CN114422438A - Link adjusting method and device of power communication network - Google Patents
Link adjusting method and device of power communication network Download PDFInfo
- Publication number
- CN114422438A CN114422438A CN202111647418.8A CN202111647418A CN114422438A CN 114422438 A CN114422438 A CN 114422438A CN 202111647418 A CN202111647418 A CN 202111647418A CN 114422438 A CN114422438 A CN 114422438A
- Authority
- CN
- China
- Prior art keywords
- link
- communication network
- power communication
- switch
- information
- 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.)
- Granted
Links
- 238000004891 communication Methods 0.000 title claims abstract description 100
- 238000000034 method Methods 0.000 title claims abstract description 36
- 239000011159 matrix material Substances 0.000 claims abstract description 112
- 230000005540 biological transmission Effects 0.000 claims abstract description 43
- 238000003062 neural network model Methods 0.000 claims abstract description 40
- 238000004458 analytical method Methods 0.000 claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims description 39
- 238000013528 artificial neural network Methods 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a link adjusting method and device of a power communication network, wherein a flow characteristic matrix is established according to switch information and link information of a connection switch and a link adjacent matrix is established according to the switch information and the link information so as to represent the flow states of a plurality of switches in the power communication network and the connection relation among a plurality of links; training a preset graph neural network model based on the flow characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model; and finally, predicting a first link flow parameter of the power communication network in a future target time period by using a link analysis model, wherein the first link flow parameter is used for adjusting a transmission link of the service data, so that the transmission link of the service data is effectively controlled.
Description
Technical Field
The present application relates to the field of power communication network control technologies, and in particular, to a link adjustment method and device for a power communication network.
Background
The power communication network is used as an indispensable support platform for the safe operation of the smart power grid, is a basis for realizing the automation of power dispatching, the marketization of power grid operation and the informatization of power grid management, and is an important means for ensuring the safe, stable and efficient economic operation of the power grid. With the continuous expansion of the scale of the power communication network, the types and the traffic of the power services are rapidly increased, and the conditions of uneven network flow distribution and the like caused by sudden and uncertain services cause great challenges to the guarantee of high-quality transmission services of the power services.
Under the conditions of sudden increase of service volume, sudden increase of service and increased service transmission requirements in the power communication network, the current power communication network only depends on a static routing algorithm, so that the reliable and safe transmission of the power communication service is difficult to guarantee, and the effective pre-control of the power service flow cannot be realized.
Disclosure of Invention
The application provides a link adjusting method and device of a power communication network, which aim to solve the technical problem that the current power communication network cannot effectively pre-control power service flow.
In order to solve the technical problem, in a first aspect, an embodiment of the present application provides a link adjustment method for a power communication network, including:
acquiring switch information of a power communication network and link information connected with a switch, wherein the switch information comprises switch flow parameters;
establishing a flow characteristic matrix according to the switch information and the link information, wherein the flow characteristic matrix is used for representing the flow states of a plurality of switches in the power communication network;
according to the link information, a link adjacency matrix is established, and the link adjacency matrix is used for representing the connection relation among a plurality of links in the power communication network;
training a preset graph neural network model based on the traffic characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model;
and predicting a first link flow parameter of the power communication network in a future target time period by using the link analysis model, wherein the first link flow parameter is used for adjusting a transmission link of the service data.
In the embodiment, the switch information of the power communication network and the link information of the connected switches are obtained, a flow characteristic matrix is established according to the switch information and the link information, and a link adjacent matrix is established according to the link information so as to represent the flow states of a plurality of switches in the power communication network and the connection relation among a plurality of links; training a preset graph neural network model based on the flow characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model, so that the flow characteristic of the power communication network can be learned by the graph neural network by utilizing the flow state of the switch and the connection relation of the links to serve as a characteristic basis for subsequently predicting the link flow parameter of the power communication network in a future time period; and finally, predicting a first link flow parameter of the power communication network in a future target time period by using a link analysis model, wherein the first link flow parameter can be used for adjusting a transmission link of the service data, so that the transmission link of the service data is effectively controlled according to the link flow parameter in the future time period, the service response speed is improved, the communication delay is reduced, the communication quality is improved, and the transmission reliability of the service data is further improved.
