A lossless network construction method supporting multiple scenarios of electric power

Abstract

The application discloses a lossless network construction method under support of various power scenes, comprising the following steps: constructing a network topology graph by acquiring node and link information of a power network, dynamically optimizing a data transmission path by using a planning algorithm, constructing a traffic prediction model to predict a future data traffic variation trend, then, dynamically adjusting network transmission parameters according to a prediction result, finally, establishing a network performance evaluation system and performing real-time evaluation on current network performance, so that the data stability and reliability in the transmission process are effectively improved, and network performance analysis and evaluation can be more comprehensively and deeply performed.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power network transmission, in particular to a lossless network construction method supporting various power scenarios. BACKGROUND

[0002] In a power system, different scenarios have increasingly diverse requirements for networks, and the traditional network has certain limitations in data transmission and cannot meet the needs of various power scenarios, therefore, a new lossless network construction method is needed to improve the performance and reliability of the network.

[0003] In Chinese Patent Publication No. CN 117596208 A, a lossless flow control method suitable for a deterministic network is disclosed, and the core technical solution is that the switch sends a flow control frame to the upper network node to control the data transmission of the best effort transmission flow queue of the upper network node. The upper network node receives the flow control frame sent by the switch, and preferentially sends high-priority flow queue data according to the integration strategy of the deterministic network, and sends best effort transmission flow queue data according to the strict priority strategy. At the same time, the switch sends the flow control frame to the lower network switch according to the received flow control frame, and preferentially sends high-priority flow queue data according to the integration strategy of the deterministic network, and sends best effort transmission flow queue data according to the strict priority strategy. The switch switches different input devices according to the waiting time to ensure that different input devices can be sent.

[0004] In the prior art, when transmitting data in different scenarios, the flow prediction model is not fully used to predict and adjust the data flow changes, and the network performance is not deeply analyzed and evaluated, so it is impossible to ensure the data stability and reliability in the transmission process.

[0005] The application provides a solution to the above-mentioned shortcomings: by obtaining the node and link information of the power network to construct a network topology graph, using a planning algorithm to dynamically optimize the data transmission path, constructing a flow prediction model to predict the future data flow trend, then dynamically adjusting the network transmission parameters according to the prediction results, and finally establishing a network performance evaluation system to evaluate the current network performance in real time, which can effectively improve the data stability and reliability in the transmission process, and can more comprehensively and deeply analyze and evaluate the network performance. SUMMARY

[0006] The application provides a lossless network construction method supporting various power scenarios.

[0007] The technical solution of the application is as follows: a lossless network construction method supporting various power scenarios, comprising:

[0008] S1, acquire node information and link information in different power scenario networks, construct a network topology graph based on the acquired node information and link information, dynamically plan a data transmission path according to the network topology graph using a planning algorithm;

[0009] S2, construct a traffic prediction model to predict data traffic changes in the power network;

[0010] S3, train the traffic prediction model and evaluate the prediction accuracy of the model;

[0011] S4, dynamically adjust network transmission parameters according to the prediction results;

[0012] S5, establish a network performance evaluation system to evaluate network performance in real time.

[0013] Further, in the S1 step, node information and link information in the power scenario network are acquired, a network topology graph is constructed based on the acquired node information and link information, and a data transmission path is dynamically planned according to the network topology graph using a planning algorithm:

[0014] Node information in different power scenario networks is acquired using an Nmap network mapper, the different power scenarios including power generation side, power transmission side and power consumption side, the node information including IP address, MAC address and location, link information in the power scenario network is acquired using an LLDP link layer discovery protocol, the link information including bandwidth parameters, delay parameters and link connection relationships, a network topology graph is constructed using Microsoft Visio drawing software, and an optimal transmission path is confirmed according to the network topology graph using a Dijkstra algorithm;

[0015] Further, the Dijkstra algorithm is set as:

[0016] Set the initial value: initialize the distance of all vertices to be infinite, and set the distance of the source point to be 0;

[0017] Mark the access state: set the access state of all vertices to be unvisited;

[0018] Loop execution:

[0019] Select a vertex with the shortest path estimate from unvisited vertices and set it as the current vertex;

[0020] Mark the current vertex as visited;

[0021] Iterate through all adjacent vertices of the current vertex, calculate the distance from the current vertex to the adjacent vertex, and if the distance is less than the current shortest path estimate value of the adjacent vertex, update the estimate value;

[0022] Repeat step: repeat the above steps until all vertices are visited.

