Optimization method and system for damage repair strategy to improve rail transit network resilience

Abstract

The present invention relates to the field of emergency management technology, and specifically discloses a method and system for optimizing a disease repair strategy for improving the resilience of a rail transit network. The method includes forwarding a disease reporting request to a main terminal upon receipt, receiving a disease treatment plan sent by the main terminal, and calculating network performance parameters of the disease treatment plan; clustering the disease reporting requests to obtain each type of disease reporting request; comparing the network performance parameters of the disease treatment plan corresponding to each disease reporting request in the same type of disease reporting request, and selecting the optimal plan; and when a new disease reporting request is received, querying the optimal plan, serving as a reference plan, and feeding it back to the main terminal. The present invention establishes a weighted interval capacity evaluation index by integrating the network topology characteristics and actual operation characteristics of the rail transit line infrastructure, and combining the corresponding maintenance processes and work efficiency of common diseases, with excellent evaluation results.

Description

Technical Field

[0001] The present invention relates to the field of emergency management technology, and specifically to a method and system for optimizing a damage repair strategy for improving the resilience of a rail transit network. Background Art

[0002] During the construction, maintenance, and management of rail transit line infrastructure, a wide range of new technologies, equipment, materials, and processes have been adopted. This has resulted in increasing improvements in construction and maintenance management, and the overall stability and reliability of infrastructure. However, due to my country's vast territory, complex climate, topography, and geology, and the diverse types of infrastructure, some areas have experienced problems that have impacted the stability and durability of infrastructure. These problems have caused train speeds to decrease within these areas, reducing rail transit capacity and impacting daily transportation operations. Improving the resilience of rail transit line infrastructure systems has attracted considerable attention in recent years from industry practitioners and researchers both domestically and internationally.

[0003] Rail transit infrastructure network resilience is the ability of a rail transit network to quickly recover from damage or potential disturbances. During this period of resilience recovery, sections of the network may not operate at their original design speeds upon commissioning, and their capacity will be limited. Assessing the resilience of rail transit infrastructure networks and simulating network performance under various maintenance strategies facilitates decision-making and is crucial for the daily operation of rail transit infrastructure networks.

[0004] Large, complex construction projects often involve multiple construction sequencing options. Currently, commonly used construction sequencing algorithms primarily optimize for construction duration. However, they often fail to consider construction effectiveness, specifically the capacity of lines during maintenance and construction. Current network resilience assessment models typically calculate link length, node degree, or average efficiency, ignoring the differences in importance between different links. Therefore, it is necessary to determine the optimal maintenance approach for multiple disasters, within limited construction capacity and within interval speed limits, to achieve the fastest possible network performance recovery. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for optimizing damage repair strategies for improving the resilience of rail transit networks, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for optimizing a damage repair strategy for improving rail transit network resilience, the method comprising:

[0008] When a disease reporting request is received, it is forwarded to the central terminal, the disease treatment plan sent by the central terminal is received, and the network performance parameters of the disease treatment plan are calculated; the disease reporting request includes the disease type, disease location and disease time;

[0009] Cluster the disease reporting requests to obtain the disease reporting requests of each category;

[0010] Compare the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and select the best solution;

[0011] When a new disease reporting request is received, the optimal solution is queried and fed back to the main terminal as a reference solution.

[0012] As a further solution of the present invention, the steps of forwarding the received disease reporting request to the central terminal, receiving the disease treatment plan sent by the central terminal, and calculating the network performance parameters of the disease treatment plan include:

[0013] Receive the disease type and disease location based on the preset request receiving port, and use the information receiving time as the disease time;

[0014] Forward the disease type, disease location and disease time to the main terminal;

[0015] Receive the fault treatment plan sent by the main terminal and calculate the throughput capacity of each line and station in the fault treatment plan;

[0016] The network performance parameters of the disease treatment solution are calculated based on the throughput capability.

[0017] As a further solution of the present invention: the step of clustering the disease reporting requests to obtain each type of disease reporting request includes:

[0018] Read the fault location requested in the fault report, and obtain the routes and stations within the preset radius with the fault location as the center;

[0019] Converting routes and sites into a graph structure; the routes correspond to connecting lines, and the sites correspond to nodes;

[0020] Get the average daily traffic volume of the line, adjust the width of the connecting line based on the average daily traffic volume, get the site size of the site, and adjust the radius of the node based on the site size;

[0021] Cluster the disease reporting requests based on the graph structure of each disease reporting request to obtain each type of disease reporting request;

[0022] The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

[0023] As a further solution of the present invention, the step of comparing the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and selecting the optimal solution includes:

[0024] For any type of disease reporting request, query the network performance parameters of the disease treatment solution corresponding to each disease reporting request;

[0025] Compare network performance parameters and select the optimal network performance parameters;

[0026] Query the disease treatment plan corresponding to the optimal network performance parameters and use it as the optimal plan.