In one embodiment, establishing a traffic characteristic matrix according to the switch information and the link information includes:
performing data processing on the switch information and the link information, determining a feature vector between every two pieces of switch information, and establishing a plurality of feature vectors as a flow feature matrix;
the feature vectors are: e (j) ═ Pa，q(t)，Pb，g(t)，Fa(t)，Fb(t)]；
Wherein e (j) represents the characteristic vector of the j link, j is the link number, the j link is connected with the q port of the switch a and the g port of the switch b, Pa，q(t) a port state parameter, P, for the q-port of switch a at time tb，g(t) a port status parameter for g port of switch b at time t, Fa(t) flow meter state parameter, F, for switch a at time tbAnd (t) represents the flow table state parameters of the switch b at the time t, wherein the switch flow parameters comprise port state parameters and flow table state parameters.
In one embodiment, establishing the link adjacency matrix according to the link information includes:
processing the link information to determine the connection relation between every two links, and establishing a plurality of connection relations as an adjacency matrix A belonging to Rj×j，Rj×jRepresenting the connection relation between j links;
and carrying out standardization processing on the adjacency matrix A to obtain a link adjacency matrix A'.
In an embodiment, training a preset graph neural network model based on a traffic characteristic matrix and a link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model, includes:
carrying out convolution processing, correction processing and graph-level operation on the flow characteristic matrix and the link adjacent matrix by using a graph neural network model to obtain a second link flow parameter of the power communication network;
calculating a loss value between the second link flow parameter and a preset marking parameter;
and if the loss value is not less than the preset threshold value, updating the model parameters of the graph neural network model, and predicting a new second link flow parameter by using the updated graph neural network model until the loss value is less than the preset threshold value to obtain a link analysis model.
In an embodiment, performing convolution processing, correction processing, and graph-level operation on the flow characteristic matrix and the link adjacency matrix by using a graph neural network model to obtain a second link flow parameter of the power communication network includes:
performing convolution processing on the flow characteristic matrix and the link adjacent matrix by using a preset ReLU function to obtain a first neural network layer;
correcting the first neural network layer by using a preset softmax function to obtain a second neural network layer;
and carrying out graph-level operation on the second neural network layer by using the preset link flow occupation level to obtain a second link flow parameter.
In an embodiment, the model parameters include a first weight matrix of the first neural network layer and a second weight matrix of the second neural network layer.
In an embodiment, after predicting the first link traffic parameter of the power communication network within the future target time period using the link analysis model, the method further includes:
acquiring service demand information of a power communication network, wherein the service demand information comprises sending switch information used for sending service data and receiving switch information used for receiving the service data;
determining flow occupancy data of a plurality of transmission paths according to the first link flow parameter, the sending switch information and the receiving switch information, wherein the transmission paths are composed of a plurality of links from the sending switch to the receiving switch;
and comparing the plurality of transmission paths according to the traffic occupancy data, and determining a target transmission path corresponding to the service demand information.
In a second aspect, an embodiment of the present application provides a link analysis device for a power communication network, including:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring switch information of a power communication network and link information connected with a switch, and the switch information comprises switch flow parameters;
the first establishing module is used for establishing a flow characteristic matrix according to the switch information and the link information, wherein the flow characteristic matrix is used for representing the flow states of a plurality of switches in the power communication network;
the second establishing module is used for establishing a link adjacency matrix according to the link information, and the link adjacency matrix is used for representing the connection relation among a plurality of links in the power communication network;
the training module is used for training a preset graph neural network model based on the traffic characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition, so as to obtain a link analysis model;
the prediction module is used for predicting a first link flow parameter of the power communication network in a future target time period by using the link analysis model, and the first link flow parameter can be used for adjusting a transmission link of the service data.
In a third aspect, an embodiment of the present application provides a computer device, including a processor and a memory, where the memory is used to store a computer program, and the computer program, when executed by the processor, implements the steps of the link adjustment method for the power communication network according to the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the link adjustment method for the power communication network according to the first aspect.
Please refer to the relevant description of the first aspect for the beneficial effects of the second to fourth aspects, which are not repeated herein.