[0023] Further, the S2 step of constructing a traffic prediction model to predict the data traffic changes in the power network:

[0024] In the S1 step, the Corelight traffic analysis tool is deployed in the power scenario network, and historical traffic data including packet size, transmission speed and link utilization are obtained, the data is cleaned and outlier processed, normalized, and an LSTM long short-term memory network algorithm is used to construct a traffic prediction model;

[0025] The collected data is processed for outliers, and the collected data including packet size data, transmission speed data and link utilization data is cleaned, and the z-score function is used to calculate the packet size data, transmission speed data and link utilization data, and the calculation formula is:

[0026] Packet size data:

[0027]

[0028] Wherein, z1 is the standardized value of the packet size data, x1 is the original value of the packet size data, μ1 is the average value of the packet size data, and σ1 is the standard deviation of the packet size data;

[0029] Transmission speed data:

[0030]

[0031] Wherein, z2 is the standardized value of the transmission speed data, x2 is the original value of the transmission speed data, μ2 is the average value of the transmission speed data, and σ2 is the standard deviation of the transmission speed data;

[0032] Link utilization data:

[0033]

[0034] Wherein, z3 is the standardized value of the link utilization data, x3 is the original value of the link utilization data, μ3 is the average value of the link utilization data, and σ3 is the standard deviation of the link utilization data;

[0035] Define |Z|>3 as an outlier, then the normal range interval of Z value is (-3, 3), when the output Z value is not in the normal range interval, it is an outlier, and the outlier is processed, the processing method includes deleting or replacing the outlier;

[0036] The data packet size data, transmission speed data and link utilization data are normalized, the minimum-maximum normalization method is used, the data is normalized to [0, 1], and the calculation formula is:

[0037]

[0038] Wherein, X' represents the normalized value, X represents the original value, X max represents the maximum value, X min represents the minimum value;

[0039] The data sequence output after the abnormal value processing and normalization processing is A1={a1, a2, …a n};

[0040] The model architecture is defined, the prediction model architecture is constructed by the LSTM algorithm, the prediction model architecture includes a plurality of memory units, each unit includes a forgetting gate, an input gate and an output gate, the data sequence A1={a1, a2, …a n} is taken as an input sequence, wherein a i is the traffic feature vector of time point i, and the calculation of the LSTM unit at t time step is:

[0041] The calculation formula of the input gate is:

[0042] i t =σ(x t W xi +h t-1 W hi +b i );

[0043]

[0044] Wherein, i t is the output of the input gate, sigma is the sigmoid activation function, W xi , W hi , W hc are the weights of the input gate, is the candidate memory cell state output, b i , b c are the biases of the input gate and the candidate memory cell respectively, h t-1 is the hidden state of the last time step, x t is the input of the current time step, and tan h represents the tangent activation function.

[0045] The calculation formula of the forgetting gate is:

[0046] f t =σ(W xf x t +W hf ht-1 +b f )

[0047] where f t is the output of the forget gate, σ is the sigmoid activation function, W xf and W hf are the weights of the forget gate, b f is the bias of the forget gate, h t-1 is the hidden state of the previous time step, and x t is the input of the current time step;

[0048] The unit state update calculation formula is:

[0049]

[0050] where C t is the unit state at time t, f t is the output of the forget gate, C t-1 is the unit state of the previous time step, i t is the output of the input gate, is the output of the unit update state;

[0051] The calculation formula of the output gate is:

[0052] o t = σ(W xo x t +W ho h t-1 +b o ),

[0053] h t = o t · tanh(c t )

[0054] where o t is the output of the output gate, σ is the sigmoid activation function, W xo and W ho are the weights of the output gate, b o is the bias of the output gate, h t-1 is the hidden state of the previous time step, h t is the current hidden layer state update output, x t is the input of the current time step, C t is the unit state at time t, and tanh represents the tangent activation function.

