A power transmission data control method and system for a smart grid

Abstract

The application discloses a power transmission data control method and system of a smart grid, and relates to the technical field of power data control. The power transmission data control method comprises the following steps: an edge control node divides power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data. The edge control node can directly process power transmission data and locally generate control instructions for emergency power transmission standard data, without relying on a remote control center, so that the effect of reducing data transmission links and avoiding emergency data preemption of resources is achieved. The response time of control instructions during a fault can be greatly shortened, line overload tripping and power supply interruption accidents caused by instruction lag can be avoided, and the stable operation of a power transmission line is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power data control, in particular to a power transmission data control method and system of a smart grid. BACKGROUND

[0002] The smart grid is the intelligentization of the power grid, which is based on an integrated and high-speed bidirectional communication network, and realizes the goals of reliability, safety, economy, high efficiency, environmental friendliness and use safety of the power grid through the application of sensing and measurement technology, equipment technology, control methods and advanced decision support system technology. Its main features include self-healing, motivation and protection of users, resistance to attacks, provision of power quality that meets user needs, allowance of the connection of various power generation forms, start of the power market and optimized and efficient operation of assets. The power transmission data control of the smart grid is a comprehensive system that realizes the collection, analysis and control of power transmission data through the cooperation of the dispatching control system of the smart grid and the smart power transmission line.

[0003] The existing power transmission data control of the smart grid mostly adopts a centralized control center unified processing mode. The power transmission lines are uniformly transmitted to the control center through data transmission for processing and analysis, and then transmitted back to the specific position of the power transmission line for execution of the instructions. Because the power transmission lines cover a wide area and have a large amount of data, and the data is not classified and prioritized for transmission, the data transmission is delayed, the emergency fault data and the normal data occupy the resources, which leads to a lag in the response of the control instructions in the event of a fault, the power transmission line cannot adjust and execute the control instructions in time, and accidents such as line overload trip and power supply interruption occur. At the same time, the real-time transmission of non-processed statistical power transmission standard data occupies the bandwidth, affecting the transmission efficiency of the control instructions, and it is difficult to meet the accurate and real-time regulation and control requirements of the power transmission data. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a power transmission data control method and system of a smart grid, which solves the problems mentioned in the background art.

[0005] To achieve the above purpose, the present application realizes the following technical scheme: a power transmission data control method of a smart grid, comprising the following steps:

[0006] S1, acquire the regional map of the power transmission line, divide the region of the power transmission line into grid regions according to the function, each grid region is provided with an edge control node, the edge control node collects the power transmission line data of the power transmission line, processes the power transmission line data to obtain power transmission standard data;

[0007] S2. The edge control node divides the power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data. The power transmission standard data of the emergency power transmission standard data is used by the edge control node to generate control commands using a multi-parameter control model. The control commands are directly transmitted to the execution terminal for execution control.

[0008] S3. For conventional transmission standard data, the edge control node generates the request instruction and transmits the request instruction and transmission standard data to the control center. The control center sends back the operation instruction to the corresponding edge control node, and the control instruction is then transmitted to the execution terminal to execute the operation instruction. For statistical transmission standard data, the transmission standard data is transmitted in batches to the control center for storage and analysis at preset intervals, and a prediction model corresponding to the transmission line is generated.

[0009] S4. When the execution terminal receives the control command, the execution terminal completes the operation corresponding to the control command and feeds back the result of the operation to the corresponding edge control node. The edge control node compares the power transmission standard data before and after the operation. If the comparison of the two power transmission standard data reaches the adjustment preset standard, a power transmission control is completed. If the comparison of the two power transmission standard data does not reach the adjustment preset standard, the control command is regenerated to perform power transmission control of the power transmission line.

[0010] Preferably, the method for constructing the multi-parameter control model includes the following steps:

[0011] Fault and normal operation data for the corresponding transmission line over the past three years are collected. The operation data includes line current, line voltage, line temperature, and power factor. The operation data is preprocessed to remove outliers and then standardized to form the original dataset.

