Emergency Dispatch Methods and Systems for Urban Rail Transit Systems Combining Big Data Analysis

Abstract

This invention provides an emergency dispatching method and system for urban rail transit systems that combines big data analysis, relating to the field of urban rail transit technology. First, it aggregates dynamic monitoring data of the urban rail transit system, including real-time train operating conditions, passenger flow distribution in station areas, track structure vibration, power supply network load, and external environmental interference data. Then, based on this data, it constructs an anomaly evolution topology, using data fluctuation characteristics as nodes and connecting them into a network through evolutionary correlation strength reflecting mutual induced relationships. Next, it identifies key transmission nodes and diffusion boundaries from the anomaly evolution topology. Then, it combines key transmission nodes, diffusion boundaries, and emergency resource layout data to generate a collaborative dispatching matrix containing multiple sequences, including resource pre-scheduling, train route reconstruction, station load evacuation, and system function compensation, linked through time-series correlation parameters. Finally, it converts the collaborative dispatching matrix into an executable dispatching instruction set and pushes it to the corresponding control terminal according to the executing entity, achieving efficient emergency dispatching.

Description

Technical Field

[0001] This invention relates to the field of urban rail transit technology, and more specifically, to an emergency dispatching method and system for urban rail transit systems that incorporates big data analysis. Background Technology

[0002] The operation of urban rail transit systems faces numerous complex and uncertain factors that can trigger various emergencies, thereby affecting the normal operation of the system and passenger safety. Traditional emergency dispatch methods for urban rail systems mainly rely on manual experience and pre-set fixed emergency plans. However, urban rail systems are large and dynamically changing complex systems, and their operational status is affected in real time by a variety of factors. For example, the real-time operating conditions of trains may become abnormal due to equipment failure or improper operation; the distribution of passenger flow in station areas may change drastically due to special events or emergencies; track structure vibration may become abnormal due to changes in geological conditions or the impact of train operation; the load on the power supply network may fluctuate due to peak electricity consumption or equipment failure; and external environmental interferences such as severe weather or foreign object intrusion can also affect the urban rail system.

[0003] Existing emergency dispatch methods, lacking comprehensive aggregation and in-depth analysis of dynamic monitoring data from urban rail transit systems, struggle to accurately grasp the complex relationships and dynamic evolution patterns among these factors. This makes it difficult to promptly identify key factors and propagation paths that may trigger systemic risks. Emergency dispatch decisions often address only single issues, lacking a holistic consideration of the entire system. This results in a lack of coordination and foresight in emergency dispatch measures, making it difficult to effectively respond to complex and ever-changing emergencies and ensuring the safe and efficient operation of the urban rail transit system in emergency situations. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide an emergency dispatching method for urban rail transit systems that incorporates big data analysis, the method comprising:

[0005] The system collects dynamic monitoring data of the urban rail transit system, which includes real-time train operating data, passenger flow distribution data in station areas, track structure vibration data, power supply network load data, and external environmental interference data.

[0006] An anomaly evolution topology is constructed based on the dynamic monitoring data of the urban rail system. The anomaly evolution topology uses data fluctuation characteristics as nodes and connects the nodes through evolutionary correlation strength to form a network. The evolutionary correlation strength reflects the mutual induction relationship between different data fluctuation characteristics over time.

[0007] Key transmission nodes and diffusion boundaries are identified from the anomalous evolutionary topology. The key transmission nodes are nodes whose evolutionary association strength exceeds the trigger threshold and are located at the center of the topology. The diffusion boundary is the boundary line formed by the set of nodes in the topology whose evolutionary association strength does not meet the transmission condition.

[0008] Combining the key transmission nodes, diffusion boundaries and emergency resource layout data of the urban rail system, a multi-sequence collaborative scheduling matrix is ​​generated. The multi-sequence collaborative scheduling matrix includes resource pre-scheduling sequence, train route reconstruction sequence, station load evacuation sequence and system function compensation sequence. Each sequence is linked through time-series correlation parameters.

[0009] The multi-sequence collaborative scheduling matrix is ​​converted into an executable scheduling instruction set, which is then pushed to the corresponding control terminal of the urban rail system according to the execution subject.

[0010] Furthermore, embodiments of the present invention also provide an emergency dispatch system for urban rail transit systems that incorporates big data analytics, characterized by comprising:

[0011] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described emergency dispatching method for urban rail transit systems incorporating big data analysis by executing the machine-executable instructions.

[0012] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of a computer device reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the computer device to execute the above-described emergency dispatching method for urban rail transit systems combined with big data analysis.

[0013] Based on the above, by comprehensively aggregating dynamic monitoring data of the urban rail system, covering various aspects such as real-time train operating conditions, passenger flow distribution in station areas, track structure vibration, power supply network load, and external environmental interference, and then constructing an anomaly evolution topology based on these data, using data fluctuation characteristics as nodes, and connecting nodes to form a network by reflecting the evolutionary correlation strength of different data fluctuation characteristics over time, it is possible to reveal the complex correlation and dynamic evolution law between various factors in the urban rail system under emergency conditions. Next, by identifying key transmission nodes and diffusion boundaries from the abnormal evolution topology, we can accurately locate the key factors and propagation paths that may trigger systemic risks. Combining key transmission nodes, diffusion boundaries, and emergency resource layout data of the urban rail system, we generate a multi-sequence collaborative scheduling matrix, which includes multiple sequences such as resource pre-scheduling, train route reconstruction, station load evacuation, and system function compensation. Through time-series correlation parameters, we achieve linkage and convert the multi-sequence collaborative scheduling matrix into an executable scheduling instruction set, which is then pushed to the corresponding control terminal of the urban rail system according to the execution subject. This enables rapid and accurate emergency scheduling, effectively responds to complex and ever-changing emergencies, ensures the safe and efficient operation of the urban rail system in emergency situations, and significantly improves the emergency management capabilities of the urban rail system. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the execution flow of the emergency dispatching method for urban rail transit systems that combines big data analysis, provided in an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of exemplary hardware and software components of an urban rail transit emergency dispatch system that combines big data analysis, provided in an embodiment of the present invention. Detailed Implementation

[0016] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating an emergency dispatching method for urban rail transit systems that incorporates big data analysis, provided in one embodiment of the present invention. The following is a detailed description of this emergency dispatching method for urban rail transit systems that incorporates big data analysis.

[0017] Step S110: Collect dynamic monitoring data of the urban rail system, which includes real-time train operating data, passenger flow distribution data in station areas, track structure vibration data, power supply network load data, and external environmental interference data.

[0018] During the daily operation of the urban rail transit system, it is necessary to continuously collect and aggregate various data related to system operation. Real-time train operating data is acquired through sensors installed on key train components, such as the traction system, braking system, and bogies, which collect parameters such as motor speed, brake pressure, and wheelset temperature. Passenger flow distribution data in station areas is obtained through cameras deployed at platforms, concourses, and entrances / exits, combined with image analysis technology, reflecting the density and flow of people in different areas. Track structure vibration data is collected by vibration sensors fixed to the track bed and rails to monitor track settlement and deformation. Power supply network load data comes from smart meters, switchgear, and other equipment in the power supply system, recording information such as current, voltage, and power in each power supply section. External environmental interference data includes wind speed, rainfall, and temperature data collected by meteorological stations along the track, as well as interference signal monitoring data from external communication systems. All of the above data are aggregated to the central data processing center through the urban rail transit system's dedicated data transmission network.

[0019] During the data collection process, for information involving passenger privacy, such as image data captured by cameras, image blurring technology is used to process sensitive information such as faces. At the same time, encrypted transmission protocols are used during data transmission, and sensitive data is de-identified during storage to protect passenger privacy and prevent data leakage.

[0020] Step S120: Construct an anomaly evolution topology based on the dynamic monitoring data of the urban rail system. The anomaly evolution topology uses data fluctuation characteristics as nodes and connects the nodes through evolution correlation strength to form a network. The evolution correlation strength reflects the mutual induction relationship between different data fluctuation characteristics over time.

[0021] After acquiring dynamic monitoring data from the urban rail transit system, this data needs to be analyzed and processed to construct an anomaly evolution topology. This anomaly evolution topology can demonstrate the interrelationships between different data fluctuation characteristics, helping to identify the propagation paths and key nodes of anomalies. The construction process first extracts fluctuation characteristics from the data, then identifies nodes, calculates the evolutionary correlation strength between nodes, and thus forms the network topology, which is then dynamically updated based on new data.

[0022] Step S121: Extract fluctuation features from the dynamic monitoring data of the urban rail system, and take the segments in each type of data that exceed the normal fluctuation range as independent data fluctuation features. Each data fluctuation feature includes the fluctuation start time, fluctuation peak value and fluctuation duration.

[0023] For various types of dynamic monitoring data from urban rail transit systems aggregated at the central data processing center, fluctuation feature extraction is required. First, for each type of data, a normal fluctuation range is determined based on its historical performance during normal operation. For example, the normal fluctuation range of motor speed in real-time train operating data varies depending on the train's operating phase (e.g., starting, accelerating, constant speed, deceleration, stopping, etc.). Statistical analysis of historical data can determine the normal fluctuation intervals for different operating phases. When the real-time collected data exceeds this normal fluctuation range, it is considered that data fluctuation has occurred. Each data segment exceeding the normal range is treated as an independent data fluctuation feature, and the fluctuation start time, peak value, and duration are recorded. For example, if the train motor speed suddenly increases and exceeds the normal range at a certain moment, the data segment from that moment until the speed returns to normal is a data fluctuation feature, where the start time is the fluctuation start time, the peak speed is the fluctuation peak value, and the time from the start to the return to normal is the fluctuation duration.

[0024] Step S122: Assign a unique identifier to each data fluctuation feature as the basic node of the abnormal evolution topology. The attribute information of the basic node includes the type of corresponding data, the collection location, and the fluctuation amplitude.