[0055] Further, the traffic prediction model is trained in the S3 step, and the prediction accuracy of the model is evaluated:

[0056] The training cycle is built, the pretreated historical traffic data is divided into 70% training set and 30% validation set, the training set is used for training the traffic prediction model, the validation set is used for evaluating the accuracy of the model, the LSTM model is trained using the back propagation training algorithm, and the hyperparameters are set, including setting the learning rate to 0.01, setting the training round to 100 times, and setting the batch size to 32, and calculating the loss function in the training cycle, the loss function is selected as the mean square error, the Adam optimizer is selected to update the weights of the model, and the specific training steps are:

[0057] S301, initialize the weights and biases of the model;

[0058] S302, select 32 sample data from the training set for training;

[0059] S303, according to the current weight and bias, forward propagation is performed to obtain the predicted value;

[0060] S304, calculate the loss between the predicted value and the true value according to the MSE loss function;

[0061] S305, calculate the gradient of the loss with respect to the model parameters using the back propagation algorithm;

[0062] S306, update the model parameters using the Adam optimizer according to the gradient;

[0063] S307, repeat steps S302 to S306 until the loss function reaches the preset training round, stop training;

[0064] The trained model is evaluated using the validation set, the K-fold cross-validation method is used to evaluate the prediction accuracy of the model, K is preset to 10, the validation set is divided into K equal subsets, and K times are repeated, each time a subset is used as the test set, and the remaining K-1 subsets are used as the training set. In each iteration, the training set is used to train the logistic regression model, the correct sample number is recorded, and the accuracy of each round is calculated: (the number of correctly predicted samples) / (the total number of samples), and then the average accuracy is calculated: (the sum of the accuracy of each round) / 10.

[0065] Further, the S4 step dynamically adjusts the network transmission parameters according to the prediction result:

[0066] Collect the current network traffic value, use the trained traffic prediction model in the S3 step to predict the traffic change in a period of time, according to the prediction result, divide into three results of traffic increase, traffic unchanged and traffic decrease, compare the predicted value with the current value to judge the category of traffic change, and judge:

[0067] When the predicted value is greater than the current traffic value, it is determined that the traffic is increasing;

[0068] When the prediction value is close to the current traffic value by 70%, it is determined that the traffic is unchanged;

[0069] When the prediction value is less than the current traffic value, it is determined that the traffic is reduced;

[0070] Adjust the network transmission parameters according to different traffic categories, and the specific measures are:

[0071] When the traffic increases, increase the network bandwidth allocation;

[0072] When the traffic is unchanged, maintain the current network parameter settings;

[0073] When the traffic is reduced, reduce the network resource allocation.

[0074] Further, the S5 step of establishing a network performance evaluation system for real-time evaluation of network performance:

[0075] Continuously collect traffic data predicted by the traffic prediction model, including packet size, transmission speed and link utilization rate, calculate the average packet size, packet size standard deviation, average transmission speed, transmission speed standard deviation, average link utilization rate and link utilization rate standard deviation according to the latest traffic data collected, define a performance evaluation function according to the calculated average and standard deviation, output the network performance according to the performance evaluation function, and define the performance evaluation function using a weighted average algorithm;

[0076] Pre-set to collect traffic data every 5 minutes, including packet size data P, transmission speed data V and link utilization rate data L, calculate the average and standard deviation of the collected traffic data, and the calculation is:

[0077] Packet size average:

[0078]

[0079] Wherein, Packet size average, P i The size of the i-th packet, n is the total number of packets;

[0080] Packet size standard deviation:

[0081]

[0082] Wherein, S P Packet size standard deviation, P i The size of the i-th packet, Packet size average, n is the total number of packets;

[0083] Transmission speed average:

[0084]

[0085] wherein, is the average of the transmission speed, V i is the transmission speed of the ith time period, n is the total number of time periods;

[0086] the standard deviation of the transmission speed:

[0087]

[0088] wherein, S V is the standard deviation of the transmission speed, V i is the transmission speed of the ith time period, is the average of the transmission speed, n is the total number of time periods;