[0012] The original dataset is divided into a training set and a test set in a 7:3 ratio. A basic model is built based on a BP neural network. The basic model includes an input layer, a hidden layer, and an output layer. The basic model is trained using the training set, and the weight parameters are iteratively optimized. The training of the basic model is stopped when the training set error is less than or equal to the corresponding threshold.

[0013] The test set is input into the trained base model to verify the accuracy of the control command output. If the error is greater than a preset threshold, the input layer, hidden layer and output layer of the base module are readjusted until the accuracy of the control command output meets the standard, thus obtaining the multi-parameter control model.

[0014] The qualified multi-parameter control model is fixedly distributed on each edge control node, and a parameter update interface for the multi-parameter control model is set on each edge control node. The multi-parameter control model supports periodic iterative optimization based on the running data through the update interface.

[0015] Preferably, the basic model is built based on a BP neural network, and the specific construction method of the basic model, including the input layer, hidden layer, and output layer, includes:

[0016] The number and function of neurons in the input layer, hidden layer, and output layer are determined to match the application scenario requirements, specifically as follows:

[0017] The input layer to the hidden layer, and the output of the hidden layer neurons are:

[0018]

[0019] in, For input parameters, For the input layer The first neuron is connected to the hidden layer. The connection weights of each neuron For the hidden layer number Bias of each neuron The activation function used for the hidden layer;

[0020] From the hidden layer to the output layer, the predicted value of the output layer is:

[0021]

[0022] in, For the output layer The predicted value of each neuron, For the hidden layer number The output value of each neuron, that is, the output result of the hidden layer neurons mentioned above. For the hidden layer number The nth neuron to the output layer The connection weights of each neuron For the output layer Bias of each neuron The activation function used for the output layer.

[0023] Preferably, the training set is used to train the base model, and the iterative optimization of the weight parameters specifically involves:

[0024] The output layer weights are optimized as follows:

[0025]

[0026] in, For the hidden layer number The first neuron to the output layer The weight update amount of each neuron, if A positive value indicates that the current weight is too small and needs to be increased. A negative value indicates that the current weight is too high and needs to be reduced. For learning rate, For the output layer The true target value of each neuron For the output layer The predicted value of each neuron, The output layer activation function The derivative of For the hidden layer number The output value of each neuron;

[0027] The input layer weights are optimized as follows:

[0028]

[0029] in, For the input layer The first neuron is connected to the hidden layer. The weight update amount of each neuron. For the hidden layer number The first neuron to the output layer The weights of each neuron, This is an approximation of the derivative of the ReLU activation function in the hidden layer. For the input layer Input data for each neuron.

[0030] Preferably, the data transmission and interaction between the edge control node and the control center for conventional and statistical transmission standard data is encrypted, specifically:

[0031] The control center generates a root key for the national cryptographic SM4 algorithm and an SM9 identifier key. It distributes the root key of the SM4 algorithm to each edge control node through an offline secure channel, deploys the public key of the SM9 identifier key to the edge control node, and stores the private key of the SM9 identifier key in the control center. The edge control node generates a local session key based on the root key.

[0032] When the edge control node and the control center transmit the request command, the control command, and the power transmission standard data, the content is encrypted with the local session key to generate first encrypted data. The first encrypted data is transmitted through the interaction gateway and then encrypted again using the AES-256 algorithm to generate second encrypted data, which is then transmitted to the control center.

[0033] After receiving the second encrypted data, the control center first decrypts it using the AES-256 algorithm to obtain the first encrypted data, and then decrypts the first encrypted data using the local session key corresponding to the edge control node to obtain the original data.

[0034] After the control center generates the control command, it encrypts the original data using the private key of the SM9 identifier key to generate encrypted data for transmission back.

[0035] The encrypted data is transmitted to the edge control node via an encrypted link. The edge control node uses the public key of the SM9 identifier key to decrypt the encrypted data to obtain the control command.

