Management method and system for node network of railway dispatching centralized system

Abstract

The invention discloses a management method for a node network of a railway centralized dispatching system. The management method comprises the following steps of: initializing network nodes: initializing a management node, a central node, a station node and a dynamic relay node in the system network according to functions and service logic of the centralized dispatching system; setting a network node grouping mechanism: calculating the importance degree of each node and a subordinate node set of a relay node by a management node through a mode of combining regular polling and event triggering, controlling dynamic grouping of station nodes and a center node, and forming a dynamic grouping mechanism; and setting a network node dynamic information routing mechanism: calculating the optimal information routing of the center node and the station nodes in an event triggering manner, and forming the dynamic routing mechanism under the coordination of the management node.

Description

Technical Field

[0001] The present invention relates to a management method and a system thereof, and in particular to a management method and a system thereof for a railway dispatching centralized system node network. Background Art

[0002] my country's railway dispatching centralized system adopts a two-layer command architecture of center-station. The center layer divides the terminals into functions in a business manner to form multiple center nodes; the station layer forms multiple station nodes based on the station. There are various information interaction requirements between center nodes, station nodes, and center and station nodes. It is a typical distributed network system. The traditional dispatching centralized system network structure is restricted by the railway communication network networking mode and channel capacity. It forces the fixed communication link between nodes to realize the collection of a large number of scattered node information and the forwarding of information between nodes, forming a multi-level network node structure such as station node set-server node set-center node set. It has the following three main problems: (1) Some nodes have multiple sub-nodes, and the failure of the main node will lead to the overall failure of the nodes under its jurisdiction; (2) The increase in the network communication level increases the failure risk of the overall network; (3) The fixed link method between nodes makes it impossible for the normal nodes affected by the faulty nodes to dynamically drift and establish normal link relationships with other related nodes. Therefore, it is urgent to carry out research on the node network management method of the dispatching centralized system with high fault tolerance to improve the reliability of information interaction in the railway dispatching centralized system. Summary of the invention

[0003] Aiming at the problem of insufficient fault tolerance of the network structure of the railway dispatching centralized system, the present invention proposes a management method and device for the node network of the railway dispatching centralized system, which improves the fault tolerance and reliability of the information interaction of the railway dispatching centralized system by adding management nodes and realizing dynamic node grouping and dynamic information routing settings. The technical solution is as follows:

[0004] A method for managing a node network of a railway dispatching centralized system, characterized by:

[0005] Step S1: Initialize network nodes: Initialize management nodes, central nodes, station nodes and dynamic relay nodes in the system network according to the functions and business logic of the centralized scheduling system;

[0006] Step S2 sets the network node grouping mechanism: the management node calculates the importance of each node, the set of nodes under the control of the relay node, controls the dynamic grouping of the station node and the central node, and forms a dynamic grouping mechanism by combining regular polling and event triggering;

[0007] Step S3 sets the dynamic information routing mechanism of the network nodes: the central node and the station node calculate the best information routing for themselves through event triggering, and form a dynamic routing mechanism under the coordination of the management node.

[0008] The invention also discloses a management device for a railway dispatching centralized system node network.

[0009] The present invention also discloses a non-volatile storage medium, characterized in that the non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the above method.

[0010] The present invention also discloses an electronic device, characterized in that it comprises a processor and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein the computer-readable instructions execute the above method when they are executed.

[0011] Beneficial Effects

[0012] (1) Improve the system's fault tolerance when some nodes fail;

[0013] (2) A management node is added to the actual centralized scheduling network system, and the grouping and information routing of each node in the network are dynamically calculated to solve the problem of information transmission failure of related nodes caused by the failure of a single node. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of a management method for a railway dispatching centralized system node network of the present invention;

[0015] Figure 2 This is a schematic diagram of a distributed node network example;

[0016] Figure 3 Set up a schematic diagram for node information routing attributes;

[0017] Figure 4 Schematic diagram of inter-node link communication failure. DETAILED DESCRIPTION

[0018] The present invention proposes a method for managing a railway dispatching centralized system node network. The method flow chart is as follows: Figure 1 shown.