[0089] the average of the link utilization:

[0090]

[0091] wherein, is the average of the link utilization, L i is the link utilization of the ith time period, n is the total number of time periods;

[0092] the standard deviation of the link utilization:

[0093]

[0094] wherein, S L is the standard deviation of the link utilization, L i is the link utilization of the ith time period, is the average of the link utilization, n is the total number of time periods;

[0095] According to the calculated S P , S V , S L , the performance evaluation function f(x) is defined as:

[0096]

[0097] wherein, a, b, g, d, e, are weight coefficients, are the target average packet size, the target average transmission speed and the target average link utilization, respectively, S Pmax , S Vmax , S LmaxThe maximum value of the data packet size standard deviation, the transmission speed standard deviation and the link utilization rate standard deviation, respectively;

[0098] The threshold of the preset performance evaluation, when the performance evaluation function value is greater than 0.8, it represents that the network performance is good, when the performance evaluation function value is less than 0.8, it represents that the current network performance is poor, and the processing measures are taken to handle, and the processing measures are:

[0099] The data compression ratio is automatically adjusted, and large data packets are divided into smaller sizes;

[0100] The optimal transmission path is recalculated;

[0101] The bandwidth resource is redistributed, and more bandwidth is preferentially allocated to critical tasks and tasks with high transmission speed requirements.

[0102] Beneficial effects:

[0103] The present application aims at the shortcomings that the data stability and reliability in the transmission process cannot be ensured in the prior art, constructs a network topology graph by acquiring node and link information of the power network, dynamically optimizes the data transmission path by using a planning algorithm, constructs a traffic prediction model to predict the future data traffic trend, then dynamically adjusts the network transmission parameters according to the prediction result, finally establishes a network performance evaluation system to evaluate the current network performance in real time, which effectively improves the data stability and reliability in the transmission process, and can more comprehensively and deeply analyze and evaluate the network performance. BRIEF DESCRIPTION OF DRAWINGS

[0104] Fig. 1 It is a structure block diagram of a lossless network construction method supporting multiple scenarios of electric power in the embodiment of the present application;

[0105] Fig. 2 It is a flow step diagram of a lossless network construction method supporting multiple scenarios of electric power in the embodiment of the present application;

[0106] Fig. 3 It is a traffic prediction model training step flow chart of a lossless network construction method supporting multiple scenarios of electric power in the embodiment of the present application. DETAILED DESCRIPTION

[0107] In order to make the purpose and advantages of the present application more clear and explicit, the present application will be further described below in combination with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present application, and do not limit the present application.

[0108] The preferred implementation methods of the present application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these implementation methods are only used to explain the technical principles of the present application, and do not limit the protection scope of the present application.

[0109] As used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. It will be further understood that the terms "comprises", "comprising", "includes" and / or "including", or the like, when used in this specification, specify the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

[0110] Referring to Figs. 1-3 As shown in the figure, a method for constructing a lossless network in various power scenarios includes:

[0111] S1, acquiring node information and link information in different power scenario networks, constructing a network topology graph based on the acquired node information and link information, and dynamically planning a data transmission path according to the network topology graph using a planning algorithm;

[0112] S2, constructing a traffic prediction model to predict data traffic changes in the power network;

[0113] S3, training the traffic prediction model and evaluating the prediction accuracy of the model;

[0114] S4, dynamically adjusting network transmission parameters according to the prediction results;

[0115] S5, establishing a network performance evaluation system to evaluate network performance in real time.

[0116] As Fig. 2 shown, in step S1, node information and link information in different power scenario networks are acquired, a network topology graph is constructed based on the acquired node information and link information, and a data transmission path is dynamically planned according to the network topology graph using a planning algorithm.