[0036] Preferably, the criteria for determining the emergency transmission standard data, routine transmission standard data, and statistical transmission standard data are as follows:

[0037] When the actual value corresponding to the transmission standard data exceeds the safety threshold, it is determined to be emergency transmission standard data. When the actual value corresponding to the transmission standard data is within the safety threshold and the power factor is less than the threshold, it is determined to be regular transmission standard data. When the transmission standard data is 24-hour periodic historical data used for the operation and maintenance of the transmission line, it is determined to be statistical transmission standard data. The transmission priority of emergency transmission standard data is higher than that of regular transmission standard data, and the transmission priority of regular transmission standard data is higher than that of statistical transmission standard data.

[0038] Preferably, the step of transmitting the statistical transmission standard data to the control center for storage and analysis in batches at preset intervals, and generating a prediction model corresponding to the transmission line, includes:

[0039] The edge control node transmits the statistical transmission standard data in batches to the control center at preset intervals.

[0040] The control center processes the power transmission standard data and extracts the daily / weekly / monthly load peak characteristics of the power transmission standard data to form a feature dataset. The feature dataset is divided into a training set and a validation set in an 8:2 ratio.

[0041] The prediction model is constructed based on an LSTM network. The prediction model uses the root mean square error (RMSE) as the loss function. The prediction model is trained using the Adam optimizer with gradient descent optimization algorithm until the RMSE of the validation set is less than a preset function threshold.

[0042] The control center calls the prediction model daily to predict the operating parameters of the transmission line at a corresponding time in the future. If the predicted value of the operating parameter is higher than a safe preset threshold, an early warning message is pushed. The prediction model is trained monthly with new incremental operating data to adapt to the load changes of the transmission line.

[0043] A power transmission data control system for a smart grid, applied to the aforementioned power transmission data control method for a smart grid, the power transmission data control system comprising: an edge control node module, an execution terminal module, a data type classification module, and a data result comparison module;

[0044] The edge control node module is deployed in each of the functionally divided grid areas and is used to perform data processing, data classification, data transmission and command execution on the transmission lines in each of the grid areas.

[0045] The execution terminal module is used to receive the control command issued by the edge control node module, execute the corresponding control operation of the transmission line, and feed back the result of the control operation to the edge control node module;

[0046] The data type classification module is integrated into the edge control node module. It is used to receive the power transmission standard data output by the edge control node module, classify the power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data, and feed the classification data back to the edge control node module.

[0047] The data result comparison module, integrated within the edge control node module, is used to receive the results of the control operation from the execution terminal module, retrieve the power transmission standard data of the edge control node module before and after the control operation, compare them according to a preset adjustment standard, and feed the comparison results back to the edge control node module to confirm the validity of the control operation from the execution terminal module.

[0048] Preferably, the power transmission data control system further includes: a data interaction module and a data acquisition module;

[0049] The data interaction module is used for transmitting request instructions and control instructions for conventional power transmission standard data between the edge control node module and the control center, and for transmitting power transmission standard data for statistical power transmission standard data from the edge control node module to the control center. It is also used for data transmission between the edge control node module and the execution terminal module.

[0050] The data acquisition module is deployed on the transmission lines in each of the grid areas to collect data on the current, voltage, line temperature, and power factor of the transmission lines, and transmit the collected data to the edge control node module.

[0051] Preferably, the data acquisition module includes a current transformer, a voltage transformer, and a fiber optic temperature sensor. The current transformer, voltage transformer, and fiber optic temperature sensor are distributed and installed on the transmission towers of each transmission line, and are connected to the edge control node module through a general data communication interface.

[0052] This invention provides a method and system for controlling power transmission data in a smart grid. It offers the following advantages:

[0053] (1) By dividing the grid according to the function of the transmission line area and deploying local edge control nodes, the edge control nodes can directly process the transmission data and generate control commands locally for emergency transmission standard data. They do not need to rely entirely on the remote control center, which reduces the data transmission links and avoids the effect of emergency data competing for resources. This can significantly shorten the response time of control commands during faults, avoid accidents such as line overload tripping and power outage caused by command delays, and ensure the stable operation of the transmission line.

