CN112766633B - Flow balance-based power wireless heterogeneous network management method and device - Google Patents
Flow balance-based power wireless heterogeneous network management method and device Download PDFInfo
- Publication number
- CN112766633B CN112766633B CN202011527098.8A CN202011527098A CN112766633B CN 112766633 B CN112766633 B CN 112766633B CN 202011527098 A CN202011527098 A CN 202011527098A CN 112766633 B CN112766633 B CN 112766633B
- Authority
- CN
- China
- Prior art keywords
- flow
- prediction result
- link
- wireless heterogeneous
- heterogeneous network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007726 management method Methods 0.000 title claims description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 45
- 238000000034 method Methods 0.000 claims abstract description 37
- 230000008859 change Effects 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 34
- 238000005070 sampling Methods 0.000 claims description 18
- 238000013507 mapping Methods 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000002592 echocardiography Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Biodiversity & Conservation Biology (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mobile Radio Communication Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application provides a method and a device for managing a wireless heterogeneous network of electric power based on flow balance, wherein the method comprises the following steps: outputting a first flow prediction result of the link through the graph neural network according to the current flow data; analyzing the change trend of the current flow data to obtain a second flow prediction result of the link; analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result; and if the final prediction result exceeds a preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network. The flow prediction is carried out by two methods of graph neural network and change trend analysis respectively, and the two flow prediction results are combined, and meanwhile, the graph neural network is utilized to carry out the flow prediction, so that the method is more suitable for the power wireless heterogeneous network with complex network topology than the traditional neural network model, and the accuracy of the flow prediction is improved.
Description
Technical Field
The application belongs to the technical field of power wireless networks, and particularly relates to a power wireless heterogeneous network management method and device based on flow balance.
Background
In recent years, along with the rapid development of smart power grids, particularly the development of big data, cloud computing, internet of things and mobile services, a great deal of real-time and high-speed data demands are caused, the acquisition frequency of a power system is also increased, and the existing communication management technology is difficult to meet the requirements of wireless heterogeneous network link management, so that the stability and instantaneity of important power services are ensured. The power wireless heterogeneous network is composed of wireless networks working in different frequency bands and different technical systems, and the function of the power wireless heterogeneous network is to provide a safe and reliable data transmission channel between a remote monitoring system and a power system server, so that most power communication requirements can be met.
The common function of the power wireless heterogeneous network is to actively switch to another link when the quality of the connected link transmission network signal cannot meet the requirement, so that the smoothness of a service channel is ensured. Therefore, the flow prediction and the flow equalization on the link are directly related to the switching effect of the power wireless heterogeneous network, and because the topology structure of the power wireless heterogeneous network is complex, the factors influencing the flow prediction result at each node in the power wireless heterogeneous network are different, the traditional flow prediction method cannot be applied to the power wireless heterogeneous network, and the requirement of link management of the power wireless heterogeneous network is difficult to meet.
Disclosure of Invention
In order to solve the defects and shortcomings in the prior art, the application provides a power wireless heterogeneous network management method based on flow balance, which comprises the following steps:
collecting current flow data at nodes of each link of the power wireless heterogeneous network;
generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
analyzing the change trend of the current flow data to obtain a second flow prediction result of the link;
analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result;
and if the final prediction result exceeds a preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network.
Optionally, the current traffic data includes link total traffic, ingress and egress link node traffic, ingress and egress port traffic, link load rate, and source IP traffic.
Optionally, the generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of the link through the graph neural network, including:
identifying the flow type at each node on the link based on a fuzzy C-means algorithm;
an input matrix is formed by the current data flow, and the input matrix is input into a graph neural network corresponding to the flow type according to the identified flow type;
and outputting a first flow prediction result of the link through the graph neural network.
Optionally, the analyzing the change trend of the current flow data to obtain a second flow prediction result of the link includes:
removing data of the link at a peak-valley value of the selected time period;
calculating the average flow rate change rate of a plurality of adjacent sampling time points;
and calculating a flow value at the next moment according to the maximum flow change rate to serve as a second flow prediction result.