[0117] Specifically, the Nmap network mapper is used to acquire node information in different power scenario networks, the different power scenarios including the power generation side, the power transmission side and the power consumption side, the node information including IP addresses, MAC addresses and locations, the LLDP link layer discovery protocol is used to acquire link information in the power scenario network, the link information including bandwidth parameters, delay parameters and link connection relationships, Microsoft Visio drawing software is used to construct a network topology graph, and Dijkstra algorithm is used to confirm the optimal transmission path according to the network topology graph;

[0118] The Dijkstra algorithm is set as:

[0119] Set initial value: initialize the distance of all vertices to be infinite, and set the distance of the source point to be 0;

[0120] Mark access state: set the access state of all vertices to be unvisited;

[0121] Loop execution:

[0122] Select a vertex with the shortest path estimate from unvisited vertices and set it as the current vertex;

[0123] Mark the current vertex as visited;

[0124] Traverse all adjacent vertices of the current vertex, calculate the distance from the current vertex to the adjacent vertex, and if the distance is less than the current shortest path estimate of the adjacent vertex, update the estimate;

[0125] Repeat step: repeat the above steps until all vertices are visited.

[0126] As shown in S2, a traffic prediction model is constructed in the S2 step to predict the data traffic changes in the power network. Fig. 2

[0127] Specifically, in the S1 step, the Corelight traffic analysis tool is deployed in the nodes of the power scenario network, historical traffic data is obtained, including packet size, transmission speed and link utilization, the data is cleaned and outlier processing is performed, the data is normalized, and an LSTM long short-term memory network algorithm is used to construct a traffic prediction model.

[0128] The collected data is subjected to outlier processing, and the collected data, including packet size data, transmission speed data and link utilization data, is cleaned. The z-score function is used to calculate the packet size data, transmission speed data and link utilization data, and the calculation formula is:

[0129] Packet size data:

[0130]

[0131] Wherein, z1 is the standardized value of the packet size data, x1 is the original value of the packet size data, μ1 is the average value of the packet size data, and σ1 is the standard deviation of the packet size data.

[0132] Transmission speed data:

[0133]

[0134] Wherein, z2 is the standardized value of the transmission speed data, x2 is the original value of the transmission speed data, μ2 is the average value of the transmission speed data, and σ2 is the standard deviation of the transmission speed data.​

[0135] Link utilization data:

[0136]

[0137] wherein z3 is the value of the link utilization data after standardization, x3 is the original value of the link utilization data, μ3 is the average value of the link utilization data, and σ3 is the standard deviation of the link utilization data;

[0138] an abnormal value is defined when |Z|>3, and the normal range interval of the Z value is (-3, 3). When the output Z value is not in the normal range interval, it is an abnormal value, and the abnormal value is processed, including deleting or replacing the abnormal value;

[0139] The packet size data, the transmission speed data and the link utilization data are normalized, and the minimum-maximum normalization method is used to normalize the data to [0, 1], and the calculation formula is:

[0140]

[0141] wherein X' represents the normalized value, X represents the original value, X max represents the maximum value, and X min represents the minimum value;

[0142] The data sequence output after the abnormal value processing and the normalization processing is A1={a1, a2, …a n};

[0143] The model architecture is defined, an LSTM algorithm is used to construct a prediction model architecture, the prediction model architecture includes a plurality of memory units, each unit includes a forgetting gate, an input gate and an output gate, and the data sequence A1={a1, a2, …a n} is used as an input sequence, wherein a i is a traffic feature vector at time point i, and the calculation of the LSTM unit at t time step is:

[0144] The calculation formula of the input gate is:

[0145] i t =σ(x t W xi +h t-1 W hi +b i );

[0146]

[0147] wherein i t is the output of the input gate, σ is a sigmoid activation function, W xi , Whi , W hc is the weight of the input gate, is the output of the candidate memory cell state, b i , b c are the biases of the input gate and the candidate memory cell, respectively, h t-1 is the hidden state of the previous time step, x t is the input of the current time step, and tanh represents the tangent activation function.

[0148] The calculation formula of the forget gate is:

[0149] f t = σ(W xf x t + W hf h t-1 + b f )

[0150] where f t is the output of the forget gate, σ is the sigmoid activation function, W xf and W hf are the weights of the forget gate, b f is the bias of the forget gate, h t-1 is the hidden state of the previous time step, and x t is the input of the current time step.