[0054] (2) By dividing the power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data through edge control nodes, different types of data are transmitted according to priority, and statistical power transmission standard data is transmitted in batches at preset intervals. This achieves the effect of avoiding non-emergency data occupying real-time transmission bandwidth and optimizing the allocation of data transmission resources, which can improve the transmission efficiency of control commands, ensure the transmission stability of regular power transmission standard data and control center, meet the needs of precise control of power transmission data, and at the same time, when emergency power transmission standard data needs to be processed, there is no need to transmit data to the control center. The local multi-parameter control model is used to directly analyze and judge the situation, and can quickly respond to and execute plans for emergency situations, saving the time of data transmission round trip and timely processing to ensure the safety of the power grid.

[0055] (3) By comparing the power transmission standard data before and after the execution of the terminal operation through the edge control node, it is determined whether the data adjustment is successful. If the data does not change after one execution, the control command is regenerated for line control, so as to realize the effect of real-time verification and dynamic optimization of power transmission control. This can avoid the control deviation caused by the failure of a single command execution, further improve the accuracy of power transmission data control, and ensure the reliability of the overall operation of the smart grid. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the control method steps of a power transmission data control method and system for a smart grid according to the present invention.

[0057] Figure 2 This is a system structure block diagram of a power transmission data control method and system for a smart grid according to the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1

[0060] Please see Figure 1 This invention provides a method and system for controlling power transmission data in a smart grid. To achieve the above objectives, this invention employs the following technical solution, comprising the following steps:

[0061] S1. Obtain the regional map of the transmission line. The region based on the transmission line is divided into grid areas according to function. Each grid area is equipped with an edge control node. The edge control node collects the transmission line data of the transmission line, processes the transmission line data, and obtains the transmission standard data.

[0062] S2. The edge control node divides the power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data. The power transmission standard data of emergency power transmission standard data is generated by the edge control node using a multi-parameter control model to generate control commands, which are directly transmitted to the execution terminal for execution control.

[0063] S3. For conventional transmission standard data, the edge control node generates a request command and transmits the request command and transmission standard data to the control center. The control center sends back the operation command to the corresponding edge control node, and then transmits the control command to the execution terminal to execute the operation command. For statistical transmission standard data, the transmission standard data is transmitted to the control center in batches at preset intervals for storage and analysis, and a prediction model for the corresponding transmission line is generated.

[0064] S4. When the execution terminal receives a control command, it completes the operation of the corresponding control command and feeds back the result of the operation to the corresponding edge control node. The edge control node compares the power transmission standard data before and after the operation. If the comparison of the two power transmission standard data reaches the adjustment preset standard, the power transmission control is completed. If the comparison of the two power transmission standard data does not reach the adjustment preset standard, the control command is regenerated to perform power transmission control of the power transmission line.

[0065] In this embodiment, specifically taking a 500kV transmission line as an example, the area of ​​the transmission line is divided into 12 grid areas according to function. An edge control node device is deployed in each grid area. The edge control node collects data on line current, voltage, temperature and power factor in real time. After processing, the transmission standard data is obtained. When the temperature of the transmission line in the corresponding grid area reaches 90℃, exceeding the safety threshold (the safety threshold is preset to 80℃), it is determined to be emergency transmission standard data. The edge control node calls the multi-parameter control model to generate a control command for the circuit breaker to reduce the load by 10%. This command is transmitted to the execution terminal (such as the intelligent frame universal circuit breaker BKW5-2000L / 3). After the execution terminal verifies that the command is correct, the load adjustment is completed in a short time. Then, the execution success command is fed back to the edge control node. If the execution success command is such that the current line temperature drops to 76℃, the edge control node compares the data before and after the adjustment, confirms that the standard is met, and completes an emergency control closed loop.

