Common mode risk search method for power communication network based on SDN

Abstract

The present invention discloses a common mode risk search method for an electric power communication network based on SDN, which includes S1 adding a service request to a service node pair set Ω, and storing working links in a network topology structure G into a working state set U; S2 judging whether Ω is empty, and if so, completing the search, otherwise deleting the links in U in G to generate a subgraph G'; S3 selecting a service request H with the highest service importance in Ω, searching whether there is a minimum common mode risk path in G' that satisfies H, and if so, entering S7, otherwise entering S4; S4 searching whether there is a minimum common mode risk path in G that satisfies H, and if so, entering S5, otherwise entering S6; S5 judging whether the minimum common mode risk path satisfies a common mode risk optimization model, and if so, entering S7; otherwise entering S6; S6 returning H that the minimum common mode risk path is not found, deleting H from U, and then returning to S2; S7 adding the minimum common mode risk path to U, allocating bandwidth, and deleting H from U, and then returning to S2.

Description

Technical Field

[0001] The present invention relates to a power grid service transmission technology, and in particular to a common mode risk search method for a power communication network based on SDN. Background Art

[0002] As a crucial component of power communications and intelligent control, the power communication network supports core grid operations such as dispatching, production, and utilization, and is the foundation for stable and reliable grid operation. With the continuous expansion of business needs and the continuous development of network topologies, the types of services carried by links are becoming more diverse. The secure transmission of various information such as power dispatching, relay protection, and production management depends entirely on network performance and hardware equipment.

[0003] Current research on risk control in power communication networks focuses primarily on maintaining system stability and reliability. On the one hand, the continuous expansion of power communication services has caused some core communication sites or fiber optic links to carry too much traffic, leading to network congestion, excessive service transmission delays, and uneven bandwidth resource allocation. This in turn increases the risk of common-mode failures in the network and reduces the stability of the communication network system. On the other hand, the power communication network differs from other networks due to the rapid dynamic changes in its services. Choosing a highly reliable, stable, and low-common-mode-risk route for the network based on real-time service changes is an important prerequisite for improving service quality and system reliability.

[0004] Therefore, the common mode risk problem has become the primary issue affecting the secure transmission of services in the power communication network, and it is necessary to design a suitable routing search algorithm based on the business characteristics of power communication to reduce the common mode risk of the system. Summary of the Invention

[0005] In response to the above-mentioned deficiencies in the prior art, the transmission path searched by the SDN-based power communication network common mode risk search method provided by the present invention can significantly reduce the common mode risk problem.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0007] A common mode risk search method for a power communication network based on SDN is provided, which includes:

[0008] S1. Obtain a service request, add the source node and destination node in the service request to a service node pair set, and store the links in the working state in the network topology structure G into a working state set;

[0009] S2. Determine whether the service node pair set is empty. If so, complete the search. Otherwise, delete the links in the working state set in the network topology structure G to generate a subgraph G'.

[0010] S3. Select the business request H with the highest importance to the centralized business from the business node, and search the subgraph G' to see if there is a path with the lowest common mode risk that satisfies the business request H. If so, proceed to S7; otherwise, proceed to S4.

[0011] S4. Search the network topology G to see if there is a path with the lowest common mode risk that satisfies the service request H. If so, proceed to S5; otherwise, proceed to S6.

[0012] S5, determine whether the lowest common mode risk path satisfies the common mode risk optimization model. If so, proceed to S7; otherwise, proceed to S6;

[0013] S6. Return that the business request H has not found the lowest common mode risk path, delete the business request H from the working status set, and then return to S2.

[0014] S7. Add the lowest common mode risk path to the working state set, allocate bandwidth, and delete the service request H from the working state set, then return to S2.

[0015] The beneficial effects of the present invention are as follows: when transmitting a service request, this scheme first searches on a basis that does not include links that are already in working state, which can greatly reduce the situation where common mode risks exist; when the lowest common mode risk path is not found, the initial network topology structure G is searched for the lowest common mode risk path that meets the common mode risk optimization model, and the risk is controlled through the common mode risk optimization model, so that the power communication network has a lower common mode risk during service transmission.

[0016] Furthermore, the method for searching for the path with the lowest common mode risk includes:

[0017] A1. Create an array visited, whose size is the number of nodes. (i) Indicates whether node i has been visited, true if visited, false if not visited, starting from the source node s, mark the current node as N c ,

[0018] A2. Judge N c Is it the destination node d? If so, output the lowest common mode risk path, otherwise go to A3;

[0019] A3. Obtain N according to the topological structure graph G or subgraph G' c All next-hop nodes v x , save to the list collection;

[0020] A4. Calculate the impact factors of remaining bandwidth utilization η and node risk R N and link risk R L , and weight the three influencing factors to obtain the evaluation index weight factor λ of the common mode risk;

[0021] A5. Calculate N based on the latitude and longitude information of the stations in the set list. c All v x The cost function value cost:

[0022] cost=f(s,v x )+λ×g(v x ,d)

[0023] Among them, f(s,v i ) is the source node s to v x The sum of the distance weights; g(v x ,d) is v x The estimated distance to the target node d in terms of latitude and longitude;

[0024] A6. Select all v x The node with the smallest cost function value is recorded as Node, and it is determined whether Node has been visited. If so, the Node is deleted from the set list and the execution continues to A6. Otherwise, it goes to A7.