Optionally, the analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result includes:
calculating a difference between the first flow prediction result and the second flow prediction result;
if the difference value is smaller than a preset threshold value, the first flow prediction result is used as a final prediction result;
and if the difference value is not smaller than the preset threshold value, carrying out weighted calculation on the first flow prediction result and the second flow prediction result based on preset weight, and taking the weighted calculation result as a final prediction result.
Optionally, the power wireless heterogeneous network management method further includes, after collecting current flow data at nodes of each link of the power wireless heterogeneous network, establishing a link load mapping table;
the link load mapping table comprises a link number, a link load value and a preset flow threshold.
Optionally, if the final prediction result exceeds the preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network, including:
judging whether a final prediction result exceeds a preset flow threshold of the link or not by inquiring a link load mapping table;
if the flow threshold exceeds the preset flow threshold, searching whether an available link exists according to the routing path;
if the available link is found, the load capacity of the available link is obtained, and if the load capacity does not exceed the preset proportion of the route, echo information is sent to inhibit the source IP flow;
and if the load capacity exceeds the preset proportion of the route and more than one available link is found, distributing the flow of the power wireless heterogeneous network according to the load capacity of the available link in proportion.
The application also provides a power wireless heterogeneous network management device based on flow balance based on the same thought, which comprises:
the acquisition unit: the method comprises the steps of acquiring current flow data at nodes of each link of the power wireless heterogeneous network;
a first prediction unit: the method comprises the steps of generating an input matrix according to current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
a second prediction unit: the method comprises the steps of analyzing the change trend of current flow data to obtain a second flow prediction result of a link;
analysis unit: the method comprises the steps of analyzing a first flow prediction result and a second flow prediction result to obtain a final prediction result;
and an equalization unit: and if the final prediction result exceeds the preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network.
Optionally, the first prediction unit is configured to:
an input matrix is formed by the current data flow, and the input matrix is input into a graph neural network corresponding to the flow type according to the identified flow type;
and outputting a first flow prediction result of the link through the graph neural network.
Optionally, the second prediction unit is configured to:
removing data of the link at a peak-valley value of the selected time period;
calculating the average flow rate change rate of a plurality of adjacent sampling time points;
and calculating a flow value at the next moment according to the maximum flow change rate to serve as a second flow prediction result.
The technical scheme provided by the application has the beneficial effects that:
the flow prediction is carried out by two methods of graph neural network and change trend analysis respectively, and the two flow prediction results are combined, and meanwhile, the graph neural network is utilized to carry out the flow prediction, so that the method is more suitable for the power wireless heterogeneous network with complex network topology than the traditional neural network model, and the accuracy of the flow prediction is improved.
And carrying out flow balancing processing on the links which are predicted to exceed the threshold value, adjusting the flow according to the load condition of the available links, and fully utilizing network resources while eliminating hidden danger of the network links in advance so as to avoid the risk of high load caused by the concentration of the flow on a part of links.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a method for managing a wireless heterogeneous network of electric power based on flow balance according to the present application;
fig. 2 is a block diagram of a power wireless heterogeneous network management device based on flow balance according to the present application.
Detailed Description
In order to make the structure and advantages of the present application more apparent, the structure of the present application will be further described with reference to the accompanying drawings.
Example 1
As shown in fig. 1, the present application provides a method for managing a wireless heterogeneous network of electric power based on flow balance, comprising:
s1: collecting current flow data at nodes of each link of the power wireless heterogeneous network;
s2: generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
s3: analyzing the change trend of the current flow data to obtain a second flow prediction result of the link;
s4: analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result;
s5: and if the final prediction result exceeds a preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network.
In this embodiment, current traffic data at the node is collected by the wireless terminal and the wireless base station network manager, where the current traffic data includes link total traffic, link node traffic, port traffic, link load rate, and source IP traffic.