[0027] As a further embodiment of the present invention, the method further comprises:

[0028] Install sensors in the track to obtain track data;

[0029] Determine the spatial range based on the disease location in the disease reporting request, query the sensors within the spatial range, and establish a connection channel with the sensors;

[0030] A time range is determined based on the time of the fault in the fault reporting request, and track data within the time range is queried from the sensor data. The track data is stored in the form of a three-dimensional matrix. The three-dimensional matrix is ​​a matrix set based on time order in the time dimension of the two-dimensional matrix. The row and column positions in the two-dimensional matrix represent the installation position of the sensor.

[0031] Read all track data for the same type of disease reporting request and calculate the mean at each location. After the calculation is completed, the average track data obtained is called data feature;

[0032] Based on data features, track defects are predicted, the optimal solution is read, prompt information containing the optimal solution is generated, and feedback is sent to the main terminal.

[0033] As a further solution of the present invention, the steps of predicting rail defects based on data features, reading the optimal solution, generating prompt information containing the optimal solution, and feeding back the information to the main terminal include:

[0034] Generate prediction instructions regularly;

[0035] Traverse the entire track area based on a preset spatial radius to locate the sensor;

[0036] Read data from the located sensors based on a preset time radius and construct orbital data;

[0037] Compare the track data with all data features and determine the data feature with the highest matching degree;

[0038] When the highest matching degree reaches the preset matching degree threshold, the optimal solution for the disease reporting request corresponding to the data feature is read, and prompt information containing the optimal solution is generated and fed back to the main terminal.

[0039] The technical solution of the present invention also provides a system for optimizing a repair strategy for improving the resilience of a rail transit network, the system comprising:

[0040] A solution analysis module is used to forward a received disease reporting request to the central terminal, receive the disease treatment solution sent by the central terminal, and calculate the network performance parameters of the disease treatment solution; the disease reporting request includes the disease type, disease location, and disease time;

[0041] The request clustering module is used to cluster the disease reporting requests and obtain the disease reporting requests of each category;

[0042] The optimal solution selection module is used to compare the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and select the optimal solution;

[0043] The solution feedback module is used to query the optimal solution when receiving a new disease reporting request, and use it as a reference solution and feedback it to the main terminal.

[0044] As a further solution of the present invention: the solution analysis module includes:

[0045] An information receiving unit, configured to receive the disease type and disease location based on a preset request receiving port, and use the time of information reception as the disease time;

[0046] Information forwarding unit, used to forward the disease type, disease location and disease time to the main terminal;

[0047] The throughput capacity calculation unit is used to receive the fault treatment plan sent by the main terminal and calculate the throughput capacity of each line and station in the fault treatment plan;

[0048] A performance parameter calculation unit is used to calculate the network performance parameters of the disease treatment solution based on the throughput.

[0049] As a further solution of the present invention: the request clustering module includes:

[0050] The point acquisition unit is used to read the disease location requested by the disease report and obtain the lines and stations within a preset radius with the disease location as the center;

[0051] A graph structure generating unit, configured to convert lines and sites into a graph structure; the lines correspond to connecting lines, and the sites correspond to nodes;

[0052] A graph structure update unit is used to obtain the average daily traffic volume of the line, adjust the width of the connection line based on the average daily traffic volume, obtain the site scale of the site, and adjust the radius of the node based on the site scale;

[0053] A clustering execution unit, configured to cluster the disease reporting requests based on the graph structure of each disease reporting request to obtain disease reporting requests of each category;

[0054] The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

[0055] As a further solution of the present invention: the optimal solution selection module includes:

[0056] A parameter query unit, used to query the network performance parameters of the disease treatment solution corresponding to each disease reporting request for any type of disease reporting request;

[0057] Parameter comparison unit, used to compare network performance parameters and select the optimal network performance parameters;

[0058] The solution query unit is used to query the disease treatment solution corresponding to the optimal network performance parameters as the optimal solution.