[0066] If the transmission standard data is within the safety threshold but the power factor is ≤0.92, it is determined to be conventional transmission standard data. The edge control node generates a request command and transmits the request command and transmission standard data to the remote control center. After the control center optimizes the command and sends it back, the edge control node forwards it to the execution terminal for execution. If the data is a historical operation record within a 24-hour period, it is determined to be statistical transmission standard data. It is temporarily stored in batches at 1-hour intervals and transmitted to the control center during periods of bandwidth idle time.

[0067] The edge control node retrieves power transmission standard data before and after the terminal operation, and performs a comprehensive comparison from multiple dimensions such as current, voltage, temperature and power factor. If the adjusted data meets the preset standard (e.g., the temperature drops to 76℃≤80℃), a power transmission control closed loop is completed. If the standard is not met (e.g., the temperature is still higher than 80℃), the control command is regenerated, and the above command transmission and execution process is repeated until the data meets the adjustment requirements, ensuring the stable operation of the power transmission line.

[0068] Example 2

[0069] Specifically: the method for constructing a multi-parameter control model, including the following steps:

[0070] Fault and normal operation data of the corresponding transmission lines were collected over the past three years. The operation data included line current, line voltage, line temperature and power factor. After preprocessing to remove outliers, the operation data was standardized to form the original dataset.

[0071] The original dataset was divided into training and test sets in a 7:3 ratio. A basic model was built based on a backpropagation neural network. The basic model includes an input layer, a hidden layer, and an output layer. The basic model was trained using the training set, and the weight parameters were iteratively optimized. Training of the basic model was stopped when the error of the training set was less than or equal to the corresponding threshold.

[0072] The test set is input into the trained base model to verify the accuracy of the control command output. If the error is greater than the preset threshold, the input layer, hidden layer and output layer of the base module are readjusted until the accuracy of the control command output meets the standard, and a multi-parameter control model is obtained.

[0073] The qualified multi-parameter control model is fixedly distributed in each edge control node, and a parameter update interface for the multi-parameter control model is set in the edge control node. The multi-parameter control model supports periodic iterative optimization based on running data through the update interface.

[0074] The basic model is built based on a backpropagation (BP) neural network. The specific construction methods for the basic model, including the input layer, hidden layer, and output layer, are as follows:

[0075] The number and function of neurons in the input layer, hidden layer, and output layer are determined to match the application scenario requirements, specifically as follows:

[0076] From the input layer to the hidden layer, the output of the hidden layer neurons is:

[0077]

[0078] in, For input parameters, For the input layer The first neuron is connected to the hidden layer. The connection weights of each neuron For the hidden layer Bias of each neuron The activation function used for the hidden layer;

[0079] From the hidden layer to the output layer, the predicted value of the output layer is:

[0080]

[0081] in, For the output layer The predicted value of each neuron, For the hidden layer The output value of each neuron, that is, the output result of the hidden layer neurons mentioned above. For the hidden layer The nth neuron to the output layer The connection weights of each neuron For the output layer Bias of each neuron The activation function used for the output layer.

[0082] The training set is used to train the basic model, and the weight optimization is performed iteratively to optimize the weight parameters. Specifically, the weight optimization involves:

[0083] The output layer weights are optimized as follows:

[0084]

[0085] in, For the hidden layer The nth neuron to the output layer The weight update amount of each neuron, if A positive value indicates that the current weight is too small and needs to be increased. A negative value indicates that the current weight is too high and needs to be reduced. For learning rate, For the output layer The true target value of each neuron. For the output layer The predicted value of each neuron. Activation function for output layer The derivative, For the hidden layer The output value of each neuron;

[0086] The input layer weights are optimized as follows:

[0087]

[0088] in, For the input layer The first neuron is connected to the hidden layer. The weight update amount of each neuron. For the hidden layer The nth neuron to the output layer The weights of each neuron, This is an approximation of the derivative of the ReLU activation function in the hidden layer. For the input layer Input data for each neuron;

