Virtual power plant communication path selection method and system

Through improved genetic algorithms, a multi-objective optimization objective function is built in a communication network, balancing latency, cost, bandwidth and security, solving the multi-objective optimization, flexibility and security problems of path selection in the prior art, and realizing network performance and improve security.

CN119030912APending Publication Date: 2024-11-26EAST CHINA ELECTRIC POWER TEST & RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410933842.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Existing communication network path selection methods ignore multi-objective optimization, lack of flexibility, uneven resource utilization and security issues, resulting in network performance bottlenecks, inefficiency and instability.

Method used

Using improved genetic algorithms, a multi-objective optimization objective function is constructed by obtaining network status information, the communication path of virtual power plants is calculated, the delay, cost, bandwidth utilization and security is balanced, and the path selection is dynamically adjusted to ensure network performance and security.

Benefits of technology

It realizes multi-objective optimization of network performance, dynamically adapts to network changes, balances resource utilization, enhances network security and reliability, and reduces maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119030912A_ABST
    Figure CN119030912A_ABST
Patent Text Reader

Abstract

The invention relates to a virtual power plant communication path selection method and system, and the method comprises the steps: obtaining the network state information of a virtual power plant, including the bandwidth use condition, delay, cost and safety related data of each link; according to the method, a network performance key index multi-objective optimization objective function is constructed according to the total delay, the total cost, the bandwidth utilization rate and the security score of a network path, an optimal path is calculated by using an improved genetic algorithm based on collected data, and the improved genetic algorithm takes the network path as an individual. According to the multi-objective optimization objective function, the fitness of each individual is obtained through calculation, genetic optimization solution is carried out, and the optimal path of virtual power plant communication is obtained; and deploying the optimal path calculated through the improved genetic algorithm into network equipment. Compared with the prior art, the invention provides a more advanced and flexible solution, which not only can improve the network performance and efficiency, but also can enhance the reliability and security of the network, thereby bringing long-term value for users.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of communication path optimization, and in particular to a method and system for selecting a communication path for a virtual power plant based on an improved genetic algorithm. Background Art

[0002] In modern communication networks, network topology refers to the geometric arrangement of nodes (such as routers, switches, terminal devices, etc.) and the links connecting these nodes. Effective communication path selection, i.e. routing, is a core issue in network design, which directly affects the performance and efficiency of the network.

[0003] Path selection method of existing communication network:

[0004] 1) Shortest path algorithm: Traditional network path selection usually uses algorithms such as Dijkstra algorithm or Bellman-Ford algorithm to find the shortest path between two points. This method mainly determines the path based on the minimum number of hops or the minimum cumulative delay, and its purpose is to reduce transmission time and improve transmission efficiency. 2) Traffic engineering technology: For example, MPLS (Multi-Protocol Label Switching) is used to create predefined paths, which allows network administrators to optimize data flows based on predicted network traffic patterns, optimize network resource utilization and increase bandwidth utilization by managing the routing of data packets.

[0005] The existing methods have the following shortcomings:

[0006] Ignoring multi-objective optimization: Although shortest path algorithms can find the path with the lowest cost, they usually ignore other key network performance indicators, which may lead to performance bottlenecks in applications such as multimedia transmission and large-scale data processing.

[0007] Lack of flexibility: Traditional algorithms often lack flexibility when the network topology changes, requiring recalculation of paths rather than dynamically adapting to changes in network status. This can lead to inefficiencies in highly dynamic modern network environments, such as mobile communications and cloud computing scenarios.

[0008] Uneven resource utilization: The shortest path algorithm may cause some links to be overloaded while other links are idle. This uneven resource utilization will further aggravate network congestion and instability.

[0009] Security issues: Security is also an important aspect when considering network paths, but traditional algorithms often do not consider security risks as a factor in path selection. This can lead to serious consequences when facing network attacks such as DDoS attacks.

[0010] Although existing network path selection technologies provide basic frameworks and solutions, there is still room for improvement in terms of multi-objective optimization, flexibility, resource balancing, and security. Summary of the invention

[0011] The purpose of the present invention is to overcome the defects of the above-mentioned prior art such as neglect of multi-objective optimization, lack of flexibility, uneven resource utilization and safety issues, and to provide a virtual power plant communication path selection method and system based on an improved genetic algorithm.