In this embodiment, the generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting, by the graph neural network, a first flow prediction result of a link includes:
the traffic type at each node on the link is identified based on a fuzzy C-means algorithm. The current flow data acquired at the ith sampling moment in S1 form a data set X i ={x 1 ,x 2 ,…,x n },x 1 、x 2 、…、x n Respectively, total traffic between link nodes, in-out link node traffic, in-out port traffic, link load rate and source IP traffic, the data are used as the characteristics of the flow to form a characteristic matrix, and the flow on the link belongs to a clustering loss function J of various flow types based on a fuzzy C-means algorithm m Minimizing the cluster loss function J m The following are provided:
wherein X is i C, obtaining current flow data at the ith sampling moment j For the clustering center of the preset flow type, m is artificially setConstant for controlling fuzzy degree of clustering result, wherein the value range is 1-m<∞,u ij The current flow data obtained at the ith sampling moment belongs to the membership degree of the jth class of flow, n is the total number of data sets, and c is the total number of flow types; the values of n and c are positive integers, X i 、c j The value ranges of the (a) are positive numbers. Those skilled in the art will know how to calculate the membership by the fuzzy C-means algorithm, and will not be described in detail here. And repeating the calculation process until the membership degree is stable within a preset range, and dividing the current flow data obtained at the ith sampling moment into the flow types with the highest membership degree.
And forming an input matrix by the current data traffic, and inputting the input matrix into the graph neural network of the corresponding traffic type according to the identified traffic type. Because different types of flow generally have different change rules, in consideration of the influence of the flow types on the prediction result, in the embodiment, different graph neural networks are trained in advance according to the different flow types, so that the prediction accuracy is further improved.
And finally, outputting a first flow prediction result of the link through the graph neural network. The traditional neural network faces a great challenge when processing the electric power wireless heterogeneous network with a complex topological structure, and because the adjacent nodes of each node are unequal in number and have association relations between the adjacent nodes, the relation of each node in the electric power wireless heterogeneous network can be described through the vertexes and the edges in the graph by adopting the graph neural network.
The pre-trained graph neural network comprises vertexes, edges connected with the vertexes and relationship features of the edges, in the embodiment, the vertexes are all nodes in a link, flow data at the nodes are the features of the vertexes, the edges are topological relations between adjacent nodes, and flow mapping relations between the two nodes are the relationship features of the edges. And sequentially passing the input matrix through an input layer, a hidden layer and an output layer of the graph neural network to obtain a first flow prediction result of the link, wherein the first flow prediction result is the flow of the link at the next moment.
In this embodiment, the analyzing the change trend of the current flow data to obtain the second flow prediction result of the link includes:
the data of the link at the peak-to-valley value of the selected time period is culled. This condition is random and not sustainable due to certain incidents resulting in a surge in network traffic. Because such an emergency would affect the accuracy of the flow trend prediction algorithm, it would need to be eliminated.
And calculating the average flow rate change rate of a plurality of adjacent sampling time points. The average change rate of five adjacent sampling time points of the link is calculated in the embodimentThe method comprises the following steps:
s represents the total flow of the link, t is the moment to be predicted, and Deltat is the sampling period.
Calculating the flow value at the next moment according to the maximum flow change rate as a second flow prediction result, namely a second flow prediction resultThe method comprises the following steps:
wherein S is t-Δt Indicating the total traffic of the link at time t- Δt.
In this embodiment, the analyzing the first flow prediction result and the second flow prediction result to obtain the final prediction result includes:
calculating a difference between the first flow prediction result and the second flow prediction result;
if the difference is smaller than the preset threshold, the result obtained by the two prediction methods is not quite different, so that a first flow result with higher probability of approaching the actual flow is selected as a final prediction result.
If the difference value is not smaller than the preset threshold value, the result obtained by the two prediction methods is larger, in order to avoid the deviation influence of a certain prediction result, the first flow prediction result and the second flow prediction result are weighted based on the preset weight, and the weighted calculation result is used as a final prediction result, so that the prediction result is corrected, and the prediction accuracy is improved.
In this embodiment, the method for managing an electric power wireless heterogeneous network further includes, after collecting current traffic data at nodes of each link of the electric power wireless heterogeneous network, establishing a link load mapping table;
the link load mapping table is shown in table 1, and includes a link number, a link load value and a preset traffic threshold.
TABLE 1
Link numbering | Link load value | Preset flow threshold |
S 12 | 40% | 75% |
S 13 | 60% | 75% |
S 14 | 20% | 75% |
S 23 | 20% | 75% |
… | … | … |
In this embodiment, the link load value in the link load mapping table is updated in real time along with the current flow data collected in S1. The preset traffic threshold refers to a bandwidth utilization threshold.