[0059] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention establishes a weighted interval capacity evaluation index by integrating the network topology characteristics and actual operation characteristics of the rail transit line infrastructure, combined with the corresponding maintenance processes and work efficiency of common diseases, and the evaluation effect is excellent; in addition, the optimal resilience recovery strategy is not only a means to deal with diseases, but also an important guarantee for the safety of rail transit operations in response to sudden natural disasters or accidents. When faced with common diseases or accidents, the resilience recovery capacity of the rail transit line infrastructure network under the random recovery strategy is low, and it takes a long time to return to normal operation. For the same traffic volume, the higher the speed of the rail line, the higher the comprehensive resilience index value, the smaller the impact of the loss of capacity when disturbed by the disease on the network, and the better the operational stability; for rail transit lines with the same maximum speed, the greater the flow weight, the lower the comprehensive resilience index value, and the worse the operational stability. In general, in the face of the threat of diseases or accidents, the disease repair strategy optimization method proposed in this invention aims to improve the network resilience of rail transit line infrastructure, analyze the network performance during the maintenance cycle and after repair, and monitor the evolution of network resilience. This can improve the resilience of the rail transit network to a certain extent, adapt to the threat of diseases to rail transit line infrastructure, and enhance safety and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention.

[0061] Figure 1 Flowchart of a method for optimizing damage repair strategies to improve rail transit network resilience.

[0062] Figure 2 Block diagram of the first sub-process of the damage repair strategy optimization method for improving rail transit network resilience.

[0063] Figure 3 Diagram of the second sub-process of the damage repair strategy optimization method for improving rail transit network resilience.

[0064] Figure 4 Diagram of the third sub-process of the damage repair strategy optimization method for improving rail transit network resilience.

[0065] Figure 5 A structural block diagram of the system for optimizing damage repair strategies to improve rail transit network resilience. DETAILED DESCRIPTION

[0066] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] Figure 1 The present invention provides a flow chart of a method for optimizing a damage repair strategy for improving rail transit network resilience. In one embodiment of the present invention, the method includes:

[0068] Step S100: When a disease reporting request is received, it is forwarded to the central terminal, the disease treatment plan sent by the central terminal is received, and the network performance parameters of the disease treatment plan are calculated; the disease reporting request includes the disease type, disease location and disease time;

[0069] The defect reporting request includes the defect type and location uploaded by the road section manager (usually front-line personnel), and the upload time is used as the defect time. The defect refers to the factors on the road section that affect the normal passage of vehicles. The execution subject of this method is equivalent to a central control terminal, which is used to manage all road sections in the entire area. It receives the defect reporting request uploaded by the road section manager at any road section and then forwards the defect reporting request to the main terminal, which is a higher-level management terminal. After receiving the defect reporting request, the main terminal issues a defect treatment plan, and the execution subject of this method then forwards the disease treatment plan to the road section manager.

[0070] In actual applications, each road section has a corresponding road section manager. The road section manager of a region (such as a town) corresponds to a central control terminal, that is, the execution subject of this method. The main terminal is at a higher level and may be set up in a county, city or higher area. One main terminal corresponds to multiple central control terminals, and one central control terminal corresponds to multiple road section managers.

[0071] Step S200: clustering the disease reporting requests to obtain disease reporting requests of each category;

[0072] Over a long time span, there are many different disease reporting requests. By comparing the disease reporting requests pairwise, similar disease reporting requests can be classified into one category, thereby obtaining different categories of disease reporting requests.

[0073] Step S300: comparing network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type, and selecting the optimal solution;

[0074] For each type of disease reporting request, the network performance parameters of the disease treatment plan corresponding to the disease reporting request are read, and they are compared pairwise to determine the optimal network performance parameters and select the optimal plan.

[0075] Step S400: When a new disease reporting request is received, the optimal solution is searched and used as a reference solution, and then fed back to the main terminal;

[0076] After internal comparison of similar disease reporting requests, each type of disease reporting request has an optimal solution. When receiving a new disease reporting request uploaded by the road section management party, the execution body of this method (central control end) will determine which category the disease reporting request belongs to, query the corresponding optimal solution, and indicate the best solution for similar disease reporting requests in historical data. It will be used as a reference solution and fed back to the main end, which can assist the main end in making a disease repair strategy more quickly. In actual scenarios, the main end is generally a signal transmitter. For the execution body of this method, it only receives the solution sent by the main end. Specifically, the working situation at the main end needs to be roughly explained. When facing a disease, the staff at the main end needs to hold a meeting to discuss and determine the final solution, and then send it to the central control end. This takes a lot of time. Under the architecture of the technical solution of the present invention, if time is tight, the main end only needs to review whether the reference solution is significantly different and can use it directly, which is equivalent to having a guaranteed better solution.