[0019] In step S1, the network structure and information transmission business logic of the railway dispatching concentration system are first disassembled. The network structure of the dispatching concentration system is divided into a central node layer (including business platform nodes such as train dispatching desks, control desks, display desks, and inquiry desks), a station node layer (including platform nodes such as station communication services, vehicle terminals, and signalman terminals) and relay nodes with information collection, preliminary processing and forwarding functions (including front-end communication machines, servers and other nodes). Various types of information in the dispatching concentration system flow in various network nodes, and the platform function nodes selectively receive and process the information according to their own business needs; secondly, in order to achieve flexible management of the information transmission needs of each business node in the network, the traditional method of determining the node link object and the direction of information transmission based on the static configuration of each node is abandoned, and management nodes are added to dynamically manage the node network at the global level; finally, according to the business characteristics of each node, the initialization of each node is realized by interacting with the management node.

[0020] Let the central node of a certain type be Where x represents the type of central node, x∈{1,2,...,W}, W represents the total number of central node types, and M represents the total number of central nodes of this type (such as train switching stations); station nodes are s1,s2,…,s N-1 ,s N , where N represents the total number of station subnodes; the various types of information in the centralized dispatching system are i1, i2, …, i P-1 ,i P , where P represents the total number of information types. Figure 2 As shown, the station node s can be constructed i For the information demand matrix L of the station node, that is

[0021]

[0022] Among them, l i,j Represents node s i With node s n The information generated p Is there an information demand relationship between them (where i∈{1,2,…,N}, j∈{1,2,…,P}). If so, then l i,j =1, otherwise l i,j = 0. Similarly, the central node c can be constructed separately i For the demand matrix between the central nodes and the demand matrix between the central and station nodes, the management node is initialized to set the information requirements of each service node. The management of the relay node is implemented in step S2.

[0023] In step S2, Figure 2Taking the centralized distributed node network as an example, the management node calculates the node grouping set including the relay node according to the initialized node information demand, which is used to determine the default direction of information between network nodes and solve the problem of reliable information exchange between nodes. The specific implementation method of step S2 is as follows.

[0024] Step S2-1: Set a relay node set.

[0025] Since railway dispatching and command adopts the command method of designated dispatching desks to manage all stations within the designated dispatching section, the division of station node groups is restricted to a certain extent by the dispatching section to which they belong, and the central node has a demand for information within the entire network. If the station node and the central node are used in a point-to-point communication mode, it will be necessary to establish a large number of information links for all-weather stable communication between nodes, occupy a large amount of link resources, and generate a large amount of various redundant information, which is easy to cause the limited bandwidth of the railway dispatching centralized system to be blocked. Therefore, relay nodes are introduced to manage a certain number of business nodes to realize the overall collection and forwarding of data from the subordinate business nodes. The specific method is as follows:

[0026] (1) Determine the number of dispatching sections SEC_N, the number of stations STA_N, and set the threshold of the number of nodes under each relay node to RLY_MAX;

[0027] (2) Set the subordinate relationship between the dispatching section and each station, and construct the relationship matrix A between the dispatching section and the station, that is,

[0028]

[0029] Among them, a i,j Represents node c i With node s j The relationship of a (where i∈{1,2,…,M}, j∈{1,2,…,N}). If there is a relationship of a i,j =1, otherwise a i,j =0. Due to the particularity of railway hubs, station nodes and dispatching nodes do not correspond one to one, and some stations are subordinate to multiple dispatching desks. The dispatching desk has frequent information exchanges with the stations under its jurisdiction, so it is preferred to set the stations under the jurisdiction of the same dispatching desk in the same grouping set. However, since the number of station nodes in different dispatching desks is not balanced, and different dispatching desks also have information exchange requirements for boundary stations, the station nodes within the jurisdiction of the dispatching desk and the dispatching section may not completely overlap.

[0030] (3) Calculate the number of relay nodes Y① First, according to the relationship matrix, calculate the number of nodes with a single subordinate station in each dispatching section; ② If there is a non-single subordinate station node in a certain section, if the node is at the boundary of two dispatching sections, it will be assigned to the dispatching section with fewer single subordinate station nodes, otherwise it will be assigned to the dispatching section; ③ Let And assign node weights according to the positions of station nodes in the dispatching section; ④ Convert the grouping into a knapsack problem and calculate the grouping results; ⑤ Set the grouping evaluation conditions (station nodes in the same dispatching section should be grouped in the same group as much as possible, the same group should not span three dispatching sections, and the number of station nodes in each group should be as consistent as possible) and evaluate the grouping results; ⑥ If the evaluation result is low, let Y =Y+1, repeat ④⑤ until the evaluation result is acceptable, and determine the number of relay nodes;

[0031] Step S2-2: Set up a node grouping mechanism, such as Figure 3 shown.