[0012] The purpose of the present invention can be achieved by the following technical solutions:

[0013] As a first aspect of the present invention, a method for selecting a communication path for a virtual power plant is provided, the steps comprising:

[0014] Obtain network status information of the virtual power plant, including bandwidth usage, latency, cost and security-related data of each link;

[0015] The multi-objective optimization objective function of key network performance indicators is constructed based on the total delay, total cost, bandwidth utilization and security score of the network path;

[0016] Based on the collected data, an improved genetic algorithm is used to calculate the optimal path. The improved genetic algorithm uses the network path as an individual, calculates the fitness of each individual according to the multi-objective optimization objective function, and performs genetic optimization to obtain the optimal path for virtual power plant communication.

[0017] The optimal path calculated by the improved genetic algorithm is deployed to the network devices.

[0018] As a preferred technical solution, the multi-objective optimization objective function is used to balance multiple key network performance indicators in network path selection:

[0019] F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p)

[0020] Where p represents the selected network path; D(p) is the total delay of path p; C(p) is the total cost of path p; L(p) is the bandwidth utilization of path p; S(p) is the security score of path p; α, β, γ, and δ are user-defined weight coefficients.

[0021] As a preferred technical solution, the constraints of the multi-objective optimization objective function include:

[0022] Bandwidth constraint: The minimum bandwidth of path p is greater than or equal to the preset minimum bandwidth requirement;

[0023] Node capacity constraint: the maximum processing capacity of node n; greater than or equal to the data load carried by the current node n;

[0024] Link cost constraint: the cost of a single link is less than or equal to the maximum acceptable cost of the path;

[0025] Safety constraint: indicates that the safety score of path p is greater than or equal to the preset minimum safety standard.

[0026] As a preferred technical solution, the solution process of the improved genetic algorithm is as follows:

[0027] Generate a set of random solutions, where each individual represents a possible network path;

[0028] Repeat the following steps until the termination condition is met:

[0029] Calculate fitness: Calculate the fitness of each individual based on the objective function;

[0030] Selection operation: individuals are selected for reproduction based on fitness, and individuals with higher fitness have a greater chance of being selected;

[0031] Crossover operation: The selected individuals generate new individuals through crossover operation, and the crossover point is randomly selected;

[0032] Mutation operation: modify certain parts of an individual with a certain probability;

[0033] Replacement operation: Use newly generated individuals to replace individuals with lower fitness in the original population.

[0034] As a preferred technical solution, the crossover operation uses single-point crossover; the selection operation uses roulette wheel selection.

[0035] As a preferred technical solution, the termination condition of the improved genetic algorithm includes: reaching a maximum number of iterations or the fitness reaching a preset threshold.

[0036] As a preferred technical solution, the method monitors the network operation status and feeds back the network status changes to the improved genetic algorithm to dynamically adjust the path selection.

[0037] As a second aspect of the present invention, a virtual power plant communication path selection system is provided, the system comprising:

[0038] Data collection layer: responsible for collecting network status information, including bandwidth usage, latency, cost, and security-related data of each link;

[0039] Path calculation layer: Based on the collected data, an improved genetic algorithm is used to calculate the optimal path. The improved genetic algorithm uses the network path as an individual, calculates the fitness of each individual according to the multi-objective optimization objective function, and performs genetic optimization to obtain the optimal path for virtual power plant communication;

[0040] Execution layer: deploys the calculated optimal path to network devices;

[0041] Monitoring and feedback layer: monitors the network operation status and dynamically adjusts path selection according to changes in network status.

[0042] As a preferred technical solution, the multi-objective optimization objective function in the path calculation layer is used to balance multiple key network performance indicators in network path selection:

[0043] F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p)

[0044] Where p represents the selected network path; D(p) is the total delay of path p; C(p) is the total cost of path p; L(p) is the bandwidth utilization of path p; S(p) is the security score of path p; α, β, γ, and δ are user-defined weight coefficients.