[0025] A7. Add Node to the lowest common mode risk path and update N c For Node, return A2.

[0026] The beneficial effect of this technical solution is that it establishes a cost function based on the directional characteristics of the source and target nodes, and by selecting the node with the lowest cost as the next hop, it avoids the possibility of traversing all breadth-first search paths. In the search implementation, the relative position of the source and target nodes is used to greatly reduce the number of redundant search paths and the time complexity of the algorithm during the search process.

[0027] Furthermore, the remaining bandwidth utilization η, node risk R N and link risk R L The calculation formulas are:

[0028] R N =f·p N (i) ,i=1,2,3,4,R L =f·A0 l

[0029]

[0030] Among them, R N is the node risk; R L is the common mode link risk; p N (i)is the failure probability of substations at different voltage levels; f is the importance factor of the service request; l is the link length; B max is the maximum bandwidth capacity of the link; S is the set of all the services in operation on the link; B i is the bandwidth corresponding to the i-th service request on the link; A0 is the fiber availability per unit length.

[0031] The beneficial effects of the above technical solution are: This solution is based on A * The idea of the algorithm is to establish the cost f from the source node to the current node and the estimated cost g from the current node to the destination node. By using the different weights of the common mode risk on the estimated cost, the algorithm dynamically searches for a routing search algorithm with the lowest common mode risk as the optimization goal. During the implementation of the algorithm, the algorithm flow is simplified and the execution efficiency of the algorithm is improved through operations such as loop detection and subgraph generation judgment.

[0032] Furthermore, before returning to S2, step S6 and step S7 further include:

[0033] B1. Based on the added timestamp, determine whether there is a route in the working state set that has completed the business request transmission at the next moment. If so, proceed to B2. Otherwise, directly return to S2.

[0034] B2. Release the link in the working state set that has completed the service request transmission, and use the common mode risk detection method to update the common mode risk value and working state set of the network topology structure G at the next moment, and then return to S2.

[0035] The beneficial effects of this technical solution are as follows: Taking into account the real-time nature of power grid services, this solution introduces timestamp detection of service transmission status, enabling the design of routes with minimal common mode risk. This solution also considers link bandwidth utilization during route searches, integrating the search process with real-time link bandwidth to achieve better balance in network service transmission.

[0036] Furthermore, when the service request H requires a backup route, a backup route search method is also included:

[0037] C1. Calculate the common mode risk value of the lowest common mode risk path as the reference common mode risk;

[0038] C2. Use the k-path search algorithm to search for all alternative paths in the network topology G whose common mode risk values are less than the reference common mode risk, and retain the alternative paths that meet the common mode risk optimization model;

[0039] C3. Determine whether the number of retained alternative paths is greater than or equal to two. If so, proceed to C4. Otherwise, return the backup route of service request H as empty.

[0040] C4. Delete the lowest common mode risk path of service request H in the working state set, select two alternative paths with the lowest common mode risk values, add them to the working state set as the primary route and the backup route, and allocate bandwidth.

[0041] The beneficial effects of the above technical solution are: in the common mode risk analysis of power communication services, some services (such as dispatching services, power protection, etc.) have extremely high requirements for real-time network transmission, and it is necessary to find primary and backup routes to prevent the inability to complete business transmission when a single path fails or to take extra time to reroute. This solution searches based on the common mode risk value of the lowest common mode risk path, which can ensure that the searched backup route has a smaller common mode risk, thereby ensuring the stability of business request transmission.

[0042] Furthermore, the calculation formula of the common mode risk optimization model is:

[0043] minR CM

[0044]

[0045] Among them, R CM is the common mode risk value; R Ni is the node risk; R Lj is the common mode link risk; N is the number of services carried on the link; A is the availability of the entire link; A min is the minimum availability of the entire link; e (i,j) is a link connecting node i and node j; for e (i,j) bandwidth capacity; for e (i,j) The maximum bandwidth capacity of the power communication network; E is the set of links in the power communication network; T Ni is the delay of the i-th node; T Lj is the delay of the jth link; T max is the maximum allowed delay limit; L Ni is the packet loss rate of the i-th node; L Lj is the packet loss rate of the jth link; L max is the maximum packet loss rate allowed by the link; V={v1,v2,…,v n} is a set of n nodes in the power communication network; i, j and s are node i, node j and request s respectively.