[0025] After extracting the data fluctuation characteristics, a unique identifier is assigned to each data fluctuation characteristic. This identifier can be a string composed of letters and numbers, used to uniquely distinguish different fluctuation characteristics. These data fluctuation characteristics with unique identifiers will serve as the basic nodes of the anomaly evolution topology. Simultaneously, each basic node also needs to contain rich attribute information, including the type of corresponding data, such as real-time train operating data, station area pedestrian flow distribution data, etc.; the collection location, i.e., the specific location in the urban rail system from which the data was collected, such as a specific motor of a train, a specific platform area of ​​a station, etc.; and the fluctuation amplitude, i.e., the degree to which the data fluctuation exceeds the normal range, which can be determined by calculating the difference between the fluctuation peak and the upper or lower limit of the normal range.

[0026] Step S123: Calculate the time-induced coefficient of any two basic nodes. The time-induced coefficient is the reciprocal of the difference between the fluctuation start time of the later basic node and the fluctuation peak time of the earlier basic node. The smaller the difference, the larger the time-induced coefficient.

[0027] For any two basic nodes in an anomalous evolution topology, it is necessary to calculate their time-induced coefficients. First, determine the order in which the fluctuations of the two basic nodes occur, i.e., which basic node fluctuates first and which fluctuates later. The basic node that fluctuates first has a peak fluctuation time, and the basic node that fluctuates later has a fluctuation start time. Calculate the difference between these two time points, i.e., the fluctuation start time of the later node minus the fluctuation peak time of the earlier node. Then, take the reciprocal of this difference; the result is the time-induced coefficient between the two basic nodes. The shorter the time interval between the fluctuations of the two basic nodes, i.e., the smaller the difference, the larger the time-induced coefficient, indicating that the later fluctuation is more likely to be induced by the earlier fluctuation in time.

[0028] Step S124: Calculate the influence correlation coefficient between any two basic nodes. The influence correlation coefficient is the physical correlation degree of the data types corresponding to the two basic nodes. The physical correlation degree is determined based on the coupling relationship of the urban rail system equipment. The tighter the coupling relationship, the greater the influence correlation coefficient.

[0029] The influence correlation coefficient measures the degree of physical correlation between the data types corresponding to two basic nodes. Urban rail transit systems are complex systems with close coupling relationships between various devices. For example, there is a direct physical correlation between the train's traction system and power supply system; the operation of the traction system depends on the electricity provided by the power supply system, thus their data types have a high degree of physical correlation. However, the physical correlation between the train's air conditioning system data and the track structure vibration data is relatively low. By analyzing the coupling relationships between urban rail transit system devices, corresponding physical correlation values ​​are assigned to different data type combinations. The larger the physical correlation value, the closer the coupling relationship between the two data types, and the larger the corresponding influence correlation coefficient.

[0030] Step S125: Multiply the time-induced coefficient and the influence correlation coefficient to obtain the evolutionary correlation strength. The larger the value of the evolutionary correlation strength, the more significant the mutual induced relationship between the two basic nodes.

[0031] After obtaining the time-induced coefficient and the influence correlation coefficient, multiplying these two coefficients yields the evolutionary correlation strength between the two basic nodes. The evolutionary correlation strength comprehensively considers both the temporal induced relationship and the physical degree of correlation, providing a more complete reflection of the mutual induced relationship between two data fluctuation features. A larger value indicates a more significant mutual induced relationship between the two basic nodes, meaning that one data fluctuation feature is more likely to induce the occurrence of the other.

[0032] Step S126: When the evolutionary association strength is greater than the association threshold, a directed connection edge is established between the two basic nodes. The direction of the connection edge is determined by the order of the start time of the fluctuations of the two basic nodes, from the node that occurs first to the node that occurs later.

[0033] A correlation threshold is set, determined based on historical anomaly data of the urban rail system and expert experience. For any two basic nodes, if the calculated evolutionary correlation strength is greater than the correlation threshold, a significant mutual induction relationship is considered to exist between the two nodes, and a directed connection edge is established between them. The direction of the connection edge is determined by the fluctuation start time of the two basic nodes, pointing from the node that fluctuated first to the node that fluctuated later, to represent the direction of anomaly propagation.

[0034] Step S127: Construct the network topology for all basic nodes and directed connections to form the initial abnormal evolution topology.

[0035] After identifying all the basic nodes and the directed edges that meet the conditions, the network topology is constructed according to the relationship between nodes and edges. Each basic node is treated as a point in the topology graph, and the directed edges are drawn on the graph according to their directions and the nodes they connect, thus forming the initial anomaly evolution topology. This anomaly evolution topology graph can intuitively show the correlation and propagation path between abnormal fluctuation characteristics in the dynamic monitoring data of the urban rail system.

[0036] Step S128: At each preset update interval, newly extracted data fluctuation features are incorporated as new nodes, evolutionary association strength is recalculated and directed connection edges are updated to achieve dynamic expansion of the abnormal evolution topology.

[0037] To ensure the anomaly evolution topology reflects the latest anomalies in the urban rail system, it needs to be dynamically updated. A preset update cycle is set, for example, every ten minutes. Within each update cycle, fluctuation characteristics are extracted from newly collected dynamic monitoring data of the urban rail system, and these newly extracted data fluctuation characteristics are added as new nodes to the existing anomaly evolution topology. Then, the evolutionary association strength between new nodes and existing nodes, as well as between existing nodes, is recalculated. Directed edges are updated based on the new evolutionary association strength values. New edges are established for node pairs with an evolutionary association strength greater than the association threshold, while existing edges are removed for node pairs with an evolutionary association strength no longer greater than the threshold. This allows for the dynamic expansion and adjustment of the anomaly evolution topology.

[0038] Step S130: Identify key transmission nodes and diffusion boundaries from the anomalous evolutionary topology. The key transmission nodes are nodes whose evolutionary association strength exceeds the trigger threshold and are located at the center of the topology. The diffusion boundaries are the boundary lines formed by the set of nodes in the topology whose evolutionary association strength has not reached the transmission condition.

[0039] After constructing the anomaly evolution topology, it is necessary to identify key propagation nodes and diffusion boundaries. Key propagation nodes play a crucial role in the propagation of anomalies and are key targets for emergency dispatch. Diffusion boundaries define the scope of the anomaly's impact. The identification process includes calculating and analyzing parameters such as the number of connected edges and topological centrality of nodes.

[0040] Step S131: Traverse all basic nodes in the abnormal evolution topology, and count the number of incoming edges and outgoing edges for each basic node. The number of incoming edges is the number of directed connection edges pointing to the basic node, and the number of outgoing edges is the number of directed connection edges pointing from the basic node.

[0041] The process iterates through all basic nodes in the constructed anomalous evolution topology. For each basic node, the number of incoming and outgoing edges is counted. The number of incoming edges refers to the number of directed connections from other nodes to that node, reflecting the degree to which that node is affected by the anomalous behavior of other nodes. The number of outgoing edges refers to the number of directed connections from that node to other nodes, reflecting the node's ability to induce anomalous behavior in other nodes. By counting the number of incoming and outgoing edges, a preliminary understanding of the connectivity of each node in the topology can be obtained.

[0042] Step S132: Calculate the topological centrality of each basic node. The topological centrality is the ratio of the sum of the number of incoming edges and outgoing edges to the average number of connected edges of all basic nodes in the topology. The larger the ratio, the closer the basic node is to the topological center.

[0043] First, calculate the average number of connected edges among all basic nodes in the anomaly evolution topology, which is the average of the sum of the number of incoming edges and outgoing edges of all nodes. Then, for each basic node, divide the sum of its incoming and outgoing edges by the average number of connected edges; the result is the topological centrality of that node. The larger the ratio of topological centrality, the more tightly connected the node is in the topology, the closer it is to the center of the topology, and the more important its role may be in the anomaly propagation process.

[0044] Step S133: Extract the evolutionary association strength corresponding to the outgoing edges of each basic node. If the evolutionary association strength of at least one outgoing edge exceeds the trigger threshold, then mark the basic node as a potential transmission node.

[0045] For each basic node, extract the evolutionary correlation strength values ​​corresponding to all its outgoing edges. A trigger threshold is set, determined based on historical data of anomaly propagation and the safety requirements of the urban rail system. If at least one evolutionary correlation strength value in the outgoing edges of a basic node exceeds this trigger threshold, it indicates that the node has a strong ability to induce anomalies in other nodes, and it is marked as a potential propagation node.

[0046] Step S134: Select the top K nodes in terms of topological centrality from the potential transmission nodes and identify them as key transmission nodes.

[0047] After identifying potential transmission nodes, these nodes are sorted according to their topological centrality. The top K nodes with the highest topological centrality are selected as key transmission nodes. The value of K can be adjusted according to the scale of the urban rail system and the needs of emergency response. Generally, it is selected to reflect the number of nodes that can reflect the main anomaly propagation path. These key transmission nodes are the most core nodes in the anomaly evolution topology and play a crucial role in promoting the propagation and spread of the anomaly.

[0048] Step S135: Collect all base nodes in the anomalous evolution topology that have not been marked as potential transmission nodes as boundary candidate nodes.

[0049] In the anomalous evolution topology, apart from the basic nodes marked as potential propagation nodes, the remaining basic nodes do not have strong anomalous propagation capabilities because the evolutionary association strength of their outgoing edges does not exceed the trigger threshold. These nodes are collected as boundary candidate nodes, and they may be located in the edge region of anomalous influence.

[0050] Step S136: Calculate the shortest path hop count between the boundary candidate node and the key transmission node, where the shortest path hop count is the minimum number of directed connection edges connecting the two nodes.

[0051] For each boundary candidate node, calculate the shortest path hop count between it and each key transmission node. The shortest path hop count is the minimum number of directed edges traversed from the boundary candidate node to the key transmission node. By calculating the shortest path hop count, the distance between the boundary candidate node and the key transmission nodes can be measured. The closer the distance, the greater the likelihood that the boundary candidate node is abnormally affected by the key transmission nodes.

[0052] Step S137: Mark the candidate boundary nodes whose shortest path hop count is greater than the boundary threshold as boundary nodes. The spatial distribution of all boundary nodes forms the initial outline of the diffusion boundary.

[0053] A boundary threshold is set, determined based on the layout of the urban rail system and the general laws governing anomaly propagation. For a candidate boundary node, if the number of hops of its shortest path to all key transmission nodes is greater than this boundary threshold, the node is considered to be far from the key transmission nodes and less affected by the anomaly, and is thus marked as a boundary node. The spatial distribution of all boundary nodes in the anomaly evolution topology will form a rough outline, i.e., the initial outline of the diffusion boundary.