[0045] As a preferred technical solution, the solution process of the improved genetic algorithm in the path calculation layer is specifically as follows:

[0046] Generate a set of random solutions, where each individual represents a possible network path;

[0047] Repeat the following steps until the termination condition is met:

[0048] Calculate fitness: Calculate the fitness of each individual based on the objective function;

[0049] Selection operation: individuals are selected for reproduction based on fitness, and individuals with higher fitness have a greater chance of being selected;

[0050] Crossover operation: The selected individuals generate new individuals through crossover operation, and the crossover point is randomly selected;

[0051] Mutation operation: modify certain parts of an individual with a certain probability;

[0052] Replacement operation: Use newly generated individuals to replace individuals with lower fitness in the original population.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1) The improved genetic algorithm proposed in the present invention can achieve multi-objective optimization of network performance. It does not simply pursue the shortest path, but optimizes based on multiple key network performance indicators, such as bandwidth, cost, delay and security. The final path may choose a route with larger bandwidth or lower cost when the delay is slightly higher, or choose a more secure link. While obtaining a shorter communication path, the communication performance and security of the network can be guaranteed.

[0055] 2) The method proposed in the present invention can dynamically adjust the decision of the improved genetic algorithm by monitoring the network status information in real time and feeding the network status information back to the improved genetic algorithm, and optimize the path selection according to the real-time feedback, which is particularly valuable in the real world environment where the network status changes frequently.

[0056] 3) The genetic algorithm proposed in the present invention helps to balance network load and reduce link congestion by taking global considerations and optimizing resource allocation. This method can balance resource utilization, extend the life of network equipment, and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flowchart of the implementation of the virtual power plant communication path selection method of the present invention;

[0058] Figure 2 The figure is a comparison chart of the results of using the improved genetic algorithm and the traditional Dijkstra algorithm for the same network topology. DETAILED DESCRIPTION

[0059] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0060] Example 1

[0061] As one implementation of the present invention, this embodiment provides a virtual power plant communication path selection method based on an improved genetic algorithm, such as Figure 1 As shown, the specific steps of the method include:

[0062] S1. Collect status information of the virtual power plant communication network, including bandwidth usage, latency, cost and security-related data of each link.

[0063] S2. Based on the collected data, an improved genetic algorithm is used to calculate the optimal path. The improved genetic algorithm takes the network path as an individual, constructs a multi-objective optimization objective function of multiple key network performance indicators including the total delay, cost, bandwidth utilization and security score of the path p, and calculates the fitness of each individual based on the objective function, performs genetic optimization solution, and obtains the optimal path for virtual power plant communication.

[0064] S3. Deploy the optimal path calculated by the improved genetic algorithm to network devices such as routers and switches.

[0065] S4. Monitor the operating status of the virtual power plant communication network in real time, and dynamically adjust path selection according to changes in network status.

[0066] 2. Objective Function

[0067] Comprehensive performance indicator formula, used to balance multiple key indicators in network path selection:

[0068] F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p)

[0069] Where p represents the selected network path; D(p) is the total delay of path p, usually measured in milliseconds; C(p) is the total cost of path p, which may include energy consumption and maintenance costs; L(p) is the bandwidth utilization of path p, expressed as the ratio of the actual used bandwidth to the total available bandwidth; S(p) is the security score of path p, and a high score indicates higher data security; α, β, γ, and δ are user-defined weight coefficients, which are used to adjust the influence of different indicators in the objective function.

[0070] 3. Constraints

[0071] Detailed constraint formula:

[0072] 3.1) Bandwidth Constraints:

[0073] Bandwidth(p)≥BW min

[0074] Where Bandwidth(p) represents the minimum bandwidth of path p; BW min The preset minimum bandwidth requirement.

[0075] 3.2) Node Capacity Constraints:

[0076] Capacity(n)≥Load(n)

[0077] In the formula, Capacity(n) represents the maximum processing capacity of node n; Load(n) is the data load carried by the current node n.

[0078] 3.3) Link cost constraint:

[0079]

[0080] Where Cost(link) represents the cost of a single link; C max is the maximum acceptable cost of the path.

[0081] 3.4) Safety constraints:

[0082] S(p)≥S min

[0083] Where S(p) represents the safety score of path p; S minThe default minimum safety standard.

[0084] 4. Improved genetic algorithm solution process:

[0085] Algorithm steps:

[0086] S4.1. Initialize the population: Generate a set of random solutions, each of which represents a possible network path.

[0087] S4.2. Calculate fitness: Calculate the fitness for each individual (network path), and the fitness is determined by the objective function.

[0088] S4.3, Selection operation: select individuals for reproduction based on fitness. Individuals with higher fitness have a greater chance of being selected. Methods such as roulette wheel selection can be used.

[0089] S4.4, Crossover operation: The selected individuals generate new individuals through the crossover operation. The crossover point is randomly selected to introduce diversity. Techniques such as single-point crossover can be used.