[0089] In this embodiment, the number of neurons in the input layer is set to 4, corresponding to the four input parameters of the transmission line: current, voltage, temperature, and power factor. Each neuron is responsible for receiving standard transmission data for one type of parameter. There are 2 hidden layers, each with 12 neurons. The first hidden layer is responsible for extracting the basic features of the data (such as the linear relationship between current and temperature), and the second hidden layer is responsible for extracting higher-order features (such as the coordinated change law of current, temperature, and power factor). The number of neurons in the output layer is set to 2, corresponding to the control command type (such as load adjustment and reactive power compensation) and the control parameter value (such as transmission amplitude adjustment and compensation capacity adjustment). The ReLU function is used in the hidden layer to solve the gradient vanishing problem, and the Sigmoid function is used in the output layer.

[0090] During the training process of the multi-parameter control model, the mean squared error (MSE) of the training set is calculated after each iteration. Training stops when MSE ≤ 2%. If MSE > 2%, the number of hidden layer neurons or the learning rate is adjusted, and training is restarted. When validating the multi-parameter control model, the test set is input into the model, and the error between the predicted command and the actual command is calculated. If the error > 3%, the grid structure of the multi-parameter control model is adjusted until the error reaches the target. After the multi-parameter control model is deployed, the edge control nodes obtain new operating data from the control center every month and fine-tune the model weights through the parameter update interface to ensure that the multi-parameter control model adapts to the load changes of the transmission line and improves the stability and accuracy of the multi-parameter control model.

[0091] Example 3

[0092] Specifically: The data transmission and interaction between the edge control nodes and the control center for both conventional and statistical transmission standard data is encrypted.

[0093] The control center generates the root key for the national cryptographic SM4 algorithm and the SM9 identifier key. The root key for the SM4 algorithm is distributed to each edge control node through an offline secure channel. The public key of the SM9 identifier key is deployed to the edge control nodes, while the private key of the SM9 identifier key is stored in the control center. The edge control nodes generate local session keys based on the root key.

[0094] When transmitting request commands, control commands, and power transmission standard data between the edge control node and the control center, the content is encrypted using a local session key to generate first encrypted data. This first encrypted data is transmitted through the interaction gateway and then further encrypted using the AES-256 algorithm to generate second encrypted data, which is then transmitted to the control center.

[0095] After receiving the second encrypted data, the control center first decrypts it using the AES-256 algorithm to obtain the first encrypted data, and then decrypts the first encrypted data using the local session key of the corresponding edge control node to obtain the original data.

[0096] After the control center generates control commands, it encrypts the raw data with the private key of the SM9 identifier key to generate encrypted data for transmission back.

[0097] The encrypted data is transmitted back to the edge control node via an encrypted link. The edge control node uses the public key of the SM9 identifier key to decrypt the encrypted data back and obtain control commands.

[0098] The specific criteria for judging emergency transmission standard data, routine transmission standard data, and statistical transmission standard data are as follows:

[0099] When the actual value corresponding to the transmission standard data exceeds the safety threshold, it is determined to be emergency transmission standard data. When the actual value corresponding to the transmission standard data is within the safety threshold and the power factor is less than the threshold, it is determined to be regular transmission standard data. When the transmission standard data is 24-hour periodic historical data used for transmission line operation and maintenance, it is determined to be statistical transmission standard data. Furthermore, the transmission priority of emergency transmission standard data is higher than that of regular transmission standard data, and the transmission priority of regular transmission standard data is higher than that of statistical transmission standard data.

[0100] The steps for transmitting statistical transmission standard data in batches to the control center for storage and analysis at preset intervals, and generating corresponding prediction models for transmission lines, include:

[0101] The edge control nodes transmit the statistical transmission standard data in batches to the control center at preset intervals.

[0102] The control center processes the transmission standard data and extracts the daily / weekly / monthly load peak characteristics to form a feature dataset. The feature dataset is then divided into a training set and a validation set in an 8:2 ratio.

[0103] A prediction model is built based on an LSTM network. The prediction model uses the root mean square error (RMSE) as the loss function and is trained using the Adam optimizer gradient descent algorithm until the RMSE on the validation set is less than a preset function threshold.