[0054] Step S138: Perform edge smoothing on the initial contour and connect adjacent boundary nodes in spatial order to form a closed curve, which serves as the final diffusion boundary.

[0055] Since the initial contour may have irregular, jagged edges, it needs to be smoothed. This is achieved using algorithms such as moving averages and spline curve fitting. Then, adjacent boundary nodes are connected sequentially according to their actual spatial positions within the urban rail system, forming a closed curve. This curve is the final determined diffusion boundary. The diffusion boundary clearly delineates the extent of the anomaly's influence.

[0056] Step S139: Record the unique identifier of the key transmission node, the corresponding data fluctuation characteristics and topological centrality, and at the same time record the coordinate sequence of the closed curve of the diffusion boundary and the identifier of the boundary node included.

[0057] After identifying key transmission nodes and determining the diffusion boundary, relevant information needs to be recorded for subsequent emergency dispatch and analysis.

[0058] For example, step S1391: Create a node information record table for each key transmission node. The fields of the node information record table include a unique identifier, the start time of data fluctuation characteristics, the peak value of fluctuation, the duration of fluctuation, the corresponding data type, the collection location, and the topological centrality value.

[0059] A node information record table is established for each key transmission node. This node information record table contains multiple fields, including a unique identifier to distinguish different key transmission nodes; fields for data fluctuation characteristics such as fluctuation start time, fluctuation peak value, and fluctuation duration to record the specific parameters of the data fluctuation characteristics corresponding to the node; a corresponding data type field to indicate which type of urban rail system dynamic monitoring data the fluctuation characteristic belongs to; a data collection location field to record the data collection location; and a topological centrality value field to record the topological centrality of the node.

[0060] Step S1392: Arrange the node information record table of all key transmission nodes in descending order of topological centrality value to form a list of key transmission nodes.

[0061] The node information records of all key transmission nodes are arranged in descending order of topological centrality. Key transmission nodes with higher topological centrality values ​​are more prominent in the central position of the abnormal evolution topology and are therefore more important. This arrangement forms a list of key transmission nodes, facilitating priority handling during emergency dispatch.

[0062] Step S1393: Perform coordinate sampling on the closed curve of the diffusion boundary, collecting a coordinate point at a preset distance to form a closed curve coordinate sequence, which contains the three-dimensional spatial coordinates of each sampling point.

[0063] To accurately describe the location of the diffusion boundary, coordinate sampling of its closed curve is required. A preset distance is set, for example, one coordinate point is collected on the closed curve every meter. Each coordinate point is represented by three-dimensional spatial coordinates, including X, Y, and Z axis coordinate values, to accurately reflect its spatial position within the urban rail system. The coordinates of these sampled points are arranged sequentially to form a closed curve coordinate sequence.

[0064] Step S1394: Collect the unique identifiers of all boundary nodes contained in the diffusion boundary, arrange them in order of their positions on the closed curve, and form a list of boundary node identifiers.

[0065] Collect the unique identifiers of all boundary nodes contained within the diffusion boundary. Arrange these unique identifiers according to the actual positional order of these boundary nodes on the closed curve of the diffusion boundary, forming a boundary node identifier list. This boundary node identifier list reflects the distribution of boundary nodes on the diffusion boundary.

[0066] Step S1395: Associate the closed curve coordinate sequence with the boundary node identifier list, where each boundary node identifier corresponds to the coordinates of the nearest sampling point in the closed curve coordinate sequence.

[0067] To map the boundary nodes to the specific locations of the diffusion boundary, a sequence of closed curve coordinates is associated with a list of boundary node identifiers. For each boundary node identifier, the coordinates of the nearest sampling point to its actual location are found in the closed curve coordinate sequence, and a correspondence is established. In this way, the approximate location of a boundary node on the diffusion boundary can be quickly determined using its identifier.

[0068] Step S1396: Create a diffusion boundary information table. The fields of the diffusion boundary information table include the closed curve coordinate sequence, the boundary node identifier list, and the boundary formation time. The boundary formation time is the time when the diffusion boundary is finally determined.

[0069] Create a diffusion boundary information table to record relevant information about diffusion boundaries. The table's fields include: a closed curve coordinate sequence to fully describe the shape and location of the diffusion boundary; a list of boundary node identifiers to record the boundary nodes contained within the diffusion boundary; and the boundary formation time, i.e., the time when the diffusion boundary was finally determined, for subsequent tracing and analysis.

[0070] Step S1397: Integrate the list of key transmission nodes with the diffusion boundary information table to form an abnormal impact range report. The abnormal impact range report also includes statistics on the spatial distance between key transmission nodes and diffusion boundaries.

[0071] The previously generated list of key transmission nodes and diffusion boundary information table are integrated to form a complete anomaly impact range report. In addition to basic information on key transmission nodes and diffusion boundaries, this report also calculates the spatial distance between each key transmission node and the diffusion boundary to reflect the positional relationship between the key nodes and the anomaly impact range boundary.

[0072] Step S1398: Store the abnormal impact range report to the urban rail transit system emergency management database. The storage path is classified according to the report generation time and abnormal event type.

[0073] The generated reports on the scope of the anomalies are stored in the urban rail transit system's emergency management database. For ease of management and retrieval, the storage path is categorized according to the report's generation time and the type of the anomaly. For example, a folder structure is created in the database with year, month, and date as hierarchical levels, and each folder further stores the corresponding reports according to the type of anomaly (such as power supply anomaly, train malfunction, etc.).

[0074] Step S1399: When the abnormal evolution topology update causes changes in key transmission nodes or diffusion boundaries, regenerate the node information record table and diffusion boundary information table, update the abnormal impact range report, and push the storage path of the abnormal impact range report to the urban rail system emergency decision-making platform.

[0075] When the abnormal evolution topology is dynamically updated, if the number of key transmission nodes, topological centrality, or the shape and location of the diffusion boundary change, it is necessary to regenerate the node information record table for key transmission nodes and the diffusion boundary information table. Based on the new node information record table and diffusion boundary information table, the abnormal impact range report is updated. Then, the updated abnormal impact range report is pushed to the urban rail transit system's emergency decision-making platform, enabling emergency decision-makers to obtain the latest abnormal impact range information in a timely manner so as to adjust emergency dispatch strategies.

[0076] Step S140: Combine the key transmission nodes, diffusion boundaries and emergency resource layout data of the urban rail system to generate a multi-sequence collaborative scheduling matrix. The multi-sequence collaborative scheduling matrix includes resource pre-scheduling sequence, train route reconstruction sequence, station load evacuation sequence and system function compensation sequence. Each sequence is linked through time-series correlation parameters.

[0077] After identifying key transmission nodes and determining diffusion boundaries, a comprehensive emergency dispatch plan needs to be developed by combining emergency resource layout data of the urban rail transit system, generating a multi-sequence collaborative dispatch matrix. This multi-sequence collaborative dispatch matrix integrates various dispatch sequences and achieves linkage between sequences through time-series correlation parameters to ensure efficient and orderly emergency dispatch.

[0078] Step S141: Call the emergency resource layout data of the urban rail transit system. The emergency resource layout data of the urban rail transit system includes the distribution location of mobile emergency teams, emergency equipment storage sites, standby train stopping locations and emergency material reserve information. Each resource is marked with response time and coverage area.

[0079] Emergency resource layout data was retrieved from the urban rail transit system's database. This data details the distribution of various emergency resources within the system. The mobile emergency team locations indicate the current positions of each emergency repair team; emergency equipment storage sites record the specific locations where various emergency equipment (such as repair tools and spare parts) are stored; backup train stopping locations indicate the stations and tracks where backup trains stop; and emergency material reserve information includes the locations and quantities of various emergency supplies (such as medical supplies, food, and drinking water). Furthermore, each emergency resource is marked with its response time and coverage area. Response time indicates the time required for the resource to reach its designated location after receiving a dispatch order, and coverage area indicates the size of the area where the resource can effectively function.

[0080] Step S142: Spatial match the collection locations of key transmission nodes with the resource locations in the urban rail system emergency resource layout data to determine the types and quantities of emergency resources available for use around each key transmission node.

[0081] Acquire the location information of key transmission nodes, which typically includes the coordinates of specific equipment installation locations or monitoring points. Spatially match these coordinates with the location coordinates of various resources in the urban rail transit system's emergency resource layout data. Using Geographic Information System (GIS) technology, calculate the distances between key transmission nodes and each emergency resource location. Based on the distances and the coverage of emergency resources, determine the types and quantities of available emergency resources within a certain range around each key transmission node. For example, if a key transmission node is located near a station, spatial matching can identify the types and quantities of emergency equipment stored at that station, as well as information on mobile emergency response teams and emergency supply depots within a certain range of the station.

[0082] Step S143: Determine the processing priority based on the topological centrality of the key transmission node. The larger the topological centrality value, the higher the processing priority. Generate a resource pre-scheduling sequence according to the priority order. The resource pre-scheduling sequence includes resource type, scheduling starting point, target key transmission node location and scheduling path.

[0083] The topological centrality of key transmission nodes reflects their importance in the anomaly evolution topology. Key transmission nodes with higher topological centrality values ​​play a more critical role in the anomaly propagation process and require priority processing. Therefore, processing priorities are determined based on the topological centrality values ​​of key transmission nodes, from highest to lowest. Following this priority order, an emergency resource scheduling scheme is planned for each key transmission node, generating a resource pre-scheduling sequence. This resource pre-scheduling sequence details the type of resource to be scheduled (e.g., mobile emergency teams, emergency equipment), the scheduling starting point (the current location of the resource), the location of the target key transmission node (the destination of the resource), and the scheduling path (the route the resource takes from the scheduling starting point to the target location).

[0084] Step S1431: Select the node with the largest topological centrality value from the key transmission nodes as the first target node and mark it as the current target node.

[0085] When generating the resource pre-scheduling sequence, the node with the highest topological centrality value is first selected from the list of key propagation nodes. Since this node occupies the most central position in the abnormal evolution topology, its anomaly may spread rapidly and affect the entire system. Therefore, it is designated as the first target node that needs to schedule resources and is marked as the current target node.