[0090] S4.5. Mutation operation: modify certain parts of an individual with a certain probability to further increase the diversity of the population.

[0091] S4.6, Replacement: Use newly generated individuals to replace individuals with lower fitness in the original population.

[0092] S4.7, termination condition: Repeat steps S4.2 to S4.6 until the termination condition is met (such as reaching the maximum number of iterations or the fitness reaches a preset threshold).

[0093] Example code (Python):

[0094]

[0095]

[0096] while len(new_population) <population_size:

[0097] p1,p2=random.sample(population,2)

[0098] c1,c2=crossover(p1,p2)

[0099] new_population.extend([mutation(c1),mutation(c2)])

[0100] population = new_population

[0101] return population

[0102] In this embodiment, a random network with 10 nodes and 20 edges is used to verify the technical effect of the present invention. Initialize the network topology: Generate a random graph in which the weights (delay, cost, bandwidth and security) of nodes and edges are randomly assigned. The improved genetic algorithm proposed in the present invention is used to find the optimal path, and the traditional Dijkstra algorithm is used to find the shortest path. The results are compared. Figure 2 As shown, the improved genetic algorithm (green solid line) of the present invention has the following advantages compared with the traditional Dijkstra algorithm (red dotted line):

[0103] 1. Network performance optimization:

[0104] Improved genetic algorithm:

[0105] The algorithm of the present invention does not simply pursue the shortest path, but optimizes based on multiple key network performance indicators (such as bandwidth, cost, delay and security). For example, the green path may choose a route with greater bandwidth or lower cost when the delay is slightly higher, or choose a more secure link.

[0106] 2. Dynamic adaptability:

[0107] The improved genetic algorithm is able to dynamically adjust its decisions to optimize path selection based on real-time feedback. This is particularly valuable in real-world environments where network status changes frequently, such as during peak traffic hours or when a link fails.

[0108] 3. Balanced resource utilization:

[0109] Genetic algorithms help balance network load and reduce link congestion by taking global considerations and optimizing resource allocation. This approach can extend the life of network equipment and reduce maintenance costs.

[0110] By comparison, the solution of the present invention provides a more advanced and flexible solution in network design and management, which is particularly suitable for modern network applications with complex requirements and dynamically changing environments. This can not only improve network performance and efficiency, but also enhance network reliability and security, thus bringing obvious long-term value to users.

[0111] Example 2

[0112] Another embodiment of the present invention provides a virtual power plant communication path selection system based on an improved genetic algorithm in this embodiment. The system executes the virtual power plant communication path selection method as described in the listing number. The specific architecture is as follows:

[0113] Data collection layer: responsible for collecting network status information, including bandwidth usage, latency, cost, and security-related data of each link.

[0114] Path calculation layer: Based on the collected data, the improved genetic algorithm proposed in the present invention is used to calculate the optimal path.

[0115] Execution layer: Deploys the calculated optimal path to network devices such as routers and switches.

[0116] Monitoring and feedback layer: monitors the network operation status and dynamically adjusts path selection according to changes in network status.

[0117] Among them, the multi-objective optimization objective function in the path calculation layer is used to balance multiple key network performance indicators in network path selection:

[0118] F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p)

[0119] Where p represents the selected network path; D(p) is the total delay of path p; C(p) is the total cost of path p; L(p) is the bandwidth utilization of path p; S(p) is the security score of path p; α, β, γ, and δ are user-defined weight coefficients.

[0120] The specific solution process of the improved genetic algorithm in the path calculation layer is as follows:

[0121] Generate a set of random solutions, where each individual represents a possible network path;

[0122] Repeat the following steps until the termination condition is met:

[0123] Calculate fitness: Calculate the fitness of each individual based on the objective function;

[0124] Selection operation: individuals are selected for reproduction based on fitness, and individuals with higher fitness have a greater chance of being selected;

[0125] Crossover operation: The selected individuals generate new individuals through crossover operation, and the crossover point is randomly selected;

[0126] Mutation operation: modify certain parts of an individual with a certain probability;

[0127] Replacement operation: Use newly generated individuals to replace individuals with lower fitness in the original population.