[0046] The beneficial effect of the above technical solution is that during the transmission of services, the risk of failure in the power communication network mainly comes from a variety of factors such as equipment and line quality, service distribution, severe extreme weather, and unexpected operations and maintenance management. Common-mode risk is mainly affected by the equipment and links at the communication site. When a device or link fails, all services carried by the device or link will fail, triggering common-mode risk. Therefore, the solution considers common-mode risk caused by fiber link failure or communication site failure, and only considers the failure of a single link. Based on the risk of equipment and link failure, a probabilistic and statistical common-mode risk assessment model for the power communication network is established. This ensures that the selected transmission paths have high security and the smooth completion of service requests.

[0047] Furthermore, the method of searching for alternative paths using the k-path search algorithm includes:

[0048] D1. Create an array visited′ and an alternative path queue B, visited ( ' i) Whether node i has been visited, true if it has been visited, false if it has not been visited;

[0049] D2. Search the topology graph G for the lowest common mode risk path that satisfies the service request H as the reference path P. r , add it to the alternative path queue B, and its common mode risk value is used as the reference common mode risk R CM (ts) ;

[0050] D3, starting from the destination node d of the business request H, tracing back in a loop, and recording the path P r The node adjacent to the destination node d is the current node N c ;

[0051] D4. Judge N c Are all next hop nodes selected? If so, go to step D7. Otherwise, select N. c No node is selected in the next hop node, and it is determined that it constitutes link P curr Is the cumulative common mode risk value greater than R CM (ts) If yes, go to D7, otherwise go to D5;

[0052] D5. Determine whether the selected node appears in the candidate path queue B. If so, proceed to D7; otherwise, proceed to D6.

[0053] D6. Determine whether the latency, packet loss rate, and bandwidth all meet the requirements. If not, proceed to step D7. If they do, add it to the candidate path B, update visited′ to become the current node, and return to D4.

[0054] D7. Determine whether the selected node is the next hop node adjacent to the destination node. If so, set the number of backtracking times k = k + 1, 0 ≤ k ≤ M, and M is the path P. r The link length is then entered into D8, otherwise the previous hop of the selected node is set to the current node N. c , return to D4;

[0055] D8. Determine whether the number of backtracking times k is greater than or equal to M. If so, output the alternative path queue B. Otherwise, return to D4.

[0056] The beneficial effects of the above technical solution are as follows: this solution adopts the idea of backtracking, and the lowest common mode risk path is used as the reference path, and its risk is used as the reference common mode risk. When the common mode risk of the search path is higher than the reference common mode risk, the path selection solution is directly discarded and subsequent operations are no longer performed, thereby reducing time complexity. Considering that the length of the path mainly affects the communication delay, by finding k paths with common mode risk values lower than the reference common mode risk, on the basis of meeting the delay, they can be used as alternative paths, and the common mode risk R of each alternative path is subsequently compared. CM , screening out the optimal path solution that meets the conditions, which can reduce risks while improving the availability of network routing.

[0057] Furthermore, the method of updating the common mode risk value and the working state set of the network topology structure G at the next moment by using the common mode risk detection method includes:

[0058] E1. Calculate the common mode risk value R of the network topology G CM , read any link P in the working status set U i ;

[0059] E2, when P i When the sum of the transmission start timestamp and the service transmission time is less than the current system time, the path P is released. i , update the work status set, and judge P i Is there a common mode? If so, go to E3; otherwise, go to E4.

[0060] E3, indicating that the lowest common mode risk path and the working state concentrated link have a node intersection, calculate P i Common mode risk R CMi , and update the network common mode risk to R CM =R CM -R CMi ;

[0061] E4. When the lowest common mode risk path is added to the working state set, the common mode condition of the lowest common mode risk path for each working link in the working state set is calculated, and the common mode risk value R at the next moment is accumulated. CM(t+1) ;

[0062] E5. When the main route and backup route are added to the working state set, the common mode of the main route and backup route to each working link in the working state set is calculated respectively, and the common mode risk value R at the next moment is accumulated. CM (t+1) ;

[0063] E6. Output the common mode risk value R at the next moment CM (t+1) and work status sets.

[0064] The beneficial effects of the above technical solution are: this solution dynamically considers the impact of link state changes on the link common mode risk value. On the one hand, for routes that have completed business transmission, it is necessary to release path resources and update the state set; on the other hand, for the path P that currently needs to complete the route search, c , you need to calculate P c The common mode conditions of each working link in the status set are accumulated and updated. This allows for dynamic changes in common mode risk that are more closely aligned with actual scenarios and is more meaningful in reducing link common mode risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 Flowchart of the common mode risk search method for SDN-based power communication network.