[0086] Step S1432: Select emergency resources from the urban rail system emergency resource layout data that match the current target node spatially and have the smallest response time value, and form a resource combination. The resource combination includes at least three of the following: mobile emergency team, emergency equipment, backup train and emergency supplies.

[0087] Based on the current target node's location, emergency resources that spatially match that location are selected from the urban rail transit system's emergency resource layout data. During the selection process, in addition to considering whether the resource's coverage area includes the current target node's location, the response time of the resource is also a key focus. The emergency resource with the lowest response time value, i.e., the resource that can reach the current target node the fastest, is selected. These resources form a resource combination, which typically includes at least three of the following: a mobile emergency team, emergency equipment, a backup train, and emergency supplies, to ensure that various abnormal situations that may occur at the current target node can be addressed.

[0088] Step S1433: Extract the current storage location of each resource in the resource combination as the scheduling starting point. The coordinate information of the scheduling starting point is obtained from the emergency resource layout data of the urban rail system.

[0089] For each emergency resource in the resource portfolio, the coordinates of its current location are extracted from the urban rail transit system's emergency resource layout data, and this coordinate information is used as the starting point for scheduling that resource. For example, the starting point for scheduling a mobile emergency team is its current location coordinates, and the starting point for scheduling emergency equipment is the coordinates of its storage site, etc.

[0090] Step S1434: Starting from the dispatching starting point and the current target node position as the endpoint, plan the dispatching path in the internal channel map of the urban rail system. The dispatching path prioritizes emergency channels parallel to the track and avoids high-risk areas within the diffusion boundary.

[0091] On the internal corridor map of the urban rail system, the dispatching route is planned from the dispatching starting point to the current target node location. During the planning process, priority is given to emergency corridors parallel to the tracks. These corridors are typically dedicated to emergency repairs and resource transportation within the urban rail system, offering high traffic efficiency. Simultaneously, high-risk areas within the diffusion boundary must be avoided. These areas may have already been affected by anomalies, posing safety hazards, and resource entry may face risks or impact dispatching efficiency.

[0092] Step S1435: Record the resource type, scheduling start coordinates, current target node position coordinates, and key path points of the scheduling path for each resource in the resource combination, forming a resource pre-scheduling subsequence for the current target node.

[0093] The relevant information for each resource in the resource combination is recorded, including resource type, scheduling start coordinates, current target node location coordinates, and key points along the scheduling path (such as intersections and transfer points). This information is arranged in a specific order to form a resource pre-scheduling subsequence for the current target node. This resource pre-scheduling subsequence is a component of the resource pre-scheduling sequence and describes the specific resource scheduling arrangements required to handle anomalies at the current target node.

[0094] Step S1436: Remove the current target node from the critical transmission nodes, mark the node whose topological centrality value immediately follows as the new current target node, and repeat the above process of selecting resource combinations, planning scheduling paths and forming resource pre-scheduling subsequences.

[0095] After generating the resource pre-scheduling subsequence for the current target node, remove that node from the list of critical transit nodes. Then, mark the node with the next highest topological centrality value in the list as the new current target node. Process the new current target node following the same procedure as for the first target node, namely, selecting resource combinations, planning scheduling paths, and forming resource pre-scheduling subsequences.

[0096] Step S1437: After all critical transmission nodes have been processed, arrange all resource pre-scheduling sub-sequences according to the processing priority of the target nodes to form a resource pre-scheduling sequence.

[0097] Once all nodes in the list of critical transmission nodes have been processed, meaning each critical transmission node has generated its corresponding resource pre-scheduling subsequence, these subsequences are arranged according to the processing priority of the target nodes (i.e., in descending order of topological centrality). The arranged subsequences are then combined to form a complete resource pre-scheduling sequence.

[0098] Step S1438: Check whether there are overlapping segments in the scheduling paths of different resources in the resource pre-scheduling sequence. If so, adjust the scheduling path or departure time of one of the resources.

[0099] The generated resource pre-scheduling sequence is examined, with a focus on whether there are overlapping segments in the scheduling paths of different resources. If two or more scheduling paths overlap in a certain area, it may lead to congestion or conflicts during resource transportation. Once an overlapping segment is found, the scheduling path of one or more resources needs to be adjusted to select alternative feasible routes; or the departure time of the resources needs to be adjusted so that the resources pass through the overlapping area at different time periods to avoid conflicts and ensure that the resources can reach their destination smoothly.

[0100] Step S144: Analyze the distribution of track lines within the diffusion boundary, and combine the location of key transmission nodes to plan alternative track paths that avoid key transmission nodes. The alternative track paths must meet the curvature and gradient requirements for safe train operation.

[0101] The diffusion boundary defines the scope of the anomaly's impact, within which track lines may be affected. Therefore, a detailed analysis of the track distribution within the diffusion boundary is necessary to determine the direction, connections, and relative positions of each line to critical transmission nodes. Based on the location information of the critical transmission nodes, alternative track routes are planned to allow trains to avoid the track sections where these nodes are located. When planning alternative routes, it is crucial to ensure that the routes meet the curvature and gradient requirements for safe train operation; the curvature should not be too large, and the gradient should not be too steep, to guarantee smooth and safe train operation.

[0102] Step S145: Extract the real-time positions of all running trains within the diffusion boundary, sort them from shortest to longest distance between the trains and key transmission nodes, determine the path reconstruction order, and generate a train path reconstruction sequence. The train path reconstruction sequence includes the train number, the original running path, the alternative track path, and the path switching time.

[0103] To ensure trains operating within the diffusion boundary can safely avoid the affected area of ​​anomalies, their routes need to be reconstructed. First, the real-time location information of all trains operating within the diffusion boundary is extracted; this information can be obtained through the trains' Global Positioning System (GPS) or track circuits. Then, the distance between each train and the nearest critical transmission node is calculated, and the trains are sorted in ascending order of distance. Trains closer to the critical transmission node are at greater risk of being affected by the anomaly and require priority for route reconstruction. Based on the sorting results, the route reconstruction order is determined, alternative track paths are planned for each train, and the timing of path switching is determined, thus generating a train route reconstruction sequence.

[0104] Step S1451: Obtain the real-time location information of all running trains within the diffusion boundary, wherein the real-time location information is the precise mileage marker of the train on the track line.

[0105] The urban rail transit system's train monitoring system acquires real-time location information for all operating trains within the diffusion boundary. This location information is represented by precise mileage markers on the track, indicating the distance from the train's current location to the track's starting point. This representation accurately reflects the train's specific position on the track.

[0106] Step S1452: Calculate the straight-line distance between the real-time position of each train and the nearest critical transmission node. The straight-line distance is determined by the difference between the train's position mileage and the track mileage corresponding to the critical transmission node.

[0107] For each operating train, the difference between its real-time location mileage marker and the mileage marker of the track position corresponding to the nearest critical transmission node is calculated. This difference represents the straight-line distance between the train and the critical transmission node (distance on the track is approximately a straight-line distance). By comparing this distance value for different trains, the urgency of the abnormal impact on the train can be determined.

[0108] Step S1453: Sort the trains in ascending order of straight-line distance value, with the trains having the smaller straight-line distance value being prioritized for path reconstruction.

[0109] The calculated straight-line distance between each train and the nearest critical transmission node is compared, and the trains are sorted in ascending order. The smaller the straight-line distance, the closer the train is to the anomaly source, and the higher the risk it faces. Therefore, its path needs to be reconstructed first to get it out of the danger zone as soon as possible.

[0110] Step S1454: For trains with the same straight-line distance, sort them from largest to smallest according to their current running speed. Trains with larger current running speed values ​​have higher reconstruction priority.

[0111] When multiple trains have the same straight-line distance to the nearest critical transmission node, their path reconstruction priorities need to be further determined. In this case, trains are sorted from highest to lowest current speed. Trains with higher speeds travel a longer distance in the same amount of time; if path reconstruction is not performed promptly, they may approach the critical transmission node more quickly, thus their reconstruction priority should be higher.

[0112] Step S1455: Determine the path reconstruction order based on the sorting results, and mark the train ranked first as the first reconstruction object.

[0113] Based on the train order obtained from the above sorting process, the order of route reconstruction is determined. The train ranked first will be marked as the first object requiring route reconstruction, and its alternative route planning and route switching time determination will be prioritized.

[0114] Step S1456: Extract the original running path of the first reconstructed object. The original running path includes the train's starting station, intermediate stations, ending station, and track sections.

[0115] The original running path information of the first reconfigurable object is extracted from the train operation scheduling system. This original running path information records in detail the train's origin station, all planned stations along the route, the destination station, and the specific track sections traversed. Knowing the original running path is the basis for planning alternative routes, ensuring that the alternative routes can meet the train's operational needs.

[0116] Step S1457: Based on the diffusion boundary and the location of key transmission nodes, plan an alternative track path for the first reconstructed object. The alternative track path must include the same starting station and ending station as the original running path, and avoid all key transmission nodes in the track section.

[0117] When planning the alternative track path for the first reconfigurable object, the locations of the diffusion boundary and key transmission nodes are used as constraints. The alternative track path must include the same starting and ending stations as the original route to ensure that the train can ultimately reach its destination. Simultaneously, the track sections traversed must avoid areas containing all key transmission nodes to prevent impact from anomalies. During the planning process, factors such as track connectivity and signaling system compatibility need to be comprehensively considered to select the optimal alternative path.

[0118] Step S1458: Analyze the track occupancy of alternative track routes, select the time period when the track is idle as the route switching time, and the route switching time must meet the safety interval requirements with the preceding and following trains.

[0119] The planned alternative track routes are analyzed for track occupancy. By consulting train timetables and real-time dispatch information, the train operation status of the alternative routes at different time periods is obtained, and the time periods with track vacancy are selected as the route switching times. At the same time, the determination of the route switching time must meet the safety distance requirements with the preceding and following trains, that is, to ensure that after the train switches to the alternative route, a sufficient safe distance is maintained between the train and the preceding and following trains to prevent rear-end collisions or accidents.

[0120] Step S1459: Record the train number of the first reconstructed object, the track section sequence of the original running path, the track section sequence of the alternative track path, and the specific time of the path switching, to form the first train path reconstruction sub-sequence.