[0077] Figure 2 The first sub-flow diagram of the method for optimizing a fault repair strategy to improve rail transit network resilience includes the following steps: when a fault reporting request is received, the fault report is forwarded to the main terminal, a fault treatment plan is received from the main terminal, and network performance parameters of the fault treatment plan are calculated:

[0078] Step S101: receiving the disease type and disease location based on a preset request receiving port, and using the information receiving time as the disease time;

[0079] Step S102: forwarding the disease type, disease location and disease time to the main terminal;

[0080] Step S103: receiving the fault treatment plan sent by the main terminal, and calculating the throughput capacity of each line and station in the fault treatment plan;

[0081] Step S104: Calculating network performance parameters of the disease treatment solution based on the throughput;

[0082] In an example of the technical solution of the present invention, a specific solution evaluation process is provided. Based on the preset request receiving port, the disease type and disease location are received, the time of information reception is used as the disease time, and the disease type, disease location and disease time are forwarded to the main end; the main end sends the disease treatment plan, and the execution subject of this method will calculate the throughput capacity of each line and station in the disease treatment plan, and then calculate the network performance parameters.

[0083] The calculation process of capacity and network performance parameters in the disease treatment plan is described in detail as follows:

[0084] The damage treatment plan includes a maintenance cycle and speed limit method. The capacity of the rail transit line during the maintenance period is calculated based on the maintenance cycle and speed limit method. Then, the resilience of the rail transit line infrastructure network is evaluated. In view of the current situation that rail transit lines find it difficult to maintain maximum capacity after damage occurs and during maintenance, the capacity-time change process is described based on infrastructure resilience theory, and the resilience of the rail transit line is evaluated.

[0085] Furthermore, regarding speed limit methods, including section speed limits for lines and station speed limits for stations, the two methods of calculating traffic capacity are different:

[0086] 1. Section speed limit:

[0087] Under local speed limit conditions, the train operation process in the section includes deceleration process, speed limit process, acceleration process and normal operation process, and their time is respectively 、 、 and . Section speed limit operation time The calculation method is: .

[0088] The speed limit occurs on the line between two stations. That is, when a train runs in a certain section, it needs to slow down to the specified speed. When the train runs in this section, it goes through four processes: deceleration process (from normal speed to speed limit), speed limit operation process (maintaining speed limit operation), acceleration process (from speed limit speed to normal speed) and normal running process (running after returning to normal speed). After that, the capacity calculation process under the interval speed limit condition is:

[0089] ; Indicates the rail transit maintenance window period (the time available for travel). Indicates the time it takes for a train to complete the entire section under speed limit conditions. Indicates the traffic capacity under speed limit conditions (the number of trains passing per unit time).

[0090] 2. Station speed limit:

[0091] Station speed limits are divided into two categories: non-stop and stop-stop. The speed limit level of the station is calculated and the additional running time caused by the speed limit is calculated. :

[0092] ;in, is the length of the main line within the station, speed limits for stations; The vehicle speed when there is no speed limit.

[0093] Reduction in throughput capacity caused by train stops during station speed restrictions The calculation method is:

[0094] ;

[0095] in, Indicates the stop time. and They represent the additional stopping time and additional starting time when the station speed is limited respectively; The minimum tracking interval time for high-speed railways;

[0096] Passing capacity under station speed limit conditions The calculation method is:

[0097] ;

[0098] in, is the number of speed-limited stations, is the number of stops at speed-limited stations, This is the window period for rail transit maintenance. is the number of stops, The reduction in capacity caused by train stops during normal station operation; 1440 is the total number of minutes in a day, which is determined by Calculated; m represents the number of speed-limited sections or speed-limited stations. The above calculation process needs to calculate the sum of the time added for each speed-limit point. The number of speed-limit points is set to m, and m is the total number of these speed-limit points traversed, which is used to sum the time added for each section.

[0099] Furthermore, the change of the rail transit line network performance function has three stages, including:

[0100] The first stage is the stage of disease occurrence. The destructive power of the disease is reflected in the local speed limit of the rail transit line, and the rail transit network performance function decreases. The second stage is the maintenance stage. The rail transit line network performance function increases with the maintenance time. The third stage is the recovery stage. After the maintenance activity, the speed of the rail transit line gradually increases from the speed limit to the normal value. The rail transit line network performance function increases with the maintenance time. The calculation method of the rail transit line network performance function within the time t during normal operation is:

[0101] ; For lines or stations in normal operation Through ability.