[0066] Figure 2 This is an example diagram of the algorithm for the node-25 link G=(V,E). DETAILED DESCRIPTION

[0067] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0068] In this scheme, the power communication network is defined as an undirected graph, and the network topology is represented by the mathematical symbol G = (V, E), where V = {v1, v2, ..., v n} represents a set of n nodes in the power communication network, each node contains information such as the latitude and longitude of the site, voltage level, etc., and |V| = n; E = {e 1,2 ,e 2,3 ,…,e n-1,n} represents the set of m links in the power communication network, e i,jrepresents the link from site i to site j. Each edge contains the weight W and the number of optical fibers on each link, and |E| = m. In actual communication networks, the length (distance) of the link is used as the weight value of the link.

[0069] The services carried by the power system communication network are mainly divided into two categories: production control services and management information services. Production control services are located in security zones I, II, and III, and mainly include power automation, power consumption information collection, distribution operation monitoring, line relay protection, and the provision of transmission lines for safety automatic devices. These services do not require high bandwidth, but are very sensitive to communication security, stability, and latency, and have a high business importance. Management information services are located in zones IV and V, and mainly include various management information systems, office automation, customer service information systems, enterprise resource management systems, financial business systems, etc. These services have high bandwidth requirements but are not sensitive to latency. Currently, there are relatively mature studies on the evaluation scheme of business importance, mainly including the analytic hierarchy process (AHP) and the superiority and inferior solution distance method (TOPSIS). The principal component analysis method is greatly affected by subjective weights, and different weights will cause fluctuations in importance data. Therefore, this scheme adopts the importance of TOPSIS related literature. Let the importance of different types of services be f (i) (i=1, 2, …, 5). Table 1 shows a typical service importance evaluation of a power communication network based on TOPSIS. The bandwidth of 4×2 Mbit / s indicates that the bandwidth is 4 Mbit / s and 1+1 dual routing is required to implement link protection.

[0070] Table 1 Bandwidth and business importance of different services

[0071]

[0072] For any service request S, we first need to consider the starting point and end point of the communication. Let s and d represent the source node and destination node of the service request respectively. If there is a path P at the current moment that allows s to d to be connected, use P sd = {s,…,i,j,…,d} represents a path from the source node to the destination node. Secondly, considering the service importance f, routes with higher importance are prioritized. Furthermore, since power communication network services are constantly changing, the estimated duration t of the service transmission must also be considered. A timestamp is added to the service transmission and the estimated duration t is accumulated to determine the current service transmission status. When the service transmission ends, the link becomes idle, releasing the link occupancy and updating the network common mode risk value.

[0073] Therefore, define the i-th service request S i ={s i ,d i ,fi ,t i}, respectively representing the source node s of the i-th business i , destination node d i , the importance of the business f i and service transmission time t i .

[0074] refer to Figure 1 , Figure 1 A common mode risk search method for power communication network based on SDN is shown; Figure 1 As shown, the method includes steps S1 to S7.

[0075] In step S1, a service request is obtained, and the source node and destination node in the service request are added to a service node pair set, and links in a working state in the network topology structure G are stored in a working state set;

[0076] In step S2, it is determined whether the service node pair set is empty. If so, the search is completed. Otherwise, the links in the working state set are deleted from the network topology structure G to generate a subgraph G'.

[0077] In step S3, the business request H with the highest importance to the centralized business is selected, and the subgraph G' is searched to see whether there is a path with the lowest common mode risk that satisfies the business request H. If so, the process proceeds to S7, otherwise, the process proceeds to S4.

[0078] In one embodiment of the present invention, the method for searching for the lowest common mode risk path includes:

[0079] A1. Create an array visited, whose size is the number of nodes. (i) Indicates whether node i has been visited, true if visited, false if not visited, starting from the source node s, mark the current node as N c ;

[0080] A2. Judge N c Is it the destination node d? If so, output the lowest common mode risk path, otherwise go to A3;

[0081] A3. Obtain N according to the topological structure graph G or subgraph G' c All next-hop nodes v x , save to the list collection;

[0082] A4. Calculate the impact factors of remaining bandwidth utilization η and node risk R N and link risk R L :

[0083] R N =f·p N(i) ,i=1,2,3,4,R L =f·A0 l

[0084]

[0085] Among them, R N is the node risk; R L is the common mode link risk; p N (i) is the failure probability of substations at different voltage levels; f is the importance factor of the service request; l is the link length; B max is the maximum bandwidth capacity of the link; S is the set of all the services in operation on the link; B i is the bandwidth corresponding to the i-th service request on the link; A0 is the fiber availability per unit length.

[0086] Then, the three influencing factors are weighted to obtain the weight factor λ of the common mode risk evaluation index.