Drawings
Fig. 1 is a schematic flowchart of a link adjustment method for a power communication network according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a link adjustment method of a power communication network according to another embodiment of the present application;
fig. 3 is a schematic structural diagram of a link analysis device of a power communication network according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
As described in the related art, under the conditions of rapid increase of traffic volume, sudden increase of traffic and increased traffic transmission demand in the power communication network, the current power communication network only depends on a static routing algorithm, which is difficult to ensure reliable and safe transmission of the power communication service, and cannot realize effective pre-control of the traffic of the power service.
The embodiment of the application provides a link adjusting method and device for an electric power communication network, wherein a flow characteristic matrix is established according to switch information and link information of a connecting switch of the electric power communication network and a link adjacent matrix is established according to the link information to represent flow states of a plurality of switches and connection relations among a plurality of links in the electric power communication network; training a preset graph neural network model based on the flow characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model, so that the flow characteristic of the power communication network can be learned by the graph neural network by utilizing the flow state of the switch and the connection relation of the links to serve as a characteristic basis for subsequently predicting the link flow parameter of the power communication network in a future time period; and finally, predicting a first link flow parameter of the power communication network in a future target time period by using a link analysis model, wherein the first link flow parameter can be used for adjusting a transmission link of the service data, so that the transmission link of the service data is effectively controlled according to the link flow parameter in the future time period, the service response speed is improved, the communication delay is reduced, the communication quality is improved, and the transmission reliability of the service data is further improved.
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a link adjusting method of an electrical power communication network according to an embodiment of the present disclosure. The link adjusting method of the power communication network in the embodiment of the application can be applied to computer equipment, and the computer equipment includes but is not limited to computing equipment such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a physical server and an independent server. As shown in fig. 1, the link adjusting method of the power communication network of the present embodiment includes steps S101 to S105, which are detailed as follows:
step S101, switch information of the power communication network and link information of a connected switch are obtained, wherein the switch information comprises switch flow parameters.
In this step, the power communication network may be regarded as being composed of a plurality of switches and links connecting the switches. The switch information includes, but is not limited to, a switch number, a port number, and switch traffic parameters, the switch traffic parameters include a total switch traffic value and port status parameters, and the port status parameters include, but are not limited to, a packet transceiving time, a number of packets to be transceived, a packet loss rate during transceiving, a number of bytes of the packets to be transceived, and a flow table status parameter. The link information includes, but is not limited to, the switch number and port number of the switch to which the link is connected.
Optionally, the control device of the power communication network (i.e., the computer device) sets the monitoring unit and the control unit on the switch and the link, monitors the switch information and the link information through the monitoring unit, and performs macro-control on the power communication network by using an analysis result obtained by analyzing the switch information and the link information.
Step S102, a flow characteristic matrix is established according to the switch information and the link information, and the flow characteristic matrix is used for representing the flow states of a plurality of switches in the power communication network.
In the present step, the first step is carried out,processing the switch information and the link information, determining a characteristic vector E (j) between every two pieces of switch information, and establishing a plurality of characteristic vectors as a flow characteristic matrix E (E (1), E (2);. E (j))]T。
Optionally, the data processing includes, but is not limited to, data cleansing, data integration, data encoding, and the like.
Optionally, the feature vector is: e (j) ═ Pa，q(t)，Pb，g(t)，Fa(t)，Fb(t)]；
Wherein e (j) represents the characteristic vector of the j link, j is the link number, the j link is connected with the q port of the switch a and the g port of the switch b, Pa，q(t) a port state parameter, P, for the q-port of switch a at time tb，g(t) a port status parameter for g port of switch b at time t, Fa(t) flow meter state parameter, F, for switch a at time tb(t) represents the flow meter state parameters of switch b at time t, and the switch flow parameters comprise the port state parameters and the flow meter state parameters.
Step S103, according to the link information, a link adjacency matrix is established, and the link adjacency matrix is used for representing the connection relation among a plurality of links in the power communication network.
In the step, the link information is subjected to data processing, the connection relation between every two links is determined, and a plurality of connection relations are established as an adjacency matrix A belonging to Rj×j，Rj×jRepresenting the connection relation between j links; and carrying out standardization processing on the adjacency matrix A to obtain the link adjacency matrix A'.