[0104] The control center calls the prediction model daily to predict the operating parameters of the transmission lines for the corresponding time in the future. When the predicted value of the operating parameters is higher than the preset safety threshold, an early warning information is pushed. The prediction model is trained monthly with new operating data increments to adapt to the load changes of the transmission lines.

[0105] In this embodiment, the classification of power transmission standard data adopts a priority-based processing principle. First, it is determined whether the data exceeds a safety threshold (e.g., current ≥ 110% of rated value, temperature ≥ 80℃). If it does, it is directly classified as emergency power transmission standard data. If it does not exceed the threshold, the power factor is then determined to be less than 0.92. If it is less than 0.92, it is classified as regular power transmission standard data. If the power factor is greater than 0.92, it is classified as emergency power transmission standard data. If the power transmission standard data is neither emergency nor regular power transmission standard data, and the data belongs to the historical record within a 24-hour period (e.g., the average current per hour of the previous day), it is classified as statistical power transmission standard data. During transmission, emergency power transmission standard data occupies more than 60% of the transmission bandwidth of the edge control node, regular power transmission standard data occupies 30% to 40% of the bandwidth, and statistical power transmission standard data is only transmitted in batches at 1-hour intervals when the bandwidth is idle (e.g., during the low electricity consumption period from 2 to 4 am) to avoid resource contention.

[0106] After receiving statistical transmission standard data, the control center first cleans the data (removing missing and duplicate values), and then extracts features such as daily load peak (e.g., maximum current from 14:00 to 16:00 daily), weekly load peak (e.g., maximum power from 8:00 to 10:00 every Monday), and monthly load peak (e.g., industrial load peak in the last week of each month) through time series analysis to form a feature dataset. A prediction model is built and trained using an LSTM network, employing the Adam optimizer (adaptively adjusting the learning rate to avoid the tedious manual parameter tuning). Training stops when the RMSE of the validation set is ≤5%. The control center uses the model to predict the operating parameters for the 24 hours of the day. If the predicted line temperature will reach 85℃ during a certain period, a warning message (including the warning period, risk parameters, and suggested measures) is pushed to the maintenance personnel 2 hours in advance. The model is incrementally trained monthly using new statistical transmission standard data to improve the prediction accuracy of the model.

[0107] Example 4

[0108] Specifically: refer to Figure 2 A power transmission data control system for a smart grid, applied to a power transmission data control method for a smart grid, the power transmission data control system includes: an edge control node module, an execution terminal module, a data type classification module and a data result comparison module;

[0109] Edge control node modules are deployed in each functionally divided grid area to perform data processing, data classification, data transmission, and command execution for the transmission lines in each grid area.

[0110] The execution terminal module is used to receive control commands issued by the edge control node module, execute the corresponding control operations of the transmission line, and feed back the results of the control operations to the edge control node module.

[0111] The data type classification module, integrated within the edge control node module, receives the power transmission standard data output by the edge control node module, classifies the power transmission standard data into emergency power transmission standard data, routine power transmission standard data, and statistical power transmission standard data, and feeds the classified data back to the edge control node module.

[0112] The data result comparison module, integrated within the edge control node module, is used to receive the results of the control operations performed by the execution terminal module, retrieve the power transmission standard data of the edge control node module before and after the control operation, compare them according to the preset adjustment standard, and feed the comparison results back to the edge control node module to confirm the effectiveness of the control operations performed by the execution terminal module.

[0113] The power transmission data control system also includes: a data interaction module and a data acquisition module;

[0114] The data interaction module is used for the transmission of request and control commands for conventional power transmission standard data between the edge control node module and the control center, and for the edge control node module to transmit statistical power transmission standard data to the control center. It is also used for data transmission between the edge control node module and the execution terminal module.

[0115] The data acquisition module is deployed on the transmission lines in each grid area to collect transmission line data such as current, voltage, line temperature, and power factor, and transmit the collected transmission line data to the edge control node module.