[0087] The calculation formula for the above-mentioned unit length optical fiber availability A0 is:

[0088]

[0089] Among them, t f is the mean time to failure; t r is the average recovery time; p f is the link failure probability;

[0090] A5. Calculate N based on the latitude and longitude information of the stations in the set list. c All v x The cost function value cost:

[0091] cost=f(s,v x )+λ×g(v x ,d)

[0092] Among them, f(s,v i ) is the source node s to v x The sum of the distance weights; g(v x ,d) is v x The estimated distance to the target node d in terms of latitude and longitude;

[0093] A6. Select all v x The node with the smallest cost function value is recorded as Node, and it is determined whether Node has been visited. If so, the Node is deleted from the set list and the execution continues to A6. Otherwise, it goes to A7.

[0094] A7. Add Node to the lowest common mode risk path and update N c For Node, return A2.

[0095] In step S4, the network topology G is searched for a path with the lowest common mode risk that satisfies the service request H. If so, the process proceeds to S5, otherwise, the process proceeds to S6;

[0096] In step S5, it is determined whether the lowest common mode risk path satisfies the common mode risk optimization model. If so, the process proceeds to S7; otherwise, the process proceeds to S6;

[0097] During implementation, the calculation formula of the preferred common model risk optimization model of this scheme is:

[0098] minR CM

[0099]

[0100] Among them, R CM is the common mode risk value; R Ni is the node risk; R Lj is the common mode link risk; N is the number of services carried on the link; A is the availability of the entire link; A min is the minimum availability of the entire link; e (i,j) is a link connecting node i and node j; B e(i,j) for e (i,j) Bandwidth capacity; B emax for e (i,j) The maximum bandwidth capacity of the power communication network; E is the set of links in the power communication network; T Ni is the delay of the i-th node; T Lj is the delay of the jth link; T max is the maximum allowed delay limit; L Ni is the packet loss rate of the i-th node; L Lj is the packet loss rate of the jth link; L max is the maximum packet loss rate allowed by the link; V={v1,v2,…,v n} is a set of n nodes in the power communication network; i, j and s are node i, node j and request s respectively.

[0101] The formula for calculating the availability of the entire link is:

[0102]

[0103] Among them, p Ni is the failure probability of substations at different voltage levels, P Nis the average failure probability of a node, where N is the abbreviation of node, not the number of services carried on the link mentioned above; i has four states; its value is determined according to the classification criteria in Table 2.

[0104] Table 2 Failure probability of substations of different levels

[0105]

[0106] In step S6, it is returned that the service request H does not find the lowest common mode risk path, and the service request H is deleted from the working status set, and then the process returns to S2;

[0107] During implementation, the solution preferably includes the following steps in step S6 and step S7 before returning to step S2:

[0108] B1. Based on the added timestamp, determine whether there is a route in the working state set that has completed the business request transmission at the next moment. If so, proceed to B2. Otherwise, directly return to S2.

[0109] B2. Release the link in the working state set that has completed the service request transmission, and use the common mode risk detection method to update the common mode risk value and working state set of the network topology structure G at the next moment, and then return to S2.

[0110] This solution addresses the real-time nature of power grid services by introducing timestamps to detect service transmission status, enabling the design of routes with minimal common mode risk. This solution also considers link bandwidth utilization during route searches, integrating the search process with real-time link bandwidth to achieve better balance in network service transmission.

[0111] In step S7 , the lowest common mode risk path is added to the working state set, bandwidth is allocated, and the service request H is deleted from the working state set, and then the process returns to S2 .

[0112] In one embodiment of the present invention, when the service request H requires a backup route, a backup route search method is also included:

[0113] C1. Calculate the common mode risk value of the lowest common mode risk path as the reference common mode risk;

[0114] C2. Use the k-path search algorithm to search for all alternative paths in the network topology G whose common mode risk values are less than the reference common mode risk, and retain the alternative paths that meet the common mode risk optimization model;

[0115] C3. Determine whether the number of retained alternative paths is greater than or equal to two. If so, proceed to C4. Otherwise, return the backup route of service request H as empty.

[0116] C4. Delete the lowest common mode risk path of service request H in the working state set, select two alternative paths with the lowest common mode risk values, add them to the working state set as the primary route and the backup route, and allocate bandwidth.