Optionally, two pieces of link information with the same switch number are determined according to the switch number in the link information, two links corresponding to the two pieces of link information are connected, and a switch corresponding to the same switch number is a link node.
And step S104, training a preset graph neural network model based on the traffic characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition, and obtaining a link analysis model.
In this step, the traffic characteristic matrix is used as a vertex set, the link adjacency matrix is used as an edge set, and the vertex set and the edge set are subjected to data processing by using the graph neural network model, so that iterative training is performed on the graph neural network model. The preset convergence condition may be that the number of iterations of the graph neural network model reaches a preset number, or that a loss function value of the graph neural network model is smaller than a preset threshold.
Step S105, predicting a first link flow parameter of the power communication network in a future target time period by using the link analysis model, wherein the first link flow parameter can be used for adjusting a transmission link of service data.
In this step, the switch information and the link information of the power communication network at the current moment are acquired, the switch information and the link information at the current moment are input into the link analysis model, and the first link flow parameter is output. Optionally, the first link traffic parameter is a link traffic occupancy probability of the power communication network at a time T in the future, and the link traffic occupancy probability is a probability that the link is transmitting data.
In an embodiment, based on the embodiment shown in fig. 1, the step S104 includes:
carrying out convolution processing, correction processing and graph-level operation on the flow characteristic matrix and the link adjacent matrix by using the graph neural network model to obtain a second link flow parameter of the power communication network;
calculating a loss value between the second link flow parameter and a preset marking parameter;
and if the loss value is not less than a preset threshold value, updating the model parameters of the graph neural network model, and predicting a new second link flow parameter by using the updated graph neural network model until the loss value is less than the preset threshold value to obtain the link analysis model.
In this embodiment, a preset ReLU function is used to perform convolution processing on the traffic characteristic matrix and the link adjacent matrix to obtain a first neural network layer; correcting the first neural network layer by using a preset softmax function to obtain a second neural network layer; and carrying out graph-level operation on the second neural network layer by utilizing a preset link flow occupation level to obtain the second link flow parameter.
Optionally, introducing the traffic characteristic matrix E and the link adjacency matrix a' into a preset ReLU function for convolution processing, to obtain:
H1＝ReLU[A′(ξE)w1]；
wherein H1Representing the first neural network layer, ξ the Bernoulli distribution coefficient, w1Is a first weight matrix;
h is to be1And importing a softmax function to perform correction processing to obtain:
H2＝softmax(A′H1w2)；
wherein H2Representing a second neural network layer, w2Is a second weight matrix;
carrying out graph-level operation on the second neural network layer, wherein the graph-level operation is to lead the preset link occupancy rate gamma of the future T time into the second neural network layer H by setting the occupancy rate gamma of the future T time2In the step (2), a second link traffic parameter is obtained, where the second link traffic parameter is a probability matrix Z whose occupancy of each link is 1, 2, 3, … …, and γ, respectively, at the time T in the future.
Optionally, the model parameters include a first weight matrix of the first neural network layer and a second weight matrix of the second neural network layer.
Optionally, the first weight matrix w is adjusted by a parameter adjustment algorithm1And a second weight matrix w2. It should be understood that the first weight matrix w1And a second weight matrix w2The parameter adjustment algorithm can be used for automatic adjustment, and the power communication network control unit can also be used for manual adjustment.
In an embodiment, on the basis of the embodiment shown in fig. 1, fig. 2 shows a flowchart of a link adjustment method of a power communication network according to another embodiment of the present application. It is understood that the same steps as those in fig. 1 are not described herein again. As shown in fig. 2, after step S105, the method further includes:
step S201, acquiring service requirement information of the power communication network, wherein the service requirement information comprises sending switch information used for sending service data and receiving switch information used for receiving the service data;
step S202, determining flow occupancy data of a plurality of transmission paths according to the first link flow parameter, the sending switch information and the receiving switch information, wherein the transmission paths are composed of a plurality of links from the sending switch to the receiving switch;
step S203, comparing the plurality of transmission paths according to the traffic occupancy data, and determining a target transmission path corresponding to the service demand information.