In this embodiment, if the final prediction result exceeds the preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network includes:
and judging whether a final prediction result exceeds a preset flow threshold of the link by inquiring a link load mapping table, and if so, searching whether an available link exists according to a routing path. For example, S 13 The final prediction result of (a) is 80%, S 13 The traffic to bandwidth ratio at the next moment is 80%, and it can be seen from table 1 that the preset traffic threshold is exceeded by 75%, so that whether there is a usable link is searched according to a routing path, that is, a link path obtained through the topology structure of the power wireless heterogeneous network. In this embodiment, a route with a link load value not exceeding 50% is selected as an available link, and S is as shown in Table 1 12 、S 14 、S 23 No more than 50% is possible, and thus the route from node 1 to node 2 to node 3 to node 4 is taken as the available link.
If the available link is found, the load capacity of the available link is obtained, and if the load capacity does not exceed the preset proportion of the route, echo information is sent to inhibit the source IP flow. In a communication network, echoes can attenuate the delayed original signal, so that the bandwidth remaining in the available link cannot afford S 13 To suppress S by echo information at the full flow rate of (2) 13 Source IP traffic of (a) to avoid link overloadAnd (5) carrying.
And if the load capacity exceeds the preset proportion of the route and more than one available link is found, distributing the flow of the power wireless heterogeneous network according to the load capacity of the available link in proportion.
Example two
As shown in fig. 2, the present application proposes a power wireless heterogeneous network management device 6 based on flow balancing, including:
acquisition unit 61: the method comprises the steps of acquiring current flow data at nodes of each link of the power wireless heterogeneous network;
the first prediction unit 62: the method comprises the steps of generating an input matrix according to current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
the second prediction unit 63: the method comprises the steps of analyzing the change trend of current flow data to obtain a second flow prediction result of a link;
analysis unit 64: the method comprises the steps of analyzing a first flow prediction result and a second flow prediction result to obtain a final prediction result;
equalization unit 65: and if the final prediction result exceeds the preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network.
In this embodiment, current traffic data at the node is collected by the wireless terminal and the wireless base station network manager, where the current traffic data includes link total traffic, link node traffic, port traffic, link load rate, and source IP traffic.
In this embodiment, the first prediction unit 62 is specifically configured to:
the traffic type at each node on the link is identified based on a fuzzy C-means algorithm. The current flow data acquired at the ith sampling moment in S1 form a data set X i ={x 1 ,x 2 ,…,x n },x 1 、x 2 、…、x n The data are respectively the total traffic between the link nodes, the traffic of the port, the link load rate and the source IP trafficAs the characteristic composition characteristic matrix of the flow, based on the fuzzy C-means algorithm, the flow on the link belongs to the clustering loss function J of various flow types m Minimizing the cluster loss function J m The following are provided:
wherein X is i C, obtaining current flow data at the ith sampling moment j For the clustering center of the preset flow type, m is a artificially set constant for controlling the fuzzy degree of the clustering result, and the value range is 1-m<∞,u ij The current flow data obtained at the ith sampling moment belongs to the membership degree of the jth class of flow, n is the total number of data sets, and c is the total number of flow types; the values of n and c are positive integers, X i 、c j The value ranges of the (a) are positive numbers. Those skilled in the art will know how to calculate the membership by the fuzzy C-means algorithm, and will not be described in detail here. And repeating the calculation process until the membership degree is stable within a preset range, and dividing the current flow data obtained at the ith sampling moment into the flow types with the highest membership degree.
And forming an input matrix by the current data traffic, and inputting the input matrix into the graph neural network of the corresponding traffic type according to the identified traffic type. Because different types of flow generally have different change rules, in consideration of the influence of the flow types on the prediction result, in the embodiment, different graph neural networks are trained in advance according to the different flow types, so that the prediction accuracy is further improved.
And finally, outputting a first flow prediction result of the link through the graph neural network. The traditional neural network faces a great challenge when processing the electric power wireless heterogeneous network with a complex topological structure, and because the adjacent nodes of each node are unequal in number and have association relations between the adjacent nodes, the relation of each node in the electric power wireless heterogeneous network can be described through the vertexes and the edges in the graph by adopting the graph neural network.