[0102] The calculation method of the rail transit line network performance function within the speed limit time t is:

[0103] ;

[0104] in, Lines or stations with speed limits Through ability.

[0105] In one embodiment of the technical solution of the present invention, network resilience can also be evaluated. The rail transit line network evaluation method is:

[0106] ; represents the final evaluation score, and A represents the total time from disease occurrence to speed recovery.

[0107] The core of the above content is the capacity calculation process during the speed limit process, which is described as follows:

[0108] Section speed limits occur between stations and are calculated by taking into account the entire process of train deceleration, speed-limited operation, acceleration, and normal operation. Station speed limits occur within stations and primarily affect the time it takes for trains to enter, stop, and exit the station, leading to a decrease in capacity. The capacity calculation process is as follows:

[0109] Section speed limit capacity = window period time / speed limit operation time;

[0110] The station speed limit capacity = the capacity after taking into account the additional time of stop.

[0111] It is worth mentioning that at the main end, the defect treatment plan generated at one time may not be unique. With the help of the above calculation process, combined with the mixed integer linear programming (MILP) method, a rail transit line maintenance decision optimization model under multi-point speed limit conditions can be established. The mixed integer linear programming method is used to form the optimal maintenance strategy with the fastest performance recovery, and then a defect treatment plan is determined and fed back to the execution entity of this method. At this time, the main end is also an autonomous intelligent device and does not require excessive human intervention. In fact, no matter what method the main end uses to determine the final defect treatment plan, for the execution entity of this method, it only receives one defect treatment plan and only performs the receiving action.

[0112] Figure 3 In the second sub-flow diagram of the method for optimizing the repair strategy for improving rail transit network resilience, the step of clustering the damage reporting requests to obtain each type of damage reporting request includes:

[0113] Step S201: Read the fault location in the fault reporting request, and obtain the routes and stations within a preset radius with the fault location as the center;

[0114] Step S202: Converting the routes and sites into a graph structure; the routes correspond to connecting lines, and the sites correspond to nodes;

[0115] Step S203: Obtain the average daily traffic volume of the line, adjust the width of the connecting line based on the average daily traffic volume, obtain the site scale of the site, and adjust the radius of the node based on the site scale;

[0116] Step S204: clustering the disease reporting requests based on the graph structure of each disease reporting request to obtain disease reporting requests of each category;

[0117] The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

[0118] In an example of the technical solution of the present invention, the clustering process of the disease reporting request is explained. The disease location of the disease reporting request is read, and the lines and stations within the preset radius are obtained with the disease location as the center. The lines and stations are converted into a graph structure. The graph structure includes nodes and connecting lines between nodes. The connecting lines in the graph structure correspond to the lines, and the stations in the graph structure correspond to the nodes. Then, the average daily traffic volume of the lines is obtained, and the width of the connecting lines is adjusted based on the average daily traffic volume. The site scale of the site is obtained, and the radius of the node is adjusted based on the site scale. The adjusted graph structure is obtained. The graph structure can be used as a feature of the disease reporting request. The graph structure is essentially two-dimensional data, and an image comparison algorithm can be applied to compare it. Comparing the graph structure of the disease reporting request is equivalent to comparing the disease reporting requests. The comparison process calculates the similarity. A subtraction function is introduced on the similarity to obtain the difference. The difference can be used as the distance in the clustering algorithm to cluster the data. In this application, the disease reporting requests are clustered. The clustering algorithm can be a conventional clustering algorithm, such as the K-means clustering algorithm with a variable K value.

[0119] Figure 4 The third sub-flow diagram of the method for optimizing a fault repair strategy to enhance rail transit network resilience includes the following steps: comparing network performance parameters of fault treatment solutions corresponding to similar fault reporting requests and selecting the optimal solution:

[0120] Step S301: For any type of disease reporting request, query the network performance parameters of the disease treatment solution corresponding to each disease reporting request;

[0121] Step S302: comparing network performance parameters and selecting the optimal network performance parameters;

[0122] Step S303: Query the disease treatment plan corresponding to the optimal network performance parameter and use it as the optimal plan.

[0123] In an example of the technical solution of the present invention, for any type of disease reporting request, the network performance parameters of the disease treatment solution corresponding to each disease reporting request are queried, the network performance parameters are compared, and the optimal network performance parameters are selected. The disease treatment solution corresponding to the optimal network performance parameters is the optimal solution.