In steps S201 to S203, performing data analysis on the received service demand information to obtain sending switch information for sending service data and receiving switch information for receiving the service data; classifying the received service requirement information; and obtaining flow occupancy data of different transmission paths, namely a link flow prediction result according to the probability matrix Z corresponding to the first link flow parameter, the sending switch information and the receiving switch information.
Optionally, a traffic occupancy comprehensive value on each transmission path is obtained, where the comprehensive value is a maximum traffic occupancy of a link on each transmission path; comparing and sequencing the comprehensive value of the flow occupancy of each transmission path, and sequencing the comprehensive values from small to large; comparing with each transmission path according to the service classification, and selecting the transmission path as a service response transmission path according to the comparison sequence based on the sequence of the above steps until the flow capacity of one transmission path meets the service classification.
Alternatively, the ranking may be based on average traffic response time and traffic flow.
In order to execute the link adjusting method of the power communication network corresponding to the method embodiment, corresponding functions and technical effects are achieved. Referring to fig. 3, fig. 3 is a block diagram illustrating a structure of a link analysis apparatus of a power communication network according to an embodiment of the present application. For convenience of explanation, only a part related to the present embodiment is shown, and the link analysis device of the power communication network provided in the embodiment of the present application includes:
an obtaining module 301, configured to obtain switch information of a power communication network and link information connected to a switch, where the switch information includes a switch flow parameter;
a first establishing module 302, configured to establish a traffic feature matrix according to the switch information and the link information, where the traffic feature matrix is used to characterize traffic states of multiple switches in the power communication network;
a second establishing module 303, configured to establish a link adjacency matrix according to the link information, where the link adjacency matrix is used to characterize a connection relationship between multiple links in the power communication network;
a training module 304, configured to train a preset graph neural network model based on the traffic feature matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition, so as to obtain a link analysis model;
a prediction module 305, configured to predict, using the link analysis model, a first link traffic parameter of the power communication network in a future target time period, where the first link traffic parameter can be used to adjust a transmission link of traffic data.
In an embodiment, the first establishing module 302 includes:
the first establishing unit is used for carrying out data processing on the switch information and the link information, determining a characteristic vector between every two pieces of switch information, and establishing a plurality of characteristic vectors as the flow characteristic matrix;
the feature vector is: e (j) ═ Pa，q(t)，Pb，g(t)，Fa(t)，Fb(t)]；
Wherein e (j) represents the bit of the j linkA eigenvector, j is a link number, the j th link is connected with the q port of the switch a and the g port of the switch b, Pa，q(t) a port state parameter, P, for the q-port of switch a at time tb，g(t) a port status parameter for g port of switch b at time t, Fa(t) flow meter state parameter, F, for switch a at time tb(t) represents the flow meter state parameters of switch b at time t, and the switch flow parameters comprise the port state parameters and the flow meter state parameters.
In an embodiment, the second establishing module 303 includes:
a second establishing unit, configured to perform data processing on the link information, determine a connection relationship between every two links, and establish a plurality of connection relationships as an adjacency matrix a belonging to Rj×j，Rj×jRepresenting the connection relation between j links;
and the normalization unit is used for performing normalization processing on the adjacency matrix A to obtain the link adjacency matrix A'.
In one embodiment, the training module 304 includes:
the processing unit is used for performing convolution processing, correction processing and graph-level operation on the flow characteristic matrix and the link adjacent matrix by using the graph neural network model to obtain a second link flow parameter of the power communication network;
the calculating unit is used for calculating a loss value between the second link flow parameter and a preset marking parameter;
and the updating unit is used for updating the model parameters of the graph neural network model if the loss value is not less than a preset threshold value, predicting new second link flow parameters by using the updated graph neural network model until the loss value is less than the preset threshold value, and obtaining the link analysis model.
In one embodiment, the processing unit includes:
the convolution subunit is configured to perform convolution processing on the traffic characteristic matrix and the link adjacent matrix by using a preset ReLU function to obtain a first neural network layer;
the correction subunit is used for correcting the first neural network layer by using a preset softmax function to obtain a second neural network layer;
and the operation subunit is used for performing graph-level operation on the second neural network layer by using a preset link traffic occupation level to obtain the second link traffic parameter.