The pre-trained graph neural network comprises vertexes, edges connected with the vertexes and relationship features of the edges, in the embodiment, the vertexes are all nodes in a link, flow data at the nodes are the features of the vertexes, the edges are topological relations between adjacent nodes, and flow mapping relations between the two nodes are the relationship features of the edges. And sequentially passing the input matrix through an input layer, a hidden layer and an output layer of the graph neural network to obtain a first flow prediction result of the link, wherein the first flow prediction result is the flow of the link at the next moment.
In this embodiment, the second prediction unit 63 is specifically configured to:
the data of the link at the peak-to-valley value of the selected time period is culled. This condition is random and not sustainable due to certain incidents resulting in a surge in network traffic. Because such an emergency would affect the accuracy of the flow trend prediction algorithm, it would need to be eliminated.
And calculating the average flow rate change rate of a plurality of adjacent sampling time points. The average change rate of five adjacent sampling time points of the link is calculated in the embodimentThe method comprises the following steps:
s represents the total flow of the link, t is the moment to be predicted, and Deltat is the sampling period.
Calculating the flow value at the next moment according to the maximum flow change rate as a second flow prediction result, namely a second flow prediction resultThe method comprises the following steps:
wherein S is t-Δt Representation chainTotal flow of the path at time t- Δt.
In the present embodiment, the analysis unit 64 is configured to:
calculating a difference between the first flow prediction result and the second flow prediction result;
if the difference is smaller than the preset threshold, the result obtained by the two prediction methods is not quite different, so that a first flow result with higher probability of approaching the actual flow is selected as a final prediction result.
If the difference value is not smaller than the preset threshold value, the result obtained by the two prediction methods is larger, in order to avoid the deviation influence of a certain prediction result, the first flow prediction result and the second flow prediction result are weighted based on the preset weight, and the weighted calculation result is used as a final prediction result, so that the prediction result is corrected, and the prediction accuracy is improved.
In this embodiment, the power wireless heterogeneous network management device 6 further includes a mapping unit, configured to establish a link load mapping table after collecting current traffic data at nodes of each link of the power wireless heterogeneous network;
the link load mapping table is shown in table 1, and includes a link number, a link load value and a preset traffic threshold.
TABLE 1
Link numbering | Link load value | Preset flow threshold |
S 12 | 40% | 75% |
S 13 | 60% | 75% |
S 14 | 20% | 75% |
S 23 | 20% | 75% |
… | … | … |
In this embodiment, the link load value in the link load mapping table is updated in real time along with the current flow data collected in the collection unit 61. The preset traffic threshold refers to a bandwidth utilization threshold.
In this embodiment, the equalizing unit 65 is configured to:
and judging whether a final prediction result exceeds a preset flow threshold of the link by inquiring a link load mapping table, and if so, searching whether an available link exists according to a routing path. For example, S 13 The final prediction result of (a) is 80%, S 13 The traffic to bandwidth ratio at the next moment is 80%, and it can be seen from table 1 that the preset traffic threshold is exceeded by 75%, so that whether there is a usable link is searched according to a routing path, that is, a link path obtained through the topology structure of the power wireless heterogeneous network. In this embodiment, a route with a link load value not exceeding 50% is selected as an available link, and S is as shown in Table 1 12 、S 14 、S 23 No more than 50% is possible, and thus the route from node 1 to node 2 to node 3 to node 4 is taken as the available link.
If the available link is found, the load capacity of the available link is obtained, and if the load capacity does not exceed the preset proportion of the route, echo information is sent to inhibit the source IP flow. In a communication network, echoes can attenuate the delayed original signal, so that the bandwidth remaining in the available link cannot afford S 13 To suppress S by echo information at the full flow rate of (2) 13 Is used to avoid link overload.
And if the load capacity exceeds the preset proportion of the route and more than one available link is found, distributing the flow of the power wireless heterogeneous network according to the load capacity of the available link in proportion.
The various numbers in the above embodiments are for illustration only and do not represent the order of assembly or use of the various components.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present application and not for limiting the same, and although the present application has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the application without departing from the spirit and scope of the application, which is intended to be covered by the claims.