[0151] The calculation formula of the cell state update is:

[0152]

[0153] where C t is the cell state at time t, f t is the output of the forget gate, C t-1 is the cell state of the previous time step, i t is the output of the input gate, and C t is the output of the cell update state.

[0154] The calculation formula of the output gate is:

[0155] o t = σ(W xo x t + W ho h t-1 + b o ),

[0156] h t = o t · tanh(c t )

[0157] where o tis the output of the output gate, σ is the sigmoid activation function, W xo and W ho is the weight of the output gate, b o is the bias of the output gate, h t-1 is the hidden state of the previous time step, h t is the current hidden layer state update output, x t is the input of the current time step, C t is the cell state at time t, and tanh represents the tangent activation function.

[0158] like Fig. 2 As shown, in step S3, the traffic prediction model is trained and the prediction accuracy of the model is evaluated.

[0159] Specifically, a training loop is built, and the preprocessed historical traffic data is split into a 70% training set and a 30% validation set. The training set is used to train the traffic prediction model, and the validation set is used to evaluate the accuracy of the model. The LSTM model is trained using the backpropagation training algorithm, and hyperparameters are set, including setting the learning rate to 0.01, the number of training rounds to 100, and the batch size to 32. The loss function in the training loop is calculated, and the mean squared error is selected as the loss function. The Adam optimizer is selected to update the model weights. The specific training steps are as follows:

[0160] S301, initialize the weights and biases of the model;

[0161] S302, select 32 sample data from the training set for training;

[0162] S303: Perform forward propagation calculation based on the current weight and bias to obtain a predicted value;

[0163] S304, calculating the loss between the predicted value and the true value according to the MSE loss function;

[0164] S305, using the back propagation algorithm to calculate the gradient of the loss with respect to the model parameters;

[0165] S306, using the Adam optimizer to update the model parameters according to the gradient;

[0166] S307, repeating steps S302 to S306 until the loss function reaches a preset number of training rounds, and then stopping the training;

[0167] The trained model is evaluated using the validation set, and the prediction accuracy of the model is evaluated using the K-fold cross-validation method. The preset K = 10, the validation set is divided into K equal subsets, and K times are repeated. In each iteration, one subset is used as the test set, and the remaining K-1 subsets are used as the training set. In each iteration, the training set is used to train the logistic regression model, the number of correct samples is recorded, the accuracy of each round is calculated: (the number of correctly predicted samples) / (the total number of samples), and then the average accuracy is calculated: (the sum of the accuracy of each round) / 10.

[0168] Wherein, the forward propagation is the process of data from the input layer through a series of calculations and processing, gradually passing to the output layer, and the back propagation is the process of updating the model parameters by calculating the gradient according to the error between the model output and the true value.

[0169] As shown in the S4 step, the network transmission parameters are dynamically adjusted according to the prediction result. Fig. 2

[0170] Specifically, the current network traffic value is collected, the traffic prediction model trained in the S3 step is used to predict the traffic change in a period of time in the future, and according to the prediction result, the traffic is divided into three results: traffic increase, traffic unchanged and traffic decrease. The predicted value is compared with the current value to judge the category of traffic change, and the judgment is:

[0171] When the predicted value is greater than the current traffic value, it is determined that the traffic is increasing;

[0172] When the predicted value is close to the current traffic value by 70%, it is determined that the traffic is unchanged;

[0173] When the predicted value is less than the current traffic value, it is determined that the traffic is decreasing;

[0174] According to different traffic categories, the network transmission parameters are adjusted, and the specific measures are:

[0175] When the traffic increases, the network bandwidth allocation is increased;

[0176] When the traffic is unchanged, the current network parameter setting is maintained;

[0177] When the traffic decreases, the network resource allocation is reduced.