In an embodiment, the model parameters include a first weight matrix of the first neural network layer and a second weight matrix of the second neural network layer.
In an embodiment, the link analysis apparatus further includes:
the second acquisition module is used for acquiring service demand information of the power communication network, wherein the service demand information comprises sending switch information used for sending service data and receiving switch information used for receiving the service data;
a determining module, configured to determine traffic occupancy data of multiple transmission paths according to the first link traffic parameter, the sending switch information, and the receiving switch information, where the transmission paths are composed of multiple links from a sending switch to a receiving switch;
and the comparison module is used for comparing the plurality of transmission paths according to the traffic occupancy data and determining a target transmission path corresponding to the service demand information.
The link analysis device of the power communication network may implement the link adjustment method of the power communication network of the above method embodiment. The alternatives in the above-described method embodiments are also applicable to this embodiment and will not be described in detail here. The rest of the embodiments of the present application may refer to the contents of the above method embodiments, and in this embodiment, details are not described again.
Fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 4, the computer device 4 of this embodiment includes: at least one processor 40 (only one shown in fig. 4), a memory 41, and a computer program 42 stored in the memory 41 and executable on the at least one processor 40, the processor 40 implementing the steps of any of the method embodiments described above when executing the computer program 42.
The computer device 4 may be a computing device such as a smart phone, a tablet computer, a desktop computer, and a cloud server. The computer device may include, but is not limited to, a processor 40, a memory 41. Those skilled in the art will appreciate that fig. 4 is merely an example of the computer device 4 and does not constitute a limitation of the computer device 4, and may include more or less components than those shown, or combine certain components, or different components, such as input output devices, network access devices, etc.
The Processor 40 may be a Central Processing Unit (CPU), and the Processor 40 may be other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 41 may in some embodiments be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 4. The memory 41 may also be an external storage device of the computer device 4 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the computer device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the computer device 4. The memory 41 is used for storing an operating system, an application program, a BootLoader (BootLoader), data, and other programs, such as program codes of the computer program. The memory 41 may also be used to temporarily store data that has been output or is to be output.
In addition, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps in any of the method embodiments described above.
The embodiments of the present application provide a computer program product, which when running on a terminal device, enables the terminal device to implement the steps in the above method embodiments when executed.
In several embodiments provided herein, it will be understood that each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a terminal device to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are further detailed to explain the objects, technical solutions and advantages of the present application, and it should be understood that the above-mentioned embodiments are only examples of the present application and are not intended to limit the scope of the present application. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the present application, may occur to those skilled in the art and are intended to be included within the scope of the present application.
Claims (10)
1. A link regulation method of a power communication network is characterized by comprising the following steps:
acquiring switch information of a power communication network and link information connected with a switch, wherein the switch information comprises switch flow parameters;
establishing a flow characteristic matrix according to the switch information and the link information, wherein the flow characteristic matrix is used for representing the flow states of a plurality of switches in the power communication network;
establishing a link adjacency matrix according to the link information, wherein the link adjacency matrix is used for representing the connection relation among a plurality of links in the power communication network;
training a preset graph neural network model based on the traffic characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model;
and predicting a first link flow parameter of the power communication network in a future target time period by using the link analysis model, wherein the first link flow parameter is used for adjusting a transmission link of service data.
2. The link adjusting method of the power communication network according to claim 1, wherein the establishing a traffic characteristic matrix according to the switch information and the link information includes:
performing data processing on the switch information and the link information, determining a feature vector between every two pieces of switch information, and establishing a plurality of feature vectors as the traffic feature matrix;
the feature vector is: e (j) ═ Pa，q(t)，Pb，g(t)，Fa(t)，Fb(t)]；
Wherein e (j) represents the characteristic vector of the j link, j is the link number, the j link is connected with the q port of the switch a and the g port of the switch b, Pa，q(t) a port state parameter, P, for the q-port of switch a at time tb，g(t) a port status parameter for g port of switch b at time t, Fa(t) flow meter state parameter, F, for switch a at time tb(t) represents the flow meter state parameters of switch b at time t, and the switch flow parameters comprise the port state parameters and the flow meter state parameters.