[0121] Record the relevant information of the first reconstructed object, including the train number (used to uniquely identify the train), the original track section sequence (listing the track sections passed through by the original route in order), the track section sequence of the alternative track route (listing the track sections passed through by the alternative route in order), and the specific time of the route switching (accurate to the minute or second). The above information is organized in a certain format to form the first train route reconstruction subsequence.

[0122] Step S14510: Remove the first reconstructed object from the sorting results, mark the trains that follow it in the sorting results as new reconstructed objects, and repeat the process of extracting the original path, planning the alternative path, and determining the switching time.

[0123] After generating the train path reconstruction subsequence for the first reconstruction object, remove that train from the sorting results. Then, mark the trains immediately following in the sorting results as new reconstruction objects. Process the new reconstruction objects using the same steps as for the first reconstruction object, namely, extracting the original running path, planning alternative track paths, analyzing track occupancy to determine the path switching time, recording information to form a train path reconstruction subsequence, etc.

[0124] Step S14511: After all trains have completed the generation of train route reconstruction sub-sequences, arrange all train route reconstruction sub-sequences in the reconstruction order to form a train route reconstruction sequence.

[0125] Once all trains requiring route reconfiguration within the diffusion boundary have generated their corresponding train route reconfiguration subsequences, these subsequences are arranged according to the route reconfiguration order (i.e., the previously determined train sorting order). The arranged subsequences are then combined to form a complete train route reconfiguration sequence, which describes the route adjustment arrangements for all trains.

[0126] Step S14512: Verify the difference between the total length of the alternative track path in the train path reconstruction sequence and the length of the original running path. If the difference exceeds the preset range, replan the alternative track path.

[0127] The generated train route reconstruction sequence is validated, with a focus on the difference between the total length of the alternative track path and the original route. A preset range for this difference is established, determined based on factors such as train travel time requirements and passenger travel experience. If the difference exceeds this preset range, it indicates that the alternative path may be too long or too short, potentially affecting train punctuality or increasing passenger travel time. In this case, a new alternative track path needs to be planned for the corresponding train to ensure the difference remains within a reasonable range.

[0128] Step S146: For stations within the diffusion boundary, calculate the station congestion index based on the pedestrian flow distribution data, and determine the evacuation priority by sorting the congestion index values ​​from largest to smallest. The station congestion index is the ratio of the number of people on the station in real time to the number of people the station is designed to accommodate.

[0129] Sites within the diffusion boundary may experience congestion due to abnormal impacts, necessitating load evacuation. First, a congestion index is calculated for each site based on pedestrian flow distribution data. The site congestion index is defined as the ratio of real-time on-site personnel to the site's designed capacity, directly reflecting the degree of congestion. Then, sites are sorted in descending order of congestion index values; sites with higher congestion index values ​​indicate more severe congestion and require higher evacuation priority.

[0130] Step S147: Construct station load evacuation paths based on evacuation priorities and generate station load evacuation sequences. The station load evacuation paths include entrance / exit selection, transfer channel guidance, and platform diversion directions. The station load evacuation sequences include station names, evacuation path planning, guidance equipment operation instructions, and evacuation completion time.

[0131] Based on the evacuation priority of each station, a load evacuation route is constructed for each station, and a station load evacuation sequence is generated. The evacuation route planning takes into account factors such as entrance and exit selection, transfer passage guidance, and platform diversion direction. Guidance equipment operation commands are used to control the guidance equipment within the station, and the evacuation completion time is estimated to be the time required for the evacuation work.

[0132] Step S1471: Select the station with the highest congestion index value from the stations within the diffusion boundary as the first evacuation target and mark it as the current evacuation station.

[0133] After determining the evacuation priorities of the stations, the station with the highest congestion index value is selected from those within the diffusion boundary. This station has the most severe congestion and requires evacuation first; it is then marked as the current evacuation station.

[0134] Step S1472: Obtain the internal structure data of the current evacuation site. The internal structure data includes the number and location of entrances and exits, the distribution of transfer passages, the platform layout, and the location of stairs and elevators.

[0135] The internal structure data of current evacuation stations is obtained from the urban rail transit system's station database. This data details the station's internal layout, including the number and specific locations of entrances and exits (e.g., which street and direction they are located on), the distribution of transfer passages (the direction and location of passages connecting different lines), the layout of platforms (e.g., island platforms, side platforms), and the location of stairs and elevators. Knowing the station's internal structure is a prerequisite for constructing a reasonable evacuation route.

[0136] Step S1473: Identify the target congested area within the current evacuation site based on the pedestrian flow distribution data. The target congested area is the area where the population density exceeds the safe density threshold.

[0137] Based on the current pedestrian flow distribution data at evacuation sites, the distribution of people within the sites is analyzed. A safe density threshold is established, determined based on fire safety regulations and practical experience in personnel evacuation. Areas where the pedestrian density exceeds this safe density threshold are identified as target congestion areas, which are the focus of evacuation efforts.

[0138] Step S1474: Construct evacuation route planning, planning multiple evacuation routes from the target congested area to each entrance and exit, each evacuation route including transfer passages, stairs and elevators, and exit signs.

[0139] For the identified target congestion areas, evacuation route planning begins. Starting from each target congestion area, multiple evacuation routes are planned towards each entrance and exit of the station. Each evacuation route details the transfer passages, stairs and elevators, and exit signs. Planning multiple routes helps avoid overcrowding during evacuation, improving efficiency. During the planning process, factors such as route length, traffic capacity, and the presence of obstacles must be considered to select the optimal evacuation route.

[0140] Step S1475: Generate guidance equipment operation instructions based on evacuation route planning. The guidance equipment operation instructions include the direction information displayed on the sign, the evacuation guidance content broadcast by the broadcast system, the opening status of the turnstile, and the running direction of the elevator.

[0141] Based on the established evacuation route plan, corresponding guidance equipment operation instructions are generated. The directional information displayed on the signs should be consistent with the direction of the evacuation route to guide passengers to evacuate in the correct direction; the evacuation guidance broadcast by the public address system should be clear and accurate, informing passengers of the evacuation route, precautions, etc.; the opening status of the turnstiles should be controlled according to evacuation needs, such as opening all turnstiles during peak evacuation periods or setting up dedicated evacuation passage turnstiles; the operating direction of elevators should be adjusted to facilitate evacuation, such as all escalators running upwards or downwards, and elevators prioritizing stops at floors with high population density.

[0142] Step S1476: Calculate the evacuation completion time based on the real-time number of people at the current evacuation site and the total capacity of the evacuation routes. The total capacity is the total number of people that can be evacuated per unit time for all evacuation routes.

[0143] The real-time number of people at current evacuation sites can be obtained from pedestrian flow data. The total capacity of evacuation routes refers to the total number of people that can be evacuated per unit of time. This number is calculated based on factors such as route width and walking speed. Dividing the real-time number of people at each site by the total capacity yields the approximate evacuation time, i.e., the evacuation completion time. Estimating the evacuation completion time helps in the rational planning of evacuation schedules and resource allocation.

[0144] Step S1477: Record the site name, evacuation route plan, guidance equipment operation instructions and evacuation completion time of the current evacuation site to form the first site load evacuation subsequence.

[0145] Record relevant information about the current evacuation site, including site name, evacuation route planning (detailed description of each evacuation route), guidance equipment operation instructions (specific operation requirements for various guidance equipment), and evacuation completion time (estimated evacuation end time). Organize the above information together to form the first site load evacuation subsequence.

[0146] Step S1478: Remove the current evacuation site from the sites within the diffusion boundary, mark the site with the next highest congestion index value as the new current evacuation site, and repeat the above process of building path, generating instructions, and calculating time.

[0147] After generating the site load evacuation subsequence for the current evacuation site, remove that site from the list of sites within the diffusion boundary. Then, mark the site with the next highest congestion index value in the list as the new current evacuation site. Process the new current evacuation site using the same steps as for the first evacuation site: acquiring internal structure data, identifying target congestion areas, constructing evacuation route plans, generating guidance equipment operation instructions, calculating evacuation completion time, and forming the site load evacuation subsequence.

[0148] Step S1479: After all stations have completed the generation of station load evacuation sub-sequences, arrange all station load evacuation sub-sequences according to evacuation priority to form station load evacuation sequences.

[0149] Once all sites requiring evacuation within the diffusion boundary have generated their corresponding site load evacuation subsequences, these subsequences are arranged according to evacuation priority (i.e., congestion index values ​​from largest to smallest). The arranged subsequences are then combined to form a complete site load evacuation sequence.

[0150] Step S148: Identify the affected urban rail system functional modules within the diffusion boundary, determine the replaceable functional modules based on system redundancy design, and generate a function switching scheme. The function switching scheme includes the switching order of primary and backup modules, data synchronization method, and switching verification steps.

[0151] Urban rail transit systems consist of multiple functional modules, such as train control systems, signaling systems, and power supply systems. When an anomaly occurs, some functional modules within the diffusion boundary may be affected, leading to functional failure or performance degradation. It is necessary to identify these affected functional modules and, based on the system's redundancy design, determine alternative functional modules and generate a function switching plan to ensure the normal operation of the urban rail transit system's core functions. The function switching plan includes the primary / backup module switching sequence, data synchronization methods, and switching verification steps.

[0152] Step S149: Determine the compensation order according to the importance of the functional modules and generate a system function compensation sequence. The greater the impact of a functional module on the safe operation of the urban rail system, the higher it is ranked. The system function compensation sequence includes the functional module name, the substitute module identifier, the switching operation steps, and the function recovery standard.

[0153] Different urban rail transit system functional modules have varying degrees of impact on system safety. Therefore, they need to be prioritized according to their importance to determine the order of compensation. Functional modules with higher importance have a greater impact on system safety and are prioritized for compensation operations. A system function compensation sequence is generated based on this order. This sequence includes information such as the functional module name, the identifier of the replacement module, the switching operation steps, and the function recovery criteria, guiding the system function compensation work.

[0154] Step S1491: Identify the urban rail system functional modules affected by key transmission nodes within the diffusion boundary. The urban rail system functional modules include train control system, signaling system, power supply system, communication system, and gate control system.