Claims (5)
1. The power wireless heterogeneous network management method based on flow balance is characterized by comprising the following steps of:
collecting current flow data at nodes of each link of the power wireless heterogeneous network;
generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
analyzing the change trend of the current flow data to obtain a second flow prediction result of the link;
analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result;
if the final prediction result exceeds a preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network;
analyzing the first flow prediction result and the second flow prediction result to obtain a final prediction result, including:
calculating a difference between the first flow prediction result and the second flow prediction result;
if the difference value is smaller than a preset threshold value, the first flow prediction result is used as a final prediction result;
if the difference value is not smaller than the preset threshold value, carrying out weighted calculation on the first flow prediction result and the second flow prediction result based on preset weight, and taking the weighted calculation result as a final prediction result;
generating an input matrix according to the current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network, wherein the method comprises the following steps:
identifying the flow type at each node on the link based on a fuzzy C-means algorithm;
an input matrix is formed by the current data flow, and the input matrix is input into a graph neural network corresponding to the flow type according to the identified flow type;
outputting a first flow prediction result of the link through the graph neural network;
analyzing the change trend of the current flow data to obtain a second flow prediction result of the link, including:
removing data of the link at a peak-valley value of the selected time period;
calculating the average flow rate change rate of a plurality of adjacent sampling time points;
and calculating a flow value at the next moment according to the maximum flow change rate to serve as a second flow prediction result.
2. The traffic balancing-based power wireless heterogeneous network management method according to claim 1, wherein the current traffic data comprises link total traffic, ingress and egress link node traffic, ingress and egress port traffic, link load rate, and source IP traffic.
3. The traffic balance-based power wireless heterogeneous network management method according to claim 1, further comprising, after collecting current traffic data at nodes of each link of the power wireless heterogeneous network, establishing a link load map;
the link load mapping table comprises a link number, a link load value and a preset flow threshold.
4. The method for managing a wireless heterogeneous network of power based on flow balancing according to claim 1, wherein if the final prediction result exceeds a preset flow threshold, performing flow balancing processing on the wireless heterogeneous network of power, comprises:
judging whether a final prediction result exceeds a preset flow threshold of the link or not by inquiring a link load mapping table;
if the flow threshold exceeds the preset flow threshold, searching whether an available link exists according to the routing path;
if the available link is found, the load capacity of the available link is obtained, and if the load capacity does not exceed the preset proportion of the route, echo information is sent to inhibit the source IP flow;
and if the load capacity exceeds the preset proportion of the route and more than one available link is found, distributing the flow of the power wireless heterogeneous network according to the load capacity of the available link in proportion.
5. The utility model provides a wireless heterogeneous network management device of electric power based on flow balance which characterized in that, wireless heterogeneous network management device of electric power includes:
the acquisition unit: the method comprises the steps of acquiring current flow data at nodes of each link of the power wireless heterogeneous network;
a first prediction unit: the method comprises the steps of generating an input matrix according to current flow data, inputting the input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
a second prediction unit: the method comprises the steps of analyzing the change trend of current flow data to obtain a second flow prediction result of a link;
analysis unit: the method comprises the steps of analyzing a first flow prediction result and a second flow prediction result to obtain a final prediction result;
and an equalization unit: if the final prediction result exceeds a preset flow threshold, performing flow equalization processing on the power wireless heterogeneous network;
the analysis unit is used for:
calculating a difference between the first flow prediction result and the second flow prediction result;
if the difference value is smaller than a preset threshold value, selecting a first flow result as a final prediction result;
if the difference value is not smaller than a preset threshold value, carrying out weighted calculation on the first flow prediction result and the second flow prediction result based on a preset weight, and taking the weighted calculation result as a final prediction result;
the first prediction unit is used for:
identifying the flow type at each node on the link based on a fuzzy C-means algorithm;
weighting current flow data at the node according to a preset flow type weight, and forming an input matrix from the weighted current flow data according to the sequence of acquisition time;