[0032] (1) The management node constructs information routing for each node based on the relay node and its subordinate nodes calculated in step S2-1 and the node information demand matrix constructed in step S1, and determines the link relationship matrix between the station node, the central node and the relay node;

[0033] (2) After each node is started, it registers with the management node using a combination of node ID and password;

[0034] (3) The management node verifies the legitimacy of the node based on its internal verification mechanism;

[0035] (4) If it is legal, establish a link with the node and synchronize the link relationship matrix and information demand matrix to the node; otherwise, ignore the node application and issue an alarm;

[0036] (5) The node establishes links with related nodes according to the synchronized link relationship matrix, and encapsulates and sends various types of information data generated internally according to the information demand matrix;

[0037] (6) The node processes the various types of information data it receives according to the information demand matrix. If the received data does not match the information demand matrix, it discards the data and sends an alarm message to the management node.

[0038] Step S3 includes the following steps.

[0039] Step S3-1: Node failure determination.

[0040] (1) Each node uses an event-driven approach combined with a link relationship matrix to monitor the link status between itself and related nodes. Once the link with a node is interrupted, it sends a corresponding alarm message to the management node;

[0041] (2) If in the cycle T If a normal link cannot be established with the node within 10 seconds, the node failure is reported to the management node.

[0042] (3) The management node is based on the link relationship matrix, such as Figure 4 As shown, the communication between nodes s2 and s3 is interrupted, and the random calculation generates K 1. The test packet is encapsulated with a completely different path (usually set to 2 or 3, such as r1→s3 or s4→s3, etc., to avoid excessive calculations affecting timeliness) leading to the node, and the test packet is sent to the node (the packet header contains the destination code with the node as the destination, and also contains the IDs of each node on the path to implement information sending on the specified path);

[0043] (4) If the node returns a confirmation packet along the original path after receiving the test packet, it can be confirmed that the node is not faulty, but the communication between specific nodes is faulty;

[0044] (5) If the management node does not receive the confirmation packet, it is considered that the node is faulty;

[0045] Step S3-2: Dynamic information routing update.

[0046] (1) If the fault is only caused by communication failure between nodes, the management node will synchronize the new routing method between the two nodes to the corresponding nodes, such as Figure 4 In the example, the information exchange route between s2 and s3 is updated by the management node to s2→r1→s3 and synchronized to s2 and s3. Then, the message transmission path ID is synchronized and updated in the packet header of the information sent by s2 and s3.

[0047] (2) If a node fails, after the failed node is determined, the set of nodes affected by the node failure is calculated based on the link relationship matrix, and the link information routing is reset; Figure 4 For example, if the relay node r1 fails, the boundary node s n Establish a link with the next relay node r2, and the information path of other nodes in the node grouping set transitions to s n , and then message passing is realized through r2, such as s2→s3→…→s n →r2;

[0048] (3) When the failed node recovers, the management node synchronizes the original link relationship to all related nodes again.

[0049] The present invention proposes a management method for a railway dispatching centralized system node network, comprising: initializing network nodes: initializing management nodes, central nodes, station nodes and dynamic relay nodes in the system network according to the functions and business logic of the dispatching centralized system; setting a network node grouping mechanism: the management node calculates the importance of each node and the subordinate node set of the relay node by combining regular polling and event triggering, controls the dynamic grouping of station nodes and central nodes, and forms a dynamic grouping mechanism; setting a network node dynamic information routing mechanism: the central node and the station node calculate their own optimal information routing by event triggering, and form a dynamic routing mechanism under the coordination of the management node.

[0050] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for managing a node network of a railway dispatching centralized system, characterized by: Step S1: Initialize network nodes: Initialize management nodes, central nodes, station nodes and dynamic relay nodes in the system network according to the functions and business logic of the centralized scheduling system; Step S2 sets the network node grouping mechanism: the management node calculates the importance of each node, the set of nodes under the control of the relay node, controls the dynamic grouping of the station node and the central node, and forms a dynamic grouping mechanism by combining regular polling and event triggering; Step S3 sets the dynamic information routing mechanism of the network nodes: the central node and the station node calculate the best information routing for themselves through event triggering, and form a dynamic routing mechanism under the coordination of the management node.

2. The method for managing a railway dispatching centralized system node network according to claim 1, characterized in that: The step 1 includes the following contents: First, the network structure and information transmission business logic of the railway dispatching centralized system are disassembled; Secondly, add management nodes to dynamically manage the node network at the global level; Finally, according to the business characteristics of each node, each node is initialized by interacting with the management node.