[0155] By analyzing the data types and acquisition locations corresponding to key transmission nodes, the functional modules of the urban rail transit system that may be affected within the diffusion boundary can be identified. These functional modules typically include train control systems, signaling systems, power supply systems, communication systems, and turnstile control systems. For example, if the data corresponding to key transmission nodes indicates abnormal parameters in the train control system, then the train control system may be an affected functional module.

[0156] Step S1492: Match a corresponding replacement module for each affected urban rail system functional module. The replacement module is a backup module with the same or similar functions in the urban rail system. The location of the replacement module must be outside the diffusion boundary or not affected by the key transmission node.

[0157] Urban rail transit systems are typically designed with redundancy in mind, equipping critical functional modules with backup modules. For each affected functional module, a corresponding replacement module needs to be matched from the system. The replacement module should have the same or similar functions as the original module, capable of taking over its operation when the original module fails. Simultaneously, the replacement module must be located outside the diffusion boundary or unaffected by critical transmission nodes to ensure its proper functioning.

[0158] Step S1493: Determine the importance ranking based on the degree of impact of the urban rail system functional modules on the safe operation of the urban rail system. The greater the value of the impact, the higher the ranking. The impact values ​​of the train control system and the signaling system are greater than those of other systems.

[0159] The various functional modules of the urban rail transit system are analyzed to assess their impact on the system's safe operation. The degree of impact can be determined numerically through expert evaluation, historical fault data analysis, and other methods. Functional modules with higher impact values ​​are more important and ranked higher in the compensation order. Generally, the train control system and signaling system are crucial to the safe operation of the urban rail transit system, and their impact values ​​are greater than those of other systems such as the power supply system, communication system, and gate control system.

[0160] Step S1494: Arrange the affected functional modules in order of importance from front to back, determine the compensation order, and prioritize the compensation operation for the functional module with the highest priority.

[0161] Based on the importance ranking of functional modules, the affected functional modules are arranged in descending order to determine the compensation order. The affected functional modules ranked highest will be prioritized for compensation operations to restore the system's critical functions as quickly as possible.

[0162] Step S1495: Select the first functional module from the sorting results as the first compensation object and mark it as the current compensation module.

[0163] Select the first functional module from the results of the compensation order and mark it as the first object that needs to be compensated, i.e., the current compensation module.

[0164] Step S1496: Generate the switching operation steps for the current compensation module. The switching operation steps include the shutdown process of the original module, the startup process of the replacement module, the data transmission process from the original module to the replacement module, and the interface docking process of the module.

[0165] Detailed switching procedures are established for the current compensation module. The shutdown procedure for the original module specifies how to safely stop its operation, avoiding data loss or system failure; the startup procedure for the replacement module describes how to start the replacement module and put it into working condition; the data transmission procedure from the original module to the replacement module ensures that critical data is accurately and completely transmitted to the replacement module, guaranteeing that the replacement module can take over the operation of the original module; the module interface docking procedure involves the interface connection between the replacement module and other modules in the system, ensuring normal interaction of data and control signals.

[0166] Step S1497: Set the function recovery standard, which is the range that the operating parameters of the replacement module need to reach, including response time, data transmission accuracy and control precision.

[0167] Functional recovery criteria are used to measure whether a replacement module has successfully taken over the functions of the original module. These criteria define the ranges that the replacement module's operating parameters must meet, such as a response time less than a certain value, a data transmission accuracy greater than a certain percentage, and control precision within a certain error range. Functional compensation is considered successful only when the replacement module's operating parameters meet these criteria.

[0168] Step S1498: Record the functional module name of the current compensation module, the unique identifier of the replacement module, the detailed sequence of switching operation steps, and the specific parameter range of the function recovery standard to form the first system function compensation sub-sequence.

[0169] Record the relevant information of the current compensation module, including the functional module name, the unique identifier of the replacement module (used to distinguish different replacement modules), the detailed sequence of switching operation steps (listing the operation content in order of step number), and the specific parameter range of the function recovery standard (the upper and lower limits or target values ​​of each parameter). Organize the above information together to form the first system function compensation subsequence.

[0170] Step S1499: Remove the current compensation module from the sorting results, mark the next functional module in the sorting results as the new current compensation module, and repeat the above process of matching replacement modules, generating steps, and setting standards.

[0171] After generating the system function compensation sub-sequence for the current compensation module, remove that module from the sorting results. Then, mark the next functional module in the sorting results as the new current compensation module. Process the new current compensation module following the same steps as processing the first compensation module, namely, matching replacement modules, generating switching operation steps, setting function recovery criteria, and forming the system function compensation sub-sequence.

[0172] Step S14910: After all affected functional modules have completed the generation of system function compensation sub-sequences, arrange all system function compensation sub-sequences in the compensation order to form a system function compensation sequence. Verify whether the total processing capacity of the replacement modules in the system function compensation sequence meets the operation requirements of the urban rail transit system. If it does not meet the requirements, increase the number of replacement modules or adjust the compensation order.

[0173] Once all affected functional modules have generated corresponding system function compensation sub-sequences, these sub-sequences are arranged in compensation order to form a system function compensation sequence. This sequence is then verified to check whether the total processing capacity of the replacement modules can meet the operational requirements of the urban rail transit system. If the total processing capacity is insufficient, it may be necessary to increase the number of replacement modules or adjust the compensation order, prioritizing the compensation of functional modules with higher processing capacity requirements.

[0174] Step S1410: Set timing association parameters, which are the time interval requirements of operation steps in different sequences. The time interval between the resource pre-scheduling sequence and the train route reconstruction sequence is such that the train can pass only after the resources are in place. The time interval between the train route reconstruction sequence and the station load evacuation sequence is such that the evacuation can only be started after the train is adjusted.

[0175] To enable coordinated operation between different scheduling sequences, timing correlation parameters need to be set. These parameters specify the time interval requirements between operational steps in different sequences. For example, there needs to be a certain interval between the time when emergency resources arrive at the key transmission node in the resource pre-scheduling sequence and the time when the train passes through that location in the train route reconfiguration sequence, ensuring that the train can only pass after the resources are in place and ready. Similarly, there needs to be a time interval between the time when the train completes the route switch in the train route reconfiguration sequence and the time when the evacuation is initiated in the station load evacuation sequence, ensuring that the train is properly adjusted and will not interfere with the evacuation process before the evacuation work is started.

[0176] Step S1411: Based on the time-series correlation parameters, integrate the resource pre-scheduling sequence with the train route reconstruction sequence, and the train route reconstruction sequence with the station load evacuation sequence into a multi-sequence collaborative scheduling matrix. The rows of the multi-sequence collaborative scheduling matrix represent time nodes, the columns represent different sequences, and the matrix elements are the operation content of the corresponding time nodes and sequences.

[0177] Based on time-series correlation parameters, resource pre-scheduling sequences, train route reconfiguration sequences, station load evacuation sequences, and system function compensation sequences are integrated to form a multi-sequence collaborative scheduling matrix. Rows in the matrix represent different time nodes, and columns represent different scheduling sequences (such as resource pre-scheduling sequences and train route reconfiguration sequences). Each element in the matrix represents the specific operation content under the corresponding time node and scheduling sequence. This matrix format clearly displays the time correlation and operation content between various scheduling sequences, enabling coordinated scheduling of all sequences.

[0178] Step S150: Convert the multi-sequence collaborative scheduling matrix into an executable scheduling instruction set, and push it to the corresponding control terminal of the urban rail system according to the execution subject.

[0179] The multi-sequence collaborative scheduling matrix is ​​an integrated representation of the emergency scheduling scheme. It needs to be converted into an executable scheduling instruction set, classified according to the executing entity, and then pushed to the corresponding control terminal in the urban rail transit system to ensure that the scheduling instructions can be executed accurately.

[0180] Step S151: Analyze each matrix element in the multi-sequence collaborative scheduling matrix and extract the execution subject information in the operation content. The execution subject includes train drivers, station staff, emergency resource dispatchers, and system maintenance personnel.

[0181] Each element of the multi-sequence collaborative scheduling matrix is ​​parsed to extract the executor information from the operation content. The executor refers to the personnel or team responsible for performing the operation. In urban rail transit emergency dispatch, common executors include train drivers (responsible for performing operations such as train route switching), station staff (responsible for evacuation guidance within the station), emergency resource dispatchers (responsible for the allocation and transportation of emergency resources), and system maintenance personnel (responsible for the switching and maintenance of system functional modules).

[0182] Step S152: Organize the operation content according to the executing entity to form a subset of dispatch instructions for different executing entities. The instruction subset for train drivers includes alternative track paths and path switching times in the train path reconstruction sequence; the instruction subset for station staff includes evacuation path planning and guidance equipment operation instructions in the station load evacuation sequence; the instruction subset for emergency resource dispatchers includes resource types, dispatch starting points, target key transmission node locations, and dispatch paths in the resource pre-schedule sequence; and the instruction subset for system maintenance personnel includes alternative module identifiers, switching operation steps, and function recovery standards in the system function compensation sequence.

[0183] Based on the extracted execution entity information, the operation content in the multi-sequence collaborative scheduling matrix is ​​categorized and organized according to the execution entity. A scheduling instruction subset is formed for each execution entity. The instruction subset corresponding to the train driver includes information such as the alternative track path and path switching time related to the train in the train path reconstruction sequence; the instruction subset corresponding to the station staff includes evacuation path planning and guidance equipment operation instructions related to the station in the station load evacuation sequence; the instruction subset corresponding to the emergency resource dispatcher includes information such as the resource type to be dispatched, dispatch starting point, target key transmission node location, and dispatch path in the resource pre-schedule sequence; and the instruction subset corresponding to the system maintenance personnel includes the alternative module identifier, switching operation steps, and function recovery standards of the affected functional modules in the system function compensation sequence.

[0184] Step S153: Add an execution timestamp to each subset of scheduling instructions, the execution timestamp being consistent with the time nodes in the multi-sequence collaborative scheduling matrix.

[0185] In a multi-sequence collaborative scheduling matrix, rows represent time nodes. To ensure that each execution entity can execute scheduling instructions at the correct time, an execution timestamp needs to be added to each subset of scheduling instructions. The execution timestamp is consistent with the time nodes in the matrix, accurate to the specific moment, such as year, month, day, hour, minute, and second.