[0178] As shown in the S5 step, a network performance evaluation system is established to evaluate the network performance in real time. Fig. 2

[0179] ​​Specifically, the traffic data predicted by the traffic prediction model is continuously collected, including packet size, transmission speed and link utilization, according to the latest collected traffic data, the average packet size, packet size standard deviation, average transmission speed, transmission speed standard deviation, average link utilization and link utilization standard deviation are calculated, the performance evaluation function is defined according to the calculated average and standard deviation, the network performance is output according to the performance evaluation function, and the performance evaluation function is defined using the weighted average algorithm;

[0180] The preset traffic data is collected every 5 minutes, including packet size data P, transmission speed data V and link utilization data L, and the average and standard deviation of the collected traffic data are calculated, and the calculation is as follows:

[0181] Packet size average:

[0182]

[0183] Wherein, Packet size average, P i The size of the i-th packet, n is the total number of packets;

[0184] Packet size standard deviation:

[0185]

[0186] Wherein, S P Packet size standard deviation, P i The size of the i-th packet, Packet size average, n is the total number of packets;

[0187] Transmission speed average:

[0188]

[0189] Wherein, Transmission speed average, V i The transmission speed of the i-th time period, n is the total number of time periods;

[0190] Transmission speed standard deviation:

[0191]

[0192] Wherein, S V Transmission speed standard deviation, V i The transmission speed of the i-th time period, Transmission speed average, n is the total number of time periods;

[0193] Link utilization average:

[0194]

[0195] wherein, is the average of link utilization, L i is the link utilization of the i th time period, n is the total number of time periods;

[0196] The standard deviation of link utilization is:

[0197]

[0198] wherein, S L is the standard deviation of link utilization, L i is the link utilization of the i th time period, is the average of link utilization, n is the total number of time periods;

[0199] According to the calculated S P , S V , S L , the performance evaluation function f(x) is defined as:

[0200]

[0201] wherein, alpha, beta, gamma, delta, epsilon, are weight coefficients, are the target average packet size, the target average transmission speed and the target average link utilization, respectively, S Pmax , S Vmax , S Lmax are the maximum values of the packet size standard deviation, the transmission speed standard deviation and the link utilization standard deviation, respectively;

[0202] The threshold of the preset performance evaluation is set, when the performance evaluation function value is greater than 0.8, it indicates that the network performance is good, when the performance evaluation function value is less than 0.8, it indicates that the current network performance is poor, and the processing measures are taken to handle, the processing measures are:

[0203] The data compression ratio is automatically adjusted, and large data packets are divided into smaller sizes;

[0204] The optimal transmission path is recalculated;

[0205] The bandwidth resource is redistributed, and more bandwidth is preferentially allocated to critical tasks and tasks with high transmission speed requirements.

[0206] As Figs. 1-3 shown, the present application is used in the scene of solving the shortcoming of the prior art that cannot guarantee the accuracy of data in the simulation analysis process, and works as follows:

[0207] acquire node information and link information in different power scenario networks, construct a network topology graph based on the acquired node information and link information, dynamically plan a data transmission path according to the network topology graph by using a planning algorithm;

[0208] construct a traffic prediction model to predict data traffic changes in the power network;

[0209] train the traffic prediction model and evaluate the prediction accuracy of the model;

[0210] dynamically adjust network transmission parameters according to the prediction results;

[0211] establish a network performance evaluation system to perform real-time evaluation of network performance.

[0212] Thus, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the accompanying drawings, but it is readily understood by those skilled in the art that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after such changes or replacements will all fall within the protection scope of the present application.

[0213] The above description is only for the preferred embodiments of the present application and is not intended to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

Claims

1. A method for constructing a lossless network supporting multiple power scenarios, characterized by: S1. Obtain node information and link information in networks of different power scenarios, construct a network topology based on the obtained node information and link information, and dynamically plan a data transmission path using a planning algorithm based on the network topology; S2. Build a traffic prediction model to predict data traffic changes in the power network; S3. Training the traffic prediction model and evaluating the prediction accuracy of the model; S4. Dynamically adjust network transmission parameters based on the prediction results; S5. Establish a network performance evaluation system to conduct real-time evaluation of network performance; In step S3, the traffic prediction model is trained and the prediction accuracy of the model is evaluated as follows: Set up a training loop and split the preprocessed historical traffic data into a 70% training set and a 30% validation set. The training set is used to train the traffic prediction model, and the validation set is used to evaluate the model's accuracy. Use the backpropagation training algorithm to train the LSTM model, set hyperparameters, calculate the loss function in the training loop, and select the Adam optimizer to update the model's weights. In step S4, the network transmission parameters are dynamically adjusted according to the prediction results: Collect the current network traffic value and use the traffic prediction model trained in step S3 to predict the traffic changes in the future. According to the prediction results, the results are divided into three categories: traffic increase, traffic unchanged, and traffic decrease. The type of traffic change is determined by comparing the predicted value with the current value.