[0117] During implementation, this solution preferably uses a k-path search algorithm to search for alternative paths, including:

[0118] D1. Create an array visited′ and an alternative path queue B, visited ( ' i) Whether node i has been visited, true if it has been visited, false if it has not been visited;

[0119] D2. Search the topology graph G for the lowest common mode risk path that satisfies the service request H as the reference path P. r , add it to the alternative path queue B, and its common mode risk value is used as the reference common mode risk R CM (ts) ;

[0120] D3, starting from the destination node d of the business request H, tracing back in a loop, and recording the path P r The node adjacent to the destination node d is the current node N c ;

[0121] D4. Judge N c Are all next-hop nodes selected? If so, proceed to step D7. Otherwise, select N. c No node is selected in the next hop node, and it is determined that it constitutes link P curr Is the cumulative common mode risk value greater than R CM (ts) If yes, go to D7, otherwise go to D5;

[0122] D5. Determine whether the selected node appears in the candidate path queue B. If so, proceed to D7; otherwise, proceed to D6.

[0123] D6. Determine whether the latency, packet loss rate, and bandwidth all meet the requirements. If not, proceed to step D7. If they do, add it to the candidate path B, update visited′ to become the current node, and return to D4.

[0124] D7. Determine whether the selected node is the next hop node adjacent to the destination node. If so, set the number of backtracking times k = k + 1, 0 ≤ k ≤ M, and M is the path P. r The link length is then entered into D8, otherwise the previous hop of the selected node is set to the current node N. c , return to D4;

[0125] D8. Determine whether the number of backtracking times k is greater than or equal to M. If so, output the alternative path queue B. Otherwise, return to D4.

[0126] This solution adopts the idea of backtracking, and the lowest common mode risk path is used as the reference path, and its risk is used as the reference common mode risk. When the common mode risk of the searched path is higher than the reference common mode risk, the path selection solution is directly discarded and subsequent operations are no longer performed, thus reducing time complexity. Considering that the length of the path mainly affects the communication delay, by finding k paths with common mode risk values lower than the reference common mode risk, on the basis of meeting the delay, they can be used as alternative paths, and the common mode risk R of each alternative path is subsequently compared. CM , screening out the optimal path solution that meets the conditions, which can reduce risks while improving the availability of network routing.

[0127] During implementation, this solution preferably adopts a common mode risk detection method to update the common mode risk value and working state set of the network topology structure G at the next moment, including:

[0128] E1. Calculate the common mode risk value R of the network topology G CM , read any link P in the working status set U i ;

[0129] E2, when P i When the sum of the transmission start timestamp and the service transmission time is less than the current system time, the path P is released. i , update the work status set, and judge P i Is there a common mode? If so, go to E3; otherwise, go to E4.

[0130] E3, indicating that the lowest common mode risk path and the working state concentrated link have a node intersection, calculate P i Common mode risk R CMi , and update the network common mode risk to R CM =R CM -R CMi ;

[0131] E4. When the lowest common mode risk path is added to the working state set, the common mode condition of the lowest common mode risk path for each working link in the working state set is calculated, and the common mode risk value R at the next moment is accumulated. CM (t+1) ;

[0132] E5. When the main route and backup route are added to the working state set, the common mode of the main route and backup route to each working link in the working state set is calculated respectively, and the common mode risk value R at the next moment is accumulated. CM (t+1) ;

[0133] E6. Output the common mode risk value R at the next moment CM (t+1) and work status sets.

[0134] The following is an example to illustrate the performance of this solution.

[0135] by Figure 2 For example, the topology consists of 15 nodes and 25 links. Assume that at time t, the service request is S = {s1, s2}, where s1 = {v2, v 15 ,0.62,3000},s2={v1,v 11 ,0.29,2500}The following introduces the routing search process for completing the above service request S.

[0136] Two business requests at the same time are sorted by business importance f, and the first business request s1 is routed first. At this time, the working state set U is an empty set, and the first routing is solved by searching for the lowest common mode risk path to obtain p1= <v2,v7,v9,v 13 ,v 15 >, check the constraints (common mode risk optimization model) and do not need backup routes. Update p1 directly to U, such as Figure 2 As shown in (a), calculate the common mode risk value R CM (t) ;

[0137] The second service request s2 first deletes the working link in U. At this time, v1 to v 11 There is still a connected link, so p2 = <v1,v4,v8,v 11 >, check the constraints (common mode risk optimization model) and no backup route is required. Update p2 to U, such as Figure 2 As shown in (b), calculate and update R CM (t+1) .

[0138] Assume that at time t+τ, there is a service request S={s3}, s3={v3,v 12 ,0.94,4000},When the third service request is made, the working link in U is first deleted. It can be found that there is no connected link in the subgraph G′ of GU. At this time, the service request has not searched for the lowest common mode risk path. In step S6, before returning to step S2, it is determined whether the current service needs a backup route. It is found that a backup route is needed. Therefore, the backup route search method is used to search the main route and the backup route. First, the method of searching the lowest common mode risk path is used to calculate the lowest common mode risk link p3 in the network topology structure G. (0) = <v3,v4,v8,v9,v 12 > and reference common mode risks

[0139] Based on p3 (0) The common mode points and common mode paths are traced back using the k-path search algorithm, and the top three paths with the lowest risk values are calculated as p3 (1) = <v3,v4,v5,v9,v 12 >, p3 (2) = <v3,v6,v8,v9,v 12 >, p3 (3) = <v3,v6,v 10 ,v 11 ,v 13 ,v 12 >. Figure 2 (df) shows that the first two paths with the lowest risk are p3 (1) and p3 (2) , therefore, p3 (1) That is the optimal main route P searched under the current request m , p3 (2) The optimal backup route P searched for in the current request m .