inputting an input matrix into a pre-trained graph neural network, and outputting a first flow prediction result of a link through the graph neural network;
the second prediction unit is used for:
removing data of the link at a peak-valley value of the selected time period;
calculating the average flow rate change rate of a plurality of adjacent sampling time points;
and calculating a flow value at the next moment according to the maximum flow change rate to serve as a second flow prediction result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011527098.8A CN112766633B (en) | 2020-12-22 | 2020-12-22 | Flow balance-based power wireless heterogeneous network management method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011527098.8A CN112766633B (en) | 2020-12-22 | 2020-12-22 | Flow balance-based power wireless heterogeneous network management method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112766633A CN112766633A (en) | 2021-05-07 |
CN112766633B true CN112766633B (en) | 2023-10-24 |
Family
ID=75694787
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011527098.8A Active CN112766633B (en) | 2020-12-22 | 2020-12-22 | Flow balance-based power wireless heterogeneous network management method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112766633B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09179902A (en) * | 1995-12-25 | 1997-07-11 | Hitachi Ltd | Highly reliable physical distribution managing method and its system |
CN104410582A (en) * | 2014-12-10 | 2015-03-11 | 国家电网公司 | Traffic balancing method for electric power communication network based on traffic prediction |
CN110084359A (en) * | 2019-03-20 | 2019-08-02 | 山东师范大学 | Dynamic intuitionistic fuzzy Cognitive Map construction method, Time Series Forecasting Methods and system |
CN110995520A (en) * | 2020-02-28 | 2020-04-10 | 清华大学 | Network flow prediction method and device, computer equipment and readable storage medium |
CN111092862A (en) * | 2019-11-29 | 2020-05-01 | 中国电力科学研究院有限公司 | Method and system for detecting abnormal communication flow of power grid terminal |
-
2020
- 2020-12-22 CN CN202011527098.8A patent/CN112766633B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09179902A (en) * | 1995-12-25 | 1997-07-11 | Hitachi Ltd | Highly reliable physical distribution managing method and its system |
CN104410582A (en) * | 2014-12-10 | 2015-03-11 | 国家电网公司 | Traffic balancing method for electric power communication network based on traffic prediction |
CN110084359A (en) * | 2019-03-20 | 2019-08-02 | 山东师范大学 | Dynamic intuitionistic fuzzy Cognitive Map construction method, Time Series Forecasting Methods and system |
CN111092862A (en) * | 2019-11-29 | 2020-05-01 | 中国电力科学研究院有限公司 | Method and system for detecting abnormal communication flow of power grid terminal |
CN110995520A (en) * | 2020-02-28 | 2020-04-10 | 清华大学 | Network flow prediction method and device, computer equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN112766633A (en) | 2021-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106850254B (en) | Method for identifying key nodes in power communication network | |
CN105933920B (en) | A kind of method and device for predicting user satisfaction | |
CN111294812A (en) | Method and system for resource capacity expansion planning | |
CN113691594B (en) | Method for solving data imbalance problem in federal learning based on second derivative | |
CN112187891A (en) | Load optimization method and device of edge computing node set based on multiple services | |
CN114513470B (en) | Network flow control method, device, equipment and computer readable storage medium | |
CN114615294A (en) | Electric power internet of things gateway edge calculation method | |
CN111901134A (en) | Method and device for predicting network quality based on recurrent neural network model (RNN) | |
CN112804702B (en) | Multi-link air-ground data exchange link performance evaluation method based on utility function | |
CN112766633B (en) | Flow balance-based power wireless heterogeneous network management method and device | |
CN113298296A (en) | Method for predicting day-ahead load probability of power transmission substation from bottom to top | |
CN116166444B (en) | Collaborative reasoning method oriented to deep learning hierarchical model | |
CN106789163A (en) | A kind of network equipment power information monitoring method, device and system | |
CN114462506B (en) | Communication network auxiliary planning method supporting preference strategy and application thereof | |
CN103096380B (en) | Wireless access point load balancing load balancing | |
CN113727385A (en) | Wireless network quality comprehensive evaluation method based on big data and expert system | |
CN114401192A (en) | Multi-SDN controller collaborative training method | |
CN114035906A (en) | Virtual machine migration method and device, electronic equipment and storage medium | |
Krishnamurthy et al. | Segregation in social networks: Markov bridge models and estimation | |
CN113434270A (en) | Data resource scheduling method and device, electronic equipment and storage medium | |
Su et al. | Research and implementation of intelligent distribution network efficiency evaluation system | |
CN112996015A (en) | Index association relationship construction method and device | |
CN111444454A (en) | Dynamic community dividing method based on spectrum method | |
CN115833116B (en) | Power distribution network reconstruction optimization method based on multi-objective optimization | |
CN114528946B (en) | Autonomous domain system sibling relationship identification method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |