenarios in road networks, based on importance sampling theory and the cross-entropy method. This method can quickly calculate an importance coefficient for each road segment in the network, thereby obtaining an importance weight. Sampling based on importance weights can effectively improve the sampling efficiency for extreme failure scenarios. Furthermore, the proposed method has low computational complexity, is simple to design and easy to implement, and can be extended to various infrastructure networks with different network structures and traffic distributions, providing an efficient and feasible solution for assessing the resilience of infrastructure networks under extreme conditions.

[0067] This invention increases the sample size for extreme failure scenarios by increasing the failure probability of road segments that have a critical impact on network performance and reducing the failure probability of non-critical road segments. This provides an effective method for assessing the network resilience of road systems under extreme conditions and taking corresponding improvement measures. Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this invention.

[0068] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0069] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0070] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for sampling extreme failure scenarios of a road network, characterized in that, The method comprises the following steps: obtaining relevant attributes and travel demand of a road network to be sampled; pre-supposing a same failure probability for all road segments, randomly sampling a specified number of failure scenarios of the road network according to the failure probability, the failure scenarios of the road network comprising failure of a plurality of road segments in the road network, and evaluating performance indicators of each failure scenario; sorting all the failure scenarios from high to low according to network delay degrees reflected by the performance indicators, selecting a specified proportion of failure scenarios with the most serious delay as risk scenarios, and statistically obtaining frequencies of each road segment in a failure road segment set of the risk scenarios and performing weighted processing to obtain an importance coefficient of each road segment; re-sampling a specified number of failure scenarios of the road network according to the importance coefficient of the road segment, screening new risk scenarios and obtaining an average travel time threshold corresponding to the new risk scenarios; comparing the average travel time threshold of the new risk scenarios with an average travel time threshold of a specified extreme failure scenario, if the average travel time threshold of the new risk scenarios is greater than the average travel time threshold of the extreme failure scenario, sampling according to the current importance coefficient and a real failure probability of the road segment to obtain an extreme failure scenario of the road network; otherwise, re-obtaining the importance coefficient of the road segment according to the current risk scenario, re-sampling a specified number of failure scenarios of the road network, and re-screening new risk scenarios; the sorting, selecting, statistically obtaining and weighted processing comprise: sorting all the failure scenarios from high to low according to network delay degrees reflected by average travel times of the network, and selecting a specified proportion of failure scenarios with the most serious delay as risk scenarios; The frequency of each road segment appearing in the failure road segment set of the risk scenario is counted, and weighted processing is performed to obtain the importance coefficient of the road segment l i ​ wherein the weighted processing formula is: where N t is the number of randomly sampled failure scenarios, N is the total number of links in the network, N fk is the number of failed links in the kth failure scenario, is a binary variable that takes the value 1 if link l i fails in the kth failure scenario, and 0 otherwise.

2. The method of claim 1, wherein, the obtaining comprises: obtaining relevant attributes and travel demand of a road network to be sampled, representing the road network as a directed graph, nodes in the directed graph representing intersections or traffic demand points, edges representing road segments, the relevant attributes comprising free-flow travel time, traffic capacity and real failure probability of the road segments, and the travel demand comprising travel start and end points and demand size between each pair of start and end points.

3. The method of claim 2, wherein, the pre-supposing, sampling and evaluating comprise: All road segments are pre-assumed to have the same failure probability. Based on this failure probability, N segments are randomly sampled using the ordinary Monte Carlo method. t The failure scenarios of the road network are identified, and the performance index of each failure scenario is evaluated. The assumed failure probability of each road segment is taken as the expected value of the actual failure probability of all road segments. The failure scenario includes the failure of multiple road segments, and the traffic capacity of the failed road segments is reduced. The performance index of the failure scenario is the average travel time of the network. The average travel time of the network is calculated by dividing the sum of the products of the travel time of all road segments and the traffic flow of each road segment by the total travel demand of the network. The traffic flow of each road segment is obtained through a traffic assignment model based on user equilibrium.

4. The method of claim 1, wherein, the re-sampling, screening and obtaining comprise: The importance coefficient of the road segment Multiply the assumed failure probability of the road segment by the assumed failure probability to obtain the new assumed failure probability, and sample N according to the new assumed failure probability. t Failure scenarios for each road network were identified, and performance metrics for each failure scenario were evaluated. N t newly sampled N t failure scenarios are sorted from high to low according to the network delay degree reflected by the performance index, and the first set proportion of failure scenarios with the most serious delay are selected as new risk scenarios, and the average travel time threshold corresponding to these new risk scenarios is obtained 5. The method of claim 4, wherein, The average travel time threshold of the new risk scenario is compared with the average travel time threshold of the set extreme failure scenario, if the average travel time threshold of the new risk scenario is greater than the average travel time threshold of the extreme failure scenario, sampling is performed according to the current importance coefficient and the real failure probability of the road section to obtain the extreme failure scenario of the road network; Otherwise, the importance coefficient of the road segment is reacquired according to the current risk scenario, N road network failure scenarios are resampled, and a new risk scenario is rescreened, comprising: t ​ average travel time threshold of the newly compared risk scenario average travel time threshold of the set extreme failure scenario when greater than multiply the current importance coefficient and the real failure probability of the road segment as the sampling probability of the road segment failure, sample and obtain the road network extreme failure scenario, otherwise, recalculate the importance coefficient of each road segment according to the current risk scenario and sample a specified number of failure scenarios, re-screen the new risk scenario until the average travel time threshold of the new risk scenario is greater than the average travel time threshold of the extreme failure scenario.

Images