[0124] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0125] Install sensors in the track to obtain track data;

[0126] Determine the spatial range based on the disease location in the disease reporting request, query the sensors within the spatial range, and establish a connection channel with the sensors;

[0127] A time range is determined based on the time of the fault in the fault reporting request, and track data within the time range is queried from the sensor data. The track data is stored in the form of a three-dimensional matrix. The three-dimensional matrix is ​​a matrix set based on time order in the time dimension of the two-dimensional matrix. The row and column positions in the two-dimensional matrix represent the installation position of the sensor.

[0128] Read all track data for the same type of disease reporting request and calculate the mean at each location. After the calculation is completed, the average track data obtained is called data feature;

[0129] Based on data features, track defects are predicted, the optimal solution is read, prompt information containing the optimal solution is generated, and feedback is sent to the main terminal.

[0130] In one example of the technical solution of the present invention, a sensor is installed in the track, and the sensor can actually obtain data. Different sensors can obtain data at different times, such as pressure and temperature. This application does not specifically limit these data, which are collectively referred to as track data; the spatial range is determined according to the disease position in the disease reporting request, the disease position is the center of the circle, and the preset value is the radius. The created circular area with a fixed position is the spatial range, the sensors within the spatial range are queried, and a connection channel with the sensor is established. The time range is determined according to the disease moment in the disease reporting request. The time range generally takes the disease moment as the end point, and obtains a period of time within a preset span as the time range, and queries the sensor data within the time range. Track data; specifically, the track data is stored in the form of a three-dimensional matrix. The three-dimensional matrix is ​​a matrix set based on time sequence of the two-dimensional matrix in the time dimension. The row and column positions in the two-dimensional matrix represent the installation position of the sensor. At this time, each position in the three-dimensional matrix is ​​represented by a three-dimensional coordinate (x, y, t). It is worth mentioning that its specific value can also be represented by an array, such as (A1, A2, ..., An). Each array element represents a sensor, which makes all types of sensor data represented by the same three-dimensional matrix. However, this method is more cumbersome. In practical applications, the same type of sensor is generally analyzed separately. At this time, the specific value is the sensor data, and the unit issue is not involved.

[0131] Based on the above content, the same data extraction method is used for each disease reporting request to construct a three-dimensional matrix of the same dimension. For any position in the three-dimensional matrix, the mean of the data of all three-dimensional matrices at that position is calculated. After the mean of each position is calculated, a three-dimensional matrix containing the mean is obtained, which can be used as a data feature. At this time, a type of disease reporting request corresponds to a data feature and also corresponds to an optimal solution. By establishing a connection between the data feature and the optimal solution, the sensor's detection data can be used as an independent variable to obtain the optimal solution. The sensor can obtain data in real time. Based on the data feature, the data obtained by the sensor is analyzed, and then the track is predicted for disease, the optimal solution is read, and prompt information containing the optimal solution is generated and fed back to the main end. This provides an early warning solution.

[0132] Specifically, the steps of predicting track defects based on data features, reading the optimal solution, generating prompt information containing the optimal solution, and feeding back the information to the main terminal include:

[0133] Generate prediction instructions regularly;

[0134] Traverse the entire track area based on a preset spatial radius to locate the sensor;

[0135] Read data from the located sensors based on a preset time radius and construct orbital data;

[0136] Compare the track data with all data features and determine the data feature with the highest matching degree;

[0137] When the highest matching degree reaches the preset matching degree threshold, the optimal solution for the disease reporting request corresponding to the data feature is read, and prompt information containing the optimal solution is generated and fed back to the main terminal.

[0138] In an example of the technical solution of the present invention, prediction instructions are generated periodically according to a preset time cycle, the entire track area is traversed based on a preset spatial radius, and the sensor is positioned. The spatial radius is the same as the radius of the determined spatial range. Data is read from the positioned sensor based on a preset time radius, and track data is constructed. The time radius is the same as the time span within the determined time range. The track data is compared with all data features to determine the data feature with the highest matching degree. When the highest matching degree reaches the preset matching degree threshold, it indicates that a corresponding disease reporting request is very likely to appear. Although the disease reporting request uploaded by the section management party has not been received at this time, a prompt message can be reported to the main end in advance, and the optimal solution can be fed back at the same time.