3. The method for managing a railway dispatching centralized system node network according to claim 1, characterized in that: The step 2 includes the following contents: Step S2-1: Set up a relay node set: (1) Determine the number of dispatching sections SEC_N, the number of stations STA_N, and set the threshold of the number of nodes under each relay node to RLY_MAX; (2) Set the subordinate relationship between the dispatching section and each station, and construct the relationship matrix A between the dispatching section and the station: Among them, a i,j Represents node c i With node s j a) is a relationship of ownership (where i∈{1,2,…,M}, j∈{1,2,…,N}); if there is a relationship of ownership, then a i,j =1, otherwise a i,j =0; (3) Calculate the number of relay nodes Y :①First, according to the relationship matrix, calculate the number of nodes with single subordinate stations in each dispatching section;②If there is a node with non-single subordinate station in a section, if the node is at the boundary of two dispatching sections, it will be assigned to the dispatching section with fewer single subordinate station nodes, otherwise it will be assigned to the dispatching section;③Let And assign node weights according to the position of the station node in the dispatching section; ④ Convert the grouping into a knapsack problem and calculate the grouping results; ⑤ Set the grouping evaluation conditions and evaluate the grouping results; ⑥ If the evaluation result is low, let Y =Y+1, repeat ④⑤ until the evaluation result is acceptable, and determine the number of relay nodes; Step S2-2: Setting up the node grouping mechanism: (1) The management node constructs information routing for each node based on the relay node and its subordinate nodes calculated in step S2-1 and the node information demand matrix constructed in step S1, and determines the link relationship matrix between the station node, the central node and the relay node; (2) After each node is started, it registers with the management node using a combination of node ID and password; (3) The management node verifies the legitimacy of the node based on its internal verification mechanism; (4) If it is legal, establish a link with the node and synchronize the link relationship matrix and information demand matrix to the node; otherwise, ignore the node application and issue an alarm; (5) The node establishes links with related nodes according to the synchronized link relationship matrix, and encapsulates and sends various types of information data generated internally according to the information demand matrix; (6) The node processes the various types of information data it receives according to the information demand matrix. If the received data does not match the information demand matrix, it discards the data and sends an alarm message to the management node.

4. The method for managing a railway dispatching centralized system node network according to claim 1, characterized in that: The step 3 includes the following contents: Step S3-1: Node failure determination: (1) Each node uses an event-driven approach combined with a link relationship matrix to monitor the link status between itself and related nodes. Once the link with a node is interrupted, it sends a corresponding alarm message to the management node; (2) If in the cycle T If a normal link cannot be established with the node within 10 seconds, the node failure is reported to the management node. (3) The management node assumes that the communication between station nodes s2 and s3 is interrupted according to the link relationship matrix, and randomly generates K A completely different path leading to the node is used to encapsulate the test packet and send it to the node; (4) If the node returns a confirmation packet along the original path after receiving the test packet, it can be confirmed that the node is not faulty, but the communication between specific nodes is faulty; (5) If the management node does not receive the confirmation packet, it is considered that the node is faulty; Step S3-2: Dynamic information routing update: (1) If only the communication failure occurs between nodes, the management node will synchronize the new routing mode between the two nodes to the corresponding nodes. The information exchange route between station node s2 and station node s3 will be updated by the management node to station node s2→r1→s3 and synchronized to station node s2 and station node s3. Then, the message transmission path ID will be synchronized and updated in the packet header of the information sent by station node s2 and station node s3. r1 is the relay node. (2) If a node fails, after the failed node is determined, the set of nodes affected by the node failure is calculated based on the link relationship matrix, and the link information routing is reset; (3) When the failed node recovers, the management node synchronizes the original link relationship to all related nodes again.

5. A management system for a railway dispatching centralized system node network, characterized in that: The management system includes the following functional modules: Initialize network node module: initialize management nodes, central nodes, station nodes and dynamic relay nodes in the system network according to the functions and business logic of the centralized dispatching system; Setting up the network node grouping mechanism module: The management node calculates the importance of each node, the set of nodes under the control of the relay node, controls the dynamic grouping of station nodes and central nodes, and forms a dynamic grouping mechanism through a combination of regular polling and event triggering; Set up a dynamic information routing mechanism module for network nodes: The central node and station node calculate their own optimal information routing through event triggering, and form a dynamic routing mechanism under the coordination of the management node.

6. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein the program controls the device where the non-volatile storage medium is located to execute the method according to any one of claims 1 to 4 when the program is executed.

7. An electronic device, characterized in that: It comprises a processor and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein the computer-readable instructions execute the method described in any one of claims 1 to 4 when executed.

Images