[0128] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for selecting a communication path for a virtual power plant, characterized in that the steps include: Obtain network status information of the virtual power plant, including bandwidth usage, latency, cost and security-related data of each link; The multi-objective optimization objective function of key network performance indicators is constructed based on the total delay, total cost, bandwidth utilization and security score of the network path; Based on the collected data, an improved genetic algorithm is used to calculate the optimal path. The improved genetic algorithm uses the network path as an individual, calculates the fitness of each individual according to the multi-objective optimization objective function, and performs genetic optimization to obtain the optimal path for virtual power plant communication. The optimal path calculated by the improved genetic algorithm is deployed to the network devices.

2. A method for selecting a virtual power plant communication path according to claim 1, characterized in that: The multi-objective optimization objective function is used to balance multiple key network performance indicators in network path selection: F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p) Where p represents the selected network path; D(p) is the total delay of path p; C(p) is the total cost of path p; L(p) is the bandwidth utilization of path p; S(p) is the security score of path p; α, β, γ, and δ are user-defined weight coefficients.

3. A method for selecting a communication path for a virtual power plant according to claim 2, characterized in that: The constraints of the multi-objective optimization objective function include: Bandwidth constraint: The minimum bandwidth of path p is greater than or equal to the preset minimum bandwidth requirement; Node capacity constraint: the maximum processing capacity of node n; greater than or equal to the data load carried by the current node n; Link cost constraint: the cost of a single link is less than or equal to the maximum acceptable cost of the path; Safety constraint: indicates that the safety score of path p is greater than or equal to the preset minimum safety standard.

4. A method for selecting a virtual power plant communication path according to claim 1, characterized in that: The solving process of the improved genetic algorithm is as follows: Generate a set of random solutions, where each individual represents a possible network path; Repeat the following steps until the termination condition is met: Calculate fitness: Calculate the fitness of each individual based on the objective function; Selection operation: individuals are selected for reproduction based on fitness, and individuals with higher fitness have a greater chance of being selected; Crossover operation: The selected individuals generate new individuals through crossover operation, and the crossover point is randomly selected; Mutation operation: modify certain parts of an individual with a certain probability; Replacement operation: Use newly generated individuals to replace individuals with lower fitness in the original population.

5. A method for selecting a virtual power plant communication path according to claim 4, characterized in that: The crossover operation uses single-point crossover; the selection operation uses roulette wheel selection.

6. A method for selecting a communication path for a virtual power plant according to claim 4, characterized in that: The termination conditions of the improved genetic algorithm include: reaching a maximum number of iterations or the fitness reaching a preset threshold.

7. A method for selecting a communication path for a virtual power plant according to claim 1, characterized in that: The method monitors the network operation status, and feeds back the network status change to the improved genetic algorithm to dynamically adjust the path selection.

8. A virtual power plant communication path selection system, characterized in that: The system comprises: Data collection layer: responsible for collecting network status information, including bandwidth usage, latency, cost, and security-related data of each link; Path calculation layer: Based on the collected data, an improved genetic algorithm is used to calculate the optimal path. The improved genetic algorithm uses the network path as an individual, calculates the fitness of each individual according to the multi-objective optimization objective function, and performs genetic optimization to obtain the optimal path for virtual power plant communication; Execution layer: deploys the calculated optimal path to network devices; Monitoring and feedback layer: monitors the network operation status and dynamically adjusts path selection according to changes in network status.

9. A virtual power plant communication path selection system according to claim 8, characterized in that: The multi-objective optimization objective function in the path calculation layer is used to balance multiple key network performance indicators in network path selection: F(p)=α×D(p)+β×C(p)+γ×L(p)-δ×S(p) Where p represents the selected network path; D(p) is the total delay of path p; C(p) is the total cost of path p; L(p) is the bandwidth utilization of path p; S(p) is the security score of path p; α, β, γ, and δ are user-defined weight coefficients.

10. A virtual power plant communication path selection system according to claim 8, characterized in that: The solving process of the improved genetic algorithm in the path calculation layer is as follows: Generate a set of random solutions, where each individual represents a possible network path; Repeat the following steps until the termination condition is met: Calculate fitness: Calculate the fitness of each individual based on the objective function; Selection operation: individuals are selected for reproduction based on fitness, and individuals with higher fitness have a greater chance of being selected; Crossover operation: The selected individuals generate new individuals through crossover operation, and the crossover point is randomly selected; Mutation operation: modify certain parts of an individual with a certain probability; Replacement operation: Use newly generated individuals to replace individuals with lower fitness in the original population.