[0140] Compared to traditional shortest path algorithms such as Dijkstra and Floyd, this solution's common-mode risk search method for power communication networks avoids the need to traverse all breadth-first searches. The method's implementation uses the relative position of source and target nodes to significantly reduce the number of redundant paths searched.

[0141] Consider the network topology of G = (V, E), where the number of nodes is n and the number of links is m. Let the maximum degree of the nodes in the network be δ, the average number of communication sites passed by the link be h, and the number of service requests be r. Then the time complexity of this solution is O(hr). For service request R i (i=1), when performing single path search (SPBA), since the working state set U of the link will not affect the search process, the algorithm searches a maximum of δ nodes each time, searches h times on average, and the time complexity is O(δh); for service request R i (i≥2), the routing situation in U needs to be considered. At this time, the running time of k-LCMP with the lowest common mode risk (k-LCMP) requires (O(δh+δh 2 r 2 (n+m))), therefore, the overall complexity of the algorithm is O(δh 2 r 2 (n+m)).

[0142] By O(δh 2 r 2The time complexity (n+m)) quantifies the time complexity of the algorithm proposed in this scheme, indicating that the algorithm of this scheme has low time complexity and good time performance.

Claims

1. A common mode risk search method for power communication network based on SDN, characterized in that: include: S1. Obtain a service request, add the source node and destination node in the service request to a service node pair set, and store the links in the working state in the network topology structure G into a working state set; S2. Determine whether the service node pair set is empty. If so, complete the search. Otherwise, delete the links in the working state set in the network topology structure G to generate a subgraph G'. S3. Select the business request H with the highest importance to the centralized business from the business node, and search the subgraph G' to see if there is a path with the lowest common mode risk that satisfies the business request H. If so, proceed to S7; otherwise, proceed to S4. S4. Search the network topology G to see if there is a path with the lowest common mode risk that satisfies the service request H. If so, proceed to S5; otherwise, proceed to S6. S5, determine whether the lowest common mode risk path satisfies the common mode risk optimization model. If so, proceed to S7; otherwise, proceed to S6; S6. Return that the business request H has not found the lowest common mode risk path, delete the business request H from the working status set, and then return to S2. S7. Add the lowest common mode risk path to the working state set, allocate bandwidth, and delete the service request H from the working state set, then return to S2.

2. The common mode risk search method for power communication network according to claim 1, characterized in that: Methods for searching for the path with the lowest common mode risk include: A1. Create an array visited, whose size is the number of nodes. (i) Indicates whether node i has been visited, true if visited, false if not visited, starting from the source node s, mark the current node as N c , A2. Judge N c Is it the destination node d? If so, output the lowest common mode risk path, otherwise go to A3; A3. Obtain N according to the topological structure graph G or subgraph G' c All next-hop nodes v x , save to the list collection; A4. Calculate the impact factors of remaining bandwidth utilization η and node risk R N and link risk R L , and weight the three influencing factors to obtain the evaluation index weight factor λ of the common mode risk; A5. Calculate N based on the latitude and longitude information of the stations in the set list. c All v x The cost function value cost: cost=f(s,v x )+λ×g(v x ,d) Among them, f(s,v i ) is the source node s to v x The sum of the distance weights; g(v x ,d) is v x The estimated distance to the target node d in terms of latitude and longitude; A6. Select all v x The node with the smallest cost function value is recorded as Node, and it is determined whether Node has been visited. If so, the Node is deleted from the set list and the execution continues to A6. Otherwise, it goes to A7. A7. Add Node to the lowest common mode risk path and update N c For Node, return A2.

3. The common mode risk search method for power communication network according to claim 2, characterized in that: Remaining bandwidth utilization η, node risk R N and link risk R L The calculation formulas are: R N =f·p N (i) ,i=1,2,3,4,R L =f·A0 l Among them, R N is the node risk; R L is the common mode link risk; p N (i) is the failure probability of substations at different voltage levels; f is the importance factor of the service request; l is the link length; B max is the maximum bandwidth capacity of the link; S is the set of all the services in operation on the link; B i is the bandwidth corresponding to the i-th service request on the link; A0 is the fiber availability per unit length.