[0139] It should be noted that one of the steps in the above content is to traverse the entire track area based on a preset spatial radius to locate the sensor. The meaning of traversal is that the track area is very large, including all areas under the jurisdiction of the central control end, and the spatial radius is relatively small. Therefore, only a part of the track area can be extracted at a time, and the sensors in a part of the area can be analyzed. The process of extracting a part of the area is a traversal process, which can be compared to a window sliding in the image. The sliding step is a preset value, and an analysis process is performed each time it slides.

[0140] Figure 5 The present invention provides a structural block diagram of a system for optimizing a repair strategy for improving rail transit network resilience. In one embodiment of the present invention, a system for optimizing a repair strategy for improving rail transit network resilience is provided. The system 10 includes:

[0141] Solution analysis module 11 is used to forward the received disease reporting request to the main terminal, receive the disease treatment solution sent by the main terminal, and calculate the network performance parameters of the disease treatment solution; the disease reporting request includes the disease type, disease location and disease time;

[0142] The request clustering module 12 is used to cluster the disease reporting requests to obtain disease reporting requests of each category;

[0143] The optimal solution selection module 13 is used to compare the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and select the optimal solution;

[0144] The solution feedback module 14 is used to query the optimal solution when receiving a new disease reporting request, and feed it back to the main terminal as a reference solution.

[0145] Furthermore, the solution analysis module 11 includes:

[0146] An information receiving unit, configured to receive the disease type and disease location based on a preset request receiving port, and use the time of information reception as the disease time;

[0147] Information forwarding unit, used to forward the disease type, disease location and disease time to the main terminal;

[0148] The throughput capacity calculation unit is used to receive the fault treatment plan sent by the main terminal and calculate the throughput capacity of each line and station in the fault treatment plan;

[0149] A performance parameter calculation unit is used to calculate the network performance parameters of the disease treatment solution based on the throughput.

[0150] Specifically, the request clustering module 12 includes:

[0151] The point acquisition unit is used to read the disease location requested by the disease report and obtain the lines and stations within a preset radius with the disease location as the center;

[0152] A graph structure generating unit, configured to convert lines and sites into a graph structure; the lines correspond to connecting lines, and the sites correspond to nodes;

[0153] A graph structure update unit is used to obtain the average daily traffic volume of the line, adjust the width of the connection line based on the average daily traffic volume, obtain the site scale of the site, and adjust the radius of the node based on the site scale;

[0154] A clustering execution unit, configured to cluster the disease reporting requests based on the graph structure of each disease reporting request to obtain disease reporting requests of each category;

[0155] The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

[0156] Furthermore, the optimal solution selection module 13 includes:

[0157] A parameter query unit, used to query the network performance parameters of the disease treatment solution corresponding to each disease reporting request for any type of disease reporting request;

[0158] Parameter comparison unit, used to compare network performance parameters and select the optimal network performance parameters;

[0159] The solution query unit is used to query the disease treatment solution corresponding to the optimal network performance parameters as the optimal solution.

[0160] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing a repair strategy for improving rail transit network resilience, characterized in that: The method comprises: When a disease reporting request is received, it is forwarded to the central terminal, the disease treatment plan sent by the central terminal is received, and the network performance parameters of the disease treatment plan are calculated; the disease reporting request includes the disease type, disease location and disease time; Cluster the disease reporting requests to obtain the disease reporting requests of each category; Compare the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and select the best solution; When receiving a new disease reporting request, the system searches for the best solution, uses it as a reference solution, and feeds it back to the main terminal; The step of clustering the disease reporting requests to obtain each type of disease reporting request includes: Read the fault location requested in the fault report, and obtain the routes and stations within the preset radius with the fault location as the center; Converting routes and sites into a graph structure; the routes correspond to connecting lines, and the sites correspond to nodes; Get the average daily traffic volume of the line, adjust the width of the connecting line based on the average daily traffic volume, get the site size of the site, and adjust the radius of the node based on the site size; Cluster the disease reporting requests based on the graph structure of each disease reporting request to obtain each type of disease reporting request; The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

2. The method for optimizing the repair strategy for improving the resilience of rail transit networks according to claim 1 is characterized in that: The steps of forwarding the received disease reporting request to the main terminal, receiving the disease treatment plan sent by the main terminal, and calculating the network performance parameters of the disease treatment plan include: Receive the disease type and disease location based on the preset request receiving port, and use the information receiving time as the disease time; Forward the disease type, disease location and disease time to the main terminal; Receive the fault treatment plan sent by the main terminal and calculate the throughput capacity of each line and station in the fault treatment plan; The network performance parameters of the disease treatment solution are calculated based on the throughput capability.