3. The link adjusting method of the power communication network according to claim 1, wherein the establishing a link adjacency matrix according to the link information comprises:
processing the link information to determine the connection relation between every two links, and establishing a plurality of connection relations as an adjacency matrix A belonging to Rj×j，Rj×jRepresenting the connection relation between j links;
and carrying out standardization processing on the adjacency matrix A to obtain the link adjacency matrix A'.
4. The method for adjusting the links of the power communication network according to any one of claims 1 to 3, wherein the training a preset graph neural network model based on the traffic feature matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition to obtain a link analysis model comprises:
carrying out convolution processing, correction processing and graph-level operation on the flow characteristic matrix and the link adjacent matrix by using the graph neural network model to obtain a second link flow parameter of the power communication network;
calculating a loss value between the second link flow parameter and a preset marking parameter;
and if the loss value is not less than a preset threshold value, updating the model parameters of the graph neural network model, and predicting a new second link flow parameter by using the updated graph neural network model until the loss value is less than the preset threshold value to obtain the link analysis model.
5. The method for adjusting the links of the power communication network according to claim 4, wherein the obtaining the second link traffic parameter of the power communication network by performing convolution processing, correction processing and graph-level operation on the traffic characteristic matrix and the link adjacency matrix by using the graph neural network model comprises:
performing convolution processing on the flow characteristic matrix and the link adjacent matrix by using a preset ReLU function to obtain a first neural network layer;
correcting the first neural network layer by using a preset softmax function to obtain a second neural network layer;
and carrying out graph-level operation on the second neural network layer by utilizing a preset link flow occupation level to obtain the second link flow parameter.
6. The method of link adjustment for a power communication network of claim 5 wherein the model parameters comprise a first weight matrix for the first neural network layer and a second weight matrix for the second neural network layer.
7. A method of link regulation for a power communication network according to any one of claims 1 to 3, wherein the predicting, using the link analysis model, the first link traffic parameter for the power communication network within a future target time period further comprises:
acquiring service demand information of the power communication network, wherein the service demand information comprises sending switch information used for sending service data and receiving switch information used for receiving the service data;
determining traffic occupancy data of a plurality of transmission paths according to the first link traffic parameter, the sending switch information and the receiving switch information, wherein the transmission paths are composed of a plurality of links from the sending switch to the receiving switch;
and comparing the plurality of transmission paths according to the traffic occupancy data, and determining a target transmission path corresponding to the service demand information.
8. A link analysis device of a power communication network, comprising:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring switch information of a power communication network and link information connected with a switch, and the switch information comprises switch flow parameters;
the first establishing module is used for establishing a flow characteristic matrix according to the switch information and the link information, wherein the flow characteristic matrix is used for representing the flow states of a plurality of switches in the power communication network;
the second establishing module is used for establishing a link adjacency matrix according to the link information, and the link adjacency matrix is used for representing the connection relation among a plurality of links in the power communication network;
the training module is used for training a preset graph neural network model based on the traffic characteristic matrix and the link adjacency matrix until the graph neural network model reaches a preset convergence condition, so as to obtain a link analysis model;
and the predicting module is used for predicting a first link flow parameter of the power communication network in a future target time period by using the link analysis model, and the first link flow parameter is used for adjusting a transmission link of service data.