4. The common mode risk search method for power communication network according to claim 1, characterized in that: Steps S6 and S7 also include the following steps before returning to S2: B1. Based on the added timestamp, determine whether there is a route in the working state set that has completed the business request transmission at the next moment. If so, proceed to B2. Otherwise, directly return to S2. B2. Release the link in the working state set that has completed the service request transmission, and use the common mode risk detection method to update the common mode risk value and working state set of the network topology structure G at the next moment, and then return to S2.

5. The common mode risk search method for power communication network according to claim 1, characterized in that: When service request H requires a backup route, the backup route search method is also included: C1. Calculate the common mode risk value of the lowest common mode risk path as the reference common mode risk; C2. Use the k-path search algorithm to search for all alternative paths in the network topology G whose common mode risk values are less than the reference common mode risk, and retain the alternative paths that meet the common mode risk optimization model; C3. Determine whether the number of retained alternative paths is greater than or equal to two. If so, proceed to C4. Otherwise, return the backup route of service request H as empty. C4. Delete the lowest common mode risk path of service request H in the working state set, select two alternative paths with the lowest common mode risk values, add them to the working state set as the primary route and the backup route, and allocate bandwidth.

6. The method for searching common mode risk in a power communication network according to claim 1 or 5, characterized in that: The calculation formula of the common model risk optimization model is: myR CM Among them, R CM is the common mode risk value; R Ni is the node risk; R Lj is the common mode link risk; N is the number of services carried on the link; A is the availability of the entire link; A min is the minimum availability of the entire link; e (i,j) is a link connecting node i and node j; for e (i,j) bandwidth capacity; for e (i,j) The maximum bandwidth capacity of the power communication network; E is the set of links in the power communication network; T Ni is the delay of the i-th node; T Lj is the delay of the jth link; T max is the maximum allowed delay limit; L Ni is the packet loss rate of the i-th node; L Lj is the packet loss rate of the jth link; L max is the maximum packet loss rate allowed by the link; V={v1,v2,…,v n } is a set of n nodes in the power communication network; i, j and s are node i, node j and request s respectively.

7. The common mode risk search method for power communication network based on SDN according to claim 1, characterized in that: Methods for searching alternative paths using the k-path search algorithm include: D1. Create an array visited′ and an alternative path queue B, visited ( ' i) Whether node i has been visited, true if it has been visited, false if it has not been visited; D2. Search the topology graph G for the lowest common mode risk path that satisfies the service request H as the reference path P. r , add it to the alternative path queue B, and its common mode risk value is used as the reference common mode risk R CM (ts) ; D3, starting from the destination node d of the business request H, tracing back in a loop, and recording the path P r The node adjacent to the destination node d is the current node N c ; D4. Judge N c Are all next hop nodes selected? If yes, go to step D7. Otherwise, select N. c No node is selected in the next hop node, and it is determined that it constitutes link P curr Is the cumulative common mode risk value greater than R CM (ts) If yes, go to D7, otherwise go to D5; D5. Determine whether the selected node appears in the candidate path queue B. If so, proceed to D7; otherwise, proceed to D6. D6. Determine whether the latency, packet loss rate, and bandwidth all meet the requirements. If not, proceed to step D7. If they do, add it to the candidate path B, update visited′ to become the current node, and return to D4. D7. Determine whether the selected node is the next hop node adjacent to the destination node. If so, set the number of backtracking times k = k + 1, 0 ≤ k ≤ M, and M is the path P. r The link length is then entered into D8, otherwise the previous hop of the selected node is set to the current node N. c , return to D4; D8. Determine whether the number of backtracking times k is greater than or equal to M. If so, output the alternative path queue B. Otherwise, return to D4.

8. The common mode risk search method for power communication network according to claim 4, characterized in that: The method for updating the common mode risk value and the working state set of the network topology structure G at the next moment by using the common mode risk detection method includes: E1. Calculate the common mode risk value R of the network topology G CM , read any link P in the working status set U i ; E2, when P i When the sum of the transmission start timestamp and the service transmission time is less than the current system time, the path P is released. i , update the work status set, and judge P i Is there a common mode? If so, go to E3; otherwise, go to E4. E3, indicating that the lowest common mode risk path and the working state concentrated link have a node intersection, calculate P i Common mode risk R CMi , and update the network common mode risk to R CM =R CM -R CMi ; E4. When the lowest common mode risk path is added to the working state set, the common mode condition of the lowest common mode risk path for each working link in the working state set is calculated, and the common mode risk value R at the next moment is accumulated. CM (t+1) ; E5. When the main route and backup route are added to the working state set, the common mode of the main route and backup route to each working link in the working state set is calculated respectively, and the common mode risk value R at the next moment is accumulated. CM (t+1) ; E6. Output the common mode risk value R at the next moment CM (t+1) and work status sets.

Images