3. The method for optimizing the repair strategy for improving the resilience of rail transit networks according to claim 1 is characterized in that: The step of comparing the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and selecting the optimal solution includes: For any type of disease reporting request, query the network performance parameters of the disease treatment solution corresponding to each disease reporting request; Compare network performance parameters and select the optimal network performance parameters; Query the disease treatment plan corresponding to the optimal network performance parameters and use it as the optimal plan.

4. The method for optimizing the repair strategy for improving the resilience of rail transit networks according to claim 1 is characterized in that: The method further comprises: Install sensors in the track to obtain track data; Determine the spatial range based on the disease location in the disease reporting request, query the sensors within the spatial range, and establish a connection channel with the sensors; A time range is determined based on the time of the fault in the fault reporting request, and track data within the time range is queried from the sensor data. The track data is stored in the form of a three-dimensional matrix. The three-dimensional matrix is ​​a matrix set based on time order in the time dimension of the two-dimensional matrix. The row and column positions in the two-dimensional matrix represent the installation position of the sensor. Read all track data for the same type of disease reporting request and calculate the mean at each location. After the calculation is completed, the average track data obtained is called data feature; Based on data features, track defects are predicted, the optimal solution is read, prompt information containing the optimal solution is generated, and feedback is sent to the main terminal.

5. The method for optimizing the repair strategy for improving the resilience of rail transit networks according to claim 4 is characterized in that: The steps of predicting track defects based on data features, reading the optimal solution, generating prompt information containing the optimal solution, and feeding back the information to the main terminal include: Generate prediction instructions regularly; Traverse the entire track area based on a preset spatial radius to locate the sensor; Read data from the located sensors based on a preset time radius and construct orbital data; Compare the track data with all data features and determine the data feature with the highest matching degree; When the highest matching degree reaches the preset matching degree threshold, the optimal solution for the disease reporting request corresponding to the data feature is read, and prompt information containing the optimal solution is generated and fed back to the main terminal.

6. A system for optimizing repair strategies for improving rail transit network resilience, characterized in that: The system comprises: A solution analysis module is used to forward a received disease reporting request to the central terminal, receive the disease treatment solution sent by the central terminal, and calculate the network performance parameters of the disease treatment solution; the disease reporting request includes the disease type, disease location, and disease time; The request clustering module is used to cluster the disease reporting requests and obtain the disease reporting requests of each category; The optimal solution selection module is used to compare the network performance parameters of the disease treatment solutions corresponding to the disease reporting requests of the same type and select the optimal solution; The solution feedback module is used to query the optimal solution when receiving a new disease reporting request, and use it as a reference solution and feedback it to the main terminal; The request clustering module includes: The point acquisition unit is used to read the disease location requested by the disease report and obtain the lines and stations within a preset radius with the disease location as the center; A graph structure generating unit, configured to convert lines and sites into a graph structure; the lines correspond to connecting lines, and the sites correspond to nodes; A graph structure update unit is used to obtain the average daily traffic volume of the line, adjust the width of the connection line based on the average daily traffic volume, obtain the site scale of the site, and adjust the radius of the node based on the site scale; A clustering execution unit, configured to cluster the disease reporting requests based on the graph structure of each disease reporting request to obtain disease reporting requests of each category; The distance parameter in the clustering process adopts the difference of the graph structure, and the difference adopts the image difference.

7. The system for optimizing the repair strategy for improving the resilience of rail transit networks according to claim 6 is characterized in that: The solution analysis module includes: An information receiving unit, configured to receive the disease type and disease location based on a preset request receiving port, and use the time of information reception as the disease time; Information forwarding unit, used to forward the disease type, disease location and disease time to the main terminal; The throughput capacity calculation unit is used to receive the fault treatment plan sent by the main terminal and calculate the throughput capacity of each line and station in the fault treatment plan; A performance parameter calculation unit is used to calculate the network performance parameters of the disease treatment solution based on the throughput.

8. The system for optimizing the repair strategy for improving rail transit network resilience according to claim 6 is characterized in that: The optimal solution selection module includes: A parameter query unit, used to query the network performance parameters of the disease treatment solution corresponding to each disease reporting request for any type of disease reporting request; Parameter comparison unit, used to compare network performance parameters and select the optimal network performance parameters; The solution query unit is used to query the disease treatment solution corresponding to the optimal network performance parameters as the optimal solution.

Images