[0186] Step S154: Convert each subset of scheduling instructions into an instruction format that conforms to the corresponding control terminal interface specification. The instruction format received by the train control terminal is track signal code, the instruction format received by the station control terminal is equipment control code, the instruction format received by the resource management terminal is resource scheduling form, and the instruction format received by the system control terminal is module operation script.

[0187] Different execution entities correspond to different control terminals, and these control terminals have different interface specifications and instruction format requirements. Therefore, each subset of scheduling instructions needs to be converted into an instruction format that conforms to the corresponding control terminal interface specification. For example, train control terminals typically receive instructions in the form of track signal codes to control train operation; station control terminals receive equipment control codes to control various guidance devices within the station; resource management terminals receive resource scheduling forms to record and execute emergency resource scheduling; and system control terminals receive module operation scripts to guide the switching operations of system function modules.

[0188] Step S155: Push the corresponding subset of scheduling instructions to the corresponding control terminal of the urban rail system in the order of execution timestamps, and record the time of instruction push and the reception status of the terminal feedback; if no reception confirmation is received from the terminal within the preset time, re-push the corresponding subset of scheduling instructions until confirmation is received or the maximum number of retries is reached.

[0189] The dispatch instruction subsets are pushed to the corresponding control terminals in the urban rail transit system according to their execution timestamps. During the push process, the time of instruction push and the terminal's reception status are recorded. A preset time is set; if no reception confirmation is received from the terminal within this time, it indicates that the instruction may have been unsuccessfully delivered, and the corresponding dispatch instruction subset needs to be pushed again. The push process is repeated until a reception confirmation is received from the terminal or the preset maximum number of retries is reached. If no confirmation is received after reaching the maximum number of retries, an alarm should be issued to notify relevant personnel for manual handling to ensure that the dispatch instructions are received and executed in a timely manner.

[0190] Based on the same inventive concept, please refer to Figure 2 This document shows a schematic block diagram of an urban rail system emergency dispatching system 100 that combines big data analysis and is used to execute the above-described emergency dispatching method for urban rail systems combining big data analysis, according to an embodiment of this application. The urban rail system emergency dispatching system 100 that combines big data analysis may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0191] In this embodiment, both the machine-readable storage medium 120 and the processor 130 are located in the urban rail transit system emergency dispatch system 100 that incorporates big data analysis, and they are separately configured. Alternatively, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110. The machine-readable storage medium 120 is used to store machine-executable instructions for executing the scheme of this application, and the processor 130 is used to execute the machine-executable instructions stored in the machine-readable storage medium 120 to implement the urban rail transit system emergency dispatch method incorporating big data analysis provided in the aforementioned method embodiment.

[0192] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for emergency dispatch of urban rail transit system combined with big data analysis, characterized in that, The method comprises: aggregating urban rail system dynamic monitoring data, the urban rail system dynamic monitoring data including train real-time working condition data, station area passenger flow distribution data, track structure vibration data, power supply network load data, and external environmental interference data; constructing an abnormal evolution topology based on the urban rail system dynamic monitoring data, the abnormal evolution topology taking data fluctuation characteristics as nodes, connecting the nodes to form a network through evolution correlation strength, and the evolution correlation strength reflecting the mutual induction relationship of different data fluctuation characteristics over time; identifying key conduction nodes and diffusion boundaries from the abnormal evolution topology, the key conduction nodes being nodes with evolution correlation strength exceeding a trigger threshold and located at the center of the topology, and the diffusion boundaries being boundary lines formed by a node set in the topology with evolution correlation strength not reaching a conduction condition; generating a multi-sequence cooperative scheduling matrix in combination with the key conduction nodes, diffusion boundaries, and urban rail system emergency resource layout data, the multi-sequence cooperative scheduling matrix including resource pre-scheduling sequences, train path reconstruction sequences, station load evacuation sequences, and system function compensation sequences, each sequence being linked through time sequence correlation parameters; converting the multi-sequence cooperative scheduling matrix into an executable scheduling instruction set and pushing it to the control terminal of the corresponding urban rail system according to the execution subject classification; the combination of the key conduction nodes, diffusion boundaries, and urban rail system emergency resource layout data to generate a multi-sequence cooperative scheduling matrix comprises: calling urban rail system emergency resource layout data, the urban rail system emergency resource layout data including mobile emergency team distribution locations, emergency equipment storage sites, backup train parking locations, and emergency material storage library information, each type of resource being labeled with response time and coverage range; spatially matching the collection locations of the key conduction nodes with the resource locations in the urban rail system emergency resource layout data to determine the types and quantities of emergency resources that can be called around each key conduction node; determining the processing priority according to the topological centrality of the key conduction nodes, with a larger topological centrality value indicating a higher processing priority, generating a resource pre-scheduling sequence in order of priority, the resource pre-scheduling sequence including resource type, scheduling starting point, target key conduction node location, and scheduling path; analyzing the track line distribution within the diffusion boundary and planning an alternative track path that avoids the key conduction nodes in combination with the locations of the key conduction nodes, the alternative track path meeting the curvature and slope requirements for safe train operation; extracting the real-time positions of all running trains within the diffusion boundary, sorting the trains by distance from the key conduction nodes from short to long to determine the path reconstruction order, and generating a train path reconstruction sequence, the train path reconstruction sequence including train number, original running path, alternative track path, and path switching time; calculating a station congestion index for the stations within the diffusion boundary according to the passenger flow distribution data, determining the evacuation priority in order of the station congestion index values from large to small, and the station congestion index being the ratio of the number of real-time passengers to the station construction passenger capacity. According to the evacuation priority, a station load evacuation path is constructed, the station load evacuation path including entrance and exit selection, transfer passage guidance and platform shunting direction, and a station load evacuation sequence is generated, the station load evacuation sequence including station name, evacuation path planning, guidance device operation instruction and evacuation completion time; An affected urban rail system function module within the diffusion boundary is identified, a replaceable function module is determined based on system redundancy design, and a function switching scheme is generated, the function switching scheme including main and standby module switching sequence, data synchronization mode and switching verification steps; A compensation sequence is determined according to the importance of the function module, and a system function compensation sequence is generated, the function module having a greater influence on the safe operation of the urban rail system being ranked higher, the system function compensation sequence including function module name, replacement module identification, switching operation step and function recovery standard; Timing correlation parameters are set, the timing correlation parameters being time interval requirements of operation steps in different sequences, the time interval of the resource pre-scheduling sequence and the train path reconstruction sequence meeting the requirement that the train can pass only after the resource is in place, and the time interval of the train path reconstruction sequence and the station load evacuation sequence meeting the requirement that evacuation is started after the train is adjusted; Based on the timing correlation parameters, the resource pre-scheduling sequence and the train path reconstruction sequence, and the train path reconstruction sequence and the station load evacuation sequence are integrated into a multi-sequence cooperative scheduling matrix, the rows of the multi-sequence cooperative scheduling matrix representing time nodes, the columns representing different sequences, and the matrix elements being operation contents corresponding to the time nodes and the sequences.

2. The emergency dispatch method for urban rail transit system integrated with big data analysis according to claim 1, characterized in that, The abnormal evolution topology is constructed based on the dynamic monitoring data of the urban rail system, the abnormal evolution topology taking data fluctuation characteristics as nodes, connecting the nodes to form a network through evolution correlation strength, including: Fluctuation characteristics are extracted from the dynamic monitoring data of the urban rail system, and the segments of each type of data that exceed the normal fluctuation range are taken as independent data fluctuation characteristics, each data fluctuation characteristic including fluctuation start time, fluctuation peak value and fluctuation duration; Each data fluctuation characteristic is assigned a unique identification as a basic node of the abnormal evolution topology, and the attribute information of the basic node includes the type, collection location and fluctuation amplitude of the corresponding data; The time-induced coefficient of any two basic nodes is calculated, the time-induced coefficient being the reciprocal of the difference between the fluctuation start time of the later-occurring basic node and the fluctuation peak time of the earlier-occurring basic node, the smaller the difference, the larger the time-induced coefficient; The influence correlation coefficient of any two basic nodes is calculated, the influence correlation coefficient being the physical correlation degree of the data types of the two basic nodes, the physical correlation degree being determined based on the coupling relationship of the urban rail system equipment, the closer the coupling relationship, the larger the influence correlation coefficient; The evolution correlation strength is obtained by multiplying the time-induced coefficient and the influence correlation coefficient, the larger the value of the evolution correlation strength, the more significant the mutual induction relationship between the two basic nodes; When the evolution correlation strength is greater than the correlation threshold, a directed connection edge is established between the two basic nodes, and the direction of the connection edge is determined by the fluctuation start time of the two basic nodes, from the earlier-occurring node to the later-occurring node. constructing a network topology of all the base nodes and the directed connection edges to form an initial abnormal evolution topology; every interval of a preset update period, incorporating a newly extracted data fluctuation feature as a new node, re-computing the evolution correlation strength and updating the directed connection edges to realize dynamic expansion of the abnormal evolution topology. 3.The urban rail transit system emergency dispatching method based on big data analysis of claim 1, wherein, The identifying the key conduction nodes and the diffusion boundary from the abnormal evolution topology comprises: traversing all the base nodes in the abnormal evolution topology, counting the number of in-edges and out-edges of each base node, the number of in-edges being the number of directed connection edges pointing to the base node, and the number of out-edges being the number of directed connection edges pointing from the base node; computing the topological centrality of each base node, the topological centrality being the ratio of the sum of the number of in-edges and the number of out-edges to the average number of connection edges of all the base nodes in the topology, the greater the ratio, the closer the base node to the topological center; extracting the evolution correlation strength corresponding to the out-edges of each base node, and marking the base node as a potential conduction node if there is at least one out-edge whose evolution correlation strength exceeds a trigger threshold; selecting the top K nodes in the topological centrality from the potential conduction nodes to determine the key conduction nodes; collecting all the base nodes in the abnormal evolution topology that are not marked as potential conduction nodes as boundary candidate nodes; computing the shortest path hop count between the boundary candidate nodes and the key conduction nodes, the shortest path hop count being the minimum number of directed connection edges connecting the two nodes; marking the boundary candidate nodes whose shortest path hop count is greater than a boundary threshold as boundary nodes, and the spatial distribution of all the boundary nodes forming an initial outline of the diffusion boundary; performing edge smoothing processing on the initial outline to connect adjacent boundary nodes in spatial position order to form a closed curve as the final diffusion boundary; recording the unique identifier, corresponding data fluctuation feature and topological centrality of the key conduction nodes, as well as the closed curve coordinate sequence and the identifiers of the boundary nodes contained in the diffusion boundary.