9. A computer device, characterized in that it comprises a processor and a memory for storing a computer program which, when executed by the processor, carries out the steps of the link adjusting method of the power communication network according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when being executed by a processor, carries out the steps of the method of link adjustment of a power communication network according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111647418.8A CN114422438B (en) | 2021-12-29 | 2021-12-29 | Link adjustment method and device for power communication network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111647418.8A CN114422438B (en) | 2021-12-29 | 2021-12-29 | Link adjustment method and device for power communication network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114422438A true CN114422438A (en) | 2022-04-29 |
| CN114422438B CN114422438B (en) | 2023-05-09 |
Family
ID=81269805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111647418.8A Active CN114422438B (en) | 2021-12-29 | 2021-12-29 | Link adjustment method and device for power communication network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114422438B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116016306A (en) * | 2022-12-15 | 2023-04-25 | 中国联合网络通信集团有限公司 | Data traffic forwarding method, device, equipment and storage medium |
| CN117294638A (en) * | 2023-09-21 | 2023-12-26 | 国网江苏省电力有限公司信息通信分公司 | Data transmission path selection method, corresponding system, device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190312898A1 (en) * | 2018-04-10 | 2019-10-10 | Cisco Technology, Inc. | SPATIO-TEMPORAL ANOMALY DETECTION IN COMPUTER NETWORKS USING GRAPH CONVOLUTIONAL RECURRENT NEURAL NETWORKS (GCRNNs) |
| CN112700056A (en) * | 2021-01-06 | 2021-04-23 | 中国互联网络信息中心 | Complex network link prediction method, complex network link prediction device, electronic equipment and medium |
| CN113572697A (en) * | 2021-07-20 | 2021-10-29 | 电子科技大学 | Load balancing method based on graph convolution neural network and deep reinforcement learning |
-
2021
- 2021-12-29 CN CN202111647418.8A patent/CN114422438B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190312898A1 (en) * | 2018-04-10 | 2019-10-10 | Cisco Technology, Inc. | SPATIO-TEMPORAL ANOMALY DETECTION IN COMPUTER NETWORKS USING GRAPH CONVOLUTIONAL RECURRENT NEURAL NETWORKS (GCRNNs) |
| CN112700056A (en) * | 2021-01-06 | 2021-04-23 | 中国互联网络信息中心 | Complex network link prediction method, complex network link prediction device, electronic equipment and medium |
| CN113572697A (en) * | 2021-07-20 | 2021-10-29 | 电子科技大学 | Load balancing method based on graph convolution neural network and deep reinforcement learning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116016306A (en) * | 2022-12-15 | 2023-04-25 | 中国联合网络通信集团有限公司 | Data traffic forwarding method, device, equipment and storage medium |
| CN117294638A (en) * | 2023-09-21 | 2023-12-26 | 国网江苏省电力有限公司信息通信分公司 | Data transmission path selection method, corresponding system, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114422438B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114422438A (en) | Link adjusting method and device of power communication network | |
| CN111612153A (en) | Method and device for training model | |
| EP3931698A1 (en) | Dynamic network configuration | |
| CN109743713B (en) | Resource allocation method and device for electric power Internet of things system | |
| CN116055406B (en) | Training method and device for congestion window prediction model | |
| CN113660325A (en) | Industrial Internet task unloading strategy based on edge calculation | |
| CN111160661B (en) | Method, system and equipment for optimizing reliability of power communication network | |
| CN114465954A (en) | Self-adaptive routing method, device and equipment for special cloud line and readable storage medium | |
| CN113329053B (en) | 5G network virtual mapping method and device based on power service characteristics | |
| CN105340318B (en) | Transmit the determination method and device of congestion | |
| CN108901047A (en) | Base station and terminal cooperation caching method and device based on content popularit variation | |
| CN112364365A (en) | Industrial data encryption method, edge server and computer readable storage medium | |
| CN109831319B (en) | Network function deployment method considering multidimensional resource constraints | |
| CN110460486A (en) | The monitoring method and system of service node | |
| CN113591999B (en) | End-edge cloud federal learning model training system and method | |
| CN113014302B (en) | Network function service chain deployment method facing satellite network | |
| CN115564055A (en) | Asynchronous joint learning training method and device, computer equipment and storage medium | |
| CN114548421A (en) | Optimization processing method and device for federal learning communication overhead | |
| CN114281544A (en) | Electric power task execution method and device based on edge calculation | |
| CN113518122A (en) | Task unloading method, device, equipment and medium for ensuring low-delay transmission by edge intelligent network | |
| CN115481752B (en) | Model training method, device, electronic equipment and storage medium | |
| CN115913991B (en) | Service data prediction method and device, electronic equipment and storage medium | |
| CN117076131B (en) | Task allocation method and device, electronic equipment and storage medium | |
| CN109190753A (en) | The processing method and processing device of neural network, storage medium, electronic device | |
| CN116016524B (en) | Data processing method and device applied to motorized command platform |
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 |