2. The method for constructing a lossless network supporting multiple power scenarios according to claim 1 is characterized by: In step S1, node information and link information in different power scenario networks are obtained, a network topology is constructed based on the obtained node information and link information, and a planning algorithm is used according to the network topology to dynamically plan the data transmission path as follows: A network mapper is used to obtain node information in different power scenario networks, where the different power scenarios include the power generation side, the power transmission side, and the power consumption side. The node information includes IP address, MAC address, and location. A link layer discovery protocol is used to obtain link information in the power scenario network, where the link information includes bandwidth parameters, delay parameters, and link connection relationships. Drawing software is used to construct a network topology diagram, and the Dijkstra algorithm is used to confirm the optimal transmission path based on the network topology diagram.

3. The method for constructing a lossless network supporting multiple power scenarios according to claim 1, characterized in that: In step S2, a traffic prediction model is constructed to predict the data traffic changes in the power network as follows: In the S1 step, a traffic analysis tool is deployed on nodes in the power scenario network to obtain historical traffic data, including packet size, transmission speed, and link utilization. The data is cleaned and outliers are processed, the data is normalized, and a traffic prediction model is constructed using the LSTM long short-term memory network algorithm.

4. The method for constructing a lossless network supporting multiple power scenarios according to claim 1 is characterized by: In step S5, a network performance evaluation system is established to perform real-time evaluation of network performance as follows: Continuously collect traffic data predicted by the traffic prediction model, including packet size, transmission speed, and link utilization. Based on the latest collected traffic data, calculate the average packet size, packet size standard deviation, average transmission speed, transmission speed standard deviation, average link utilization, and link utilization standard deviation. Define a performance evaluation function based on the calculated average and standard deviation. Evaluate network performance based on the output of the performance evaluation function. Define the performance evaluation function to use a weighted average algorithm.

5. The method for constructing a lossless network supporting multiple power scenarios according to claim 2 is characterized in that: The Dijkstra algorithm is set up as: Set initial values: Initialize the distances of all vertices to infinity and the distance of the source point to 0; Mark the visited state: the visited state of all vertices is set to unvisited; Loop execution: Select the vertex with the shortest path estimate from the unvisited vertices and set it as the current vertex; Mark the current vertex as visited; Traverse all adjacent vertices of the current vertex and calculate the distance from the current vertex to the adjacent vertex. If the distance is less than the current shortest path estimate of the adjacent vertex, update the estimate. Repeat steps: Repeat the above steps until all vertices are visited.

6. The method for constructing a lossless network supporting multiple power scenarios according to claim 1, characterized in that: The training steps are: S301, initialize the weights and biases of the model; S302, selecting sample data from the training set for training; S303: Perform forward propagation calculation based on the current weight and bias to obtain a predicted value; S304, calculating the loss between the predicted value and the true value according to the loss function; S305, using the back propagation algorithm to calculate the gradient of the loss with respect to the model parameters; S306, using the Adam optimizer to update the model parameters according to the gradient; S307: Repeat steps S302 to S306 until the loss function reaches a preset number of training rounds, and then stop training.

7. The method for constructing a lossless network supporting multiple power scenarios according to claim 1, characterized in that: Traffic flow is determined as follows: When the predicted value is greater than the current flow value, it is determined that the flow rate has increased; When the predicted value is 70% close to the current flow value, the flow is determined to be unchanged; When the predicted value is less than the current flow value, it is determined that the flow is decreasing.

8. The method for constructing a lossless network supporting multiple power scenarios according to claim 4, characterized in that: The weighted average algorithm is: in, is the weight coefficient, are the target average packet size, target average transmission speed, and target average link utilization, respectively. are the maximum values ​​of the standard deviation of packet size, transmission speed and link utilization respectively.

Images