4. The emergency dispatch method for urban rail transit system integrated with big data analysis according to claim 1, characterized in that, The generating a resource pre-scheduling sequence in priority order comprises: selecting the node with the largest topological centrality value from the key conduction nodes as the first target node and marking it as the current target node; selecting the emergency resource with the smallest response time value that spatially matches the current target node from the urban rail system emergency resource layout data to form a resource combination, the resource combination containing at least three of a mobile emergency team, emergency equipment, a backup train and emergency supplies; extracting the current storage location of each resource in the resource combination as a scheduling starting point, the coordinate information of the scheduling starting point being obtained from the urban rail system emergency resource layout data; planning a scheduling path in the internal passage map of the urban rail system with the scheduling starting point as the starting point and the location of the current target node as the end point, the scheduling path preferentially selecting an emergency passage parallel to the track and avoiding high-risk areas within the diffusion boundary; recording the resource type, scheduling starting point coordinates, current target node location coordinates and key passing points of the scheduling path of each resource in the resource combination to form a resource pre-scheduling sub-sequence for the current target node; Removing the current target node from the key conduction nodes, marking the node with the second largest value of the topological centrality as the new current target node, and repeating the above process of screening resource combinations, planning scheduling paths, and forming resource pre-scheduling sub-sequences; After all the key conduction nodes are processed, arranging all the resource pre-scheduling sub-sequences in the order of the processing priority of the target nodes to form a resource pre-scheduling sequence; Checking whether there are overlapping sections in the scheduling paths of different resources in the resource pre-scheduling sequence, and adjusting the scheduling path or the departure time of one of the resources if there are overlapping sections; Adding a scheduling start time and a predicted arrival time to each resource pre-scheduling sub-sequence in the resource pre-scheduling sequence, determining the scheduling start time according to the current state of the resource, and calculating the predicted arrival time according to the length of the scheduling path and the moving speed of the resource. 5.The urban rail transit system emergency dispatching method in conjunction with big data analysis of claim 1, wherein, The train path reconstruction sequence is generated by sorting the trains according to the distance from the train to the key conduction node from short to long, determining the path reconstruction order, including: Obtaining real-time position information of all running trains within the diffusion boundary, the real-time position information being accurate mileage markers of the trains on the track line; Calculating the straight-line distance between the real-time position of each train and the nearest key conduction node, the straight-line distance being determined by the difference between the train position mileage and the corresponding track mileage of the key conduction node; Sorting the trains according to the straight-line distance values from small to large, the smaller the straight-line distance value, the higher the priority of the train for path reconstruction; For trains with the same straight-line distance value, sorting the trains according to the train current running speed value from large to small, the larger the train current running speed value, the higher the priority of the train for path reconstruction; Determining the path reconstruction order according to the sorting result, the train ranked first being marked as the first reconstruction object; Extracting the original running path of the first reconstruction object, the original running path including the starting station, the passing stations, the terminal station, and the track sections of the train; Planning an alternative track path for the first reconstruction object based on the diffusion boundary and the position of the key conduction node, the alternative track path needing to include the same starting station and terminal station as the original running path, and the passing track sections avoiding all key conduction nodes; Analyzing the track occupancy of the alternative track path, and selecting a period when the track is idle as the path switching time, the path switching time needing to meet the safety interval requirements with the preceding and following trains; Recording the train number of the first reconstruction object, the track section sequence of the original running path, the track section sequence of the alternative track path, and the specific time of the path switching time, forming a first train path reconstruction sub-sequence; Removing the first reconstruction object from the sorting result, marking the train immediately following the first reconstruction object in the sorting result as a new reconstruction object, and repeating the above process of extracting the original path, planning the alternative path, and determining the switching time; After all the train path reconstruction sub-sequences are generated, arranging all the train path reconstruction sub-sequences in the reconstruction order to form a train path reconstruction sequence; Verifying the difference between the total length of the alternative track path and the length of the original running path in the train path reconstruction sequence, and re-planning the alternative track path if the difference exceeds a preset range.

6. The emergency dispatch method for urban rail transit system integrated with big data analysis according to claim 1, characterized in that, The station load evacuation path is constructed according to the evacuation priority, including: select a station with the largest congestion index value from the stations within the diffusion boundary as the first evacuation object, and mark it as the current evacuation station; obtain internal structure data of the current evacuation station, the internal structure data including the number and location of entrances and exits, the distribution of transfer passages, the layout of platforms, and the location of stairs and elevators; identify a target crowded area in the current evacuation station according to the people flow distribution data, the target crowded area being an area with a personnel density exceeding a safety density threshold; construct an evacuation path plan, the evacuation path plan including multiple evacuation paths from the target crowded area to each entrance and exit, each evacuation path including the transfer passages, stairs, elevators, and exit signs to be passed through; generate guidance device operation instructions based on the evacuation path plan, the guidance device operation instructions including direction information displayed on a signboard, evacuation guidance content broadcast by a broadcast system, the opening state of a gate, and the running direction of an elevator; calculate an evacuation completion time according to the real-time number of people in the current evacuation station and the total traffic capacity of the evacuation paths, the total traffic capacity being the total number of people that can be evacuated per unit time on all the evacuation paths; record the station name of the current evacuation station, the channel number sequence of the evacuation path plan, the specific content of the guidance device operation instructions, and the specific time of the evacuation completion time, to form a first station load evacuation subsequence; remove the current evacuation station from the stations within the diffusion boundary, mark a station with a congestion index value next to the current evacuation station as a new current evacuation station, and repeat the process of constructing a path, generating instructions, and calculating a time; arrange all the station load evacuation subsequences in a station load evacuation sequence according to an evacuation priority when all the station load evacuation subsequences are generated, and arrange all the station load evacuation subsequences in a station load evacuation sequence according to an evacuation priority when all the station load evacuation subsequences are generated; check whether the evacuation paths of adjacent stations in the station load evacuation sequence share an external passage, and adjust the selection of an entrance or exit of one of the stations if an external passage is shared.

7. The emergency dispatch method for urban rail transit system integrated with big data analysis according to claim 1, characterized in that, determine a compensation order according to the importance ranking of the functional modules, and generate a system function compensation sequence, including: identify urban rail system functional modules affected by the key conduction nodes within the diffusion boundary, the urban rail system functional modules including a train control system, a signal system, a power supply system, a communication system, and a gate control system; match a corresponding replacement module to each affected urban rail system functional module, the replacement module being a standby module with the same or similar function in the urban rail system, the replacement module being located outside the diffusion boundary or not being affected by the key conduction nodes; determine an importance ranking according to the influence of the urban rail system functional modules on the safe operation of the urban rail system, the greater the influence value, the higher the ranking, the influence value of the train control system and the signal system being greater than that of other systems; arrange the affected functional modules in order from front to back according to the importance ranking to determine a compensation order, the functional module with the highest ranking being given priority for compensation operation; select the first functional module from the ranking result as the first compensation object, and mark it as the current compensation module; generate a switching operation step for the current compensation module, the switching operation step including a shutdown process of the original module, a startup process of the replacement module, a data transmission process from the original module to the replacement module, and an interfacing process of the modules. Setting a function recovery standard, the function recovery standard is the range of operating parameters of the replacement module, including response time, data transmission accuracy and control accuracy; Recording the function module name of the current compensation module, the unique identification of the replacement module, the detailed sequence of the switching operation steps and the specific parameter range of the function recovery standard, forming the first system function compensation sub-sequence; Removing the current compensation module from the sorting result, marking the function module immediately following it in the sorting result as the new current compensation module, and repeating the above process of matching the replacement module, generating the steps, and setting the standard; When all affected function modules complete the system function compensation sub-sequence generation, arrange all system function compensation sub-sequences in the compensation order to form a system function compensation sequence, and verify whether the total processing capacity of the replacement modules in the system function compensation sequence meets the operation requirements of the urban rail system. If not, increase the number of replacement modules or adjust the compensation order.

8. The urban rail system emergency dispatching method in conjunction with big data analysis according to claim 1, characterized in that, The method comprises the following steps: Analyzing each matrix element in the multi-sequence cooperative scheduling matrix, and extracting the execution subject information in the operation content, wherein the execution subject includes train drivers, station staff, emergency resource dispatchers and system maintenance personnel; Classifying and organizing the operation content according to the execution subject to form a scheduling instruction subset for different execution subjects. The instruction subset corresponding to the train driver includes the replacement track path and path switching time in the train path reconstruction sequence; the instruction subset corresponding to the station staff includes the evacuation path planning and guidance device operation instruction in the station load evacuation sequence; the instruction subset corresponding to the emergency resource dispatcher includes the resource type, dispatch starting point, target key conduction node position and dispatch path in the resource pre-dispatch sequence; and the instruction subset corresponding to the system maintenance personnel includes the replacement module identification, switching operation step and function recovery standard in the system function compensation sequence; Adding an execution timestamp to each scheduling instruction subset, which is consistent with the time node in the multi-sequence cooperative scheduling matrix; Converting each scheduling instruction subset into an instruction format conforming to the interface specification of the corresponding control terminal. The instruction format received by the train control terminal is track signal coding, the instruction format received by the station control terminal is device control code, the instruction format received by the resource management terminal is resource scheduling form, and the instruction format received by the system control terminal is module operation script; According to the order of the execution timestamp, the corresponding scheduling instruction subset is pushed to the control terminal corresponding to the urban rail system, and the time of instruction pushing and the receiving state of terminal feedback are recorded. If no receiving confirmation is received from the terminal within a predetermined time, the corresponding scheduling instruction subset is pushed again until the confirmation is received or the maximum number of retries is reached.

9. An emergency dispatch system for urban rail transit system combined with big data analysis, characterized in that, The method comprises the following steps: A processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the machine-executable instructions to perform the urban rail system emergency scheduling method combined with big data analysis according to any one of claims 1 to 8.

Images