[0116] The data acquisition module includes current transformers, voltage transformers, and fiber optic temperature sensors. The current transformers, voltage transformers, and fiber optic temperature sensors are distributed and installed on the transmission towers of each transmission line and are connected to the edge control node module through a general data communication interface.

[0117] In this embodiment, specifically, the data acquisition module transmits the acquired raw transmission line data to the edge control node module. After processing the transmission standard data, the edge control node module sends it to the data type classification module. The data type classification module completes the classification and feeds back the results to the edge control node module. Based on the classification results, the edge control node module directly generates control commands for emergency transmission standard data and transmits them to the execution terminal module for execution via the data interaction module. For regular transmission standard data, it generates request commands and transmits them to the control center via the data interaction module. The control center then classifies the request commands and the transmission standard data they contain. The system analyzes and judges the data, then generates corresponding control commands or does not require issuing control commands. For statistical transmission standard data, the data is transmitted to the control center for temporary storage through the data interaction module. Then, the statistical transmission standard data is used to simulate and predict the transmission lines of the power grid, and a continuously optimized and updated prediction model is generated to predict the subsequent operation data and status of the power grid's transmission lines. After receiving the control command, the execution terminal module executes the operation, and then feeds back the results to the edge control node module through the data interaction module. The edge control node module calls the data result comparison module to compare the transmission line data before and after the operation and generates the final control data.

[0118] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for controlling power transmission data in a smart grid, characterized in that: Includes the following steps: S1. Obtain a regional map of the transmission line. Based on the region of the transmission line, divide it into grid regions according to function. Each grid region is equipped with an edge control node. The edge control node collects the transmission line data of the transmission line and processes the transmission line data to obtain transmission standard data. S2. The edge control node classifies the power transmission standard data into emergency power transmission standard data, regular power transmission standard data, and statistical power transmission standard data. For the emergency power transmission standard data, the edge control node generates control commands using a multi-parameter control model. These control commands are directly transmitted to the execution terminal for execution control. The specific criteria for determining the emergency, regular, and statistical power transmission standard data are as follows: When the actual value corresponding to the transmission standard data exceeds the safety threshold, it is determined to be emergency transmission standard data. When the actual value corresponding to the transmission standard data is within the safety threshold and the power factor is less than the threshold, it is determined to be regular transmission standard data. When the transmission standard data is 24-hour periodic historical data used for the operation and maintenance of the transmission line, it is determined to be statistical transmission standard data. The transmission priority of emergency transmission standard data is higher than that of regular transmission standard data, and the transmission priority of regular transmission standard data is higher than that of statistical transmission standard data. S3. For conventional transmission standard data, the edge control node generates a request command and transmits the request command and the transmission standard data to the control center. The control center sends back the operation command to the corresponding edge control node, and then transmits it to the execution terminal to execute the operation command. For statistical transmission standard data, the transmission standard data is transmitted in batches to the control center for storage and analysis at preset intervals, and a prediction model corresponding to the transmission line is generated. S4. When the execution terminal receives the control command, the execution terminal completes the operation corresponding to the control command and feeds back the result of the operation to the corresponding edge control node. The edge control node compares the power transmission standard data before and after the operation. If the comparison of the two power transmission standard data reaches the adjustment preset standard, a power transmission control is completed. If the comparison of the two power transmission standard data does not reach the adjustment preset standard, the control command is regenerated to perform power transmission control of the power transmission line.

2. The power transmission data control method for a smart grid according to claim 1, characterized in that: The method for constructing the multi-parameter control model includes the following specific steps: Fault and normal operation data for the corresponding transmission line over the past three years are collected. The operation data includes line current, line voltage, line temperature, and power factor. The operation data is preprocessed to remove outliers and then standardized to form the original dataset. The original dataset is divided into a training set and a test set in a 7:3 ratio. A basic model is built based on a BP neural network. The basic model includes an input layer, a hidden layer, and an output layer. The basic model is trained using the training set, and the weight parameters are iteratively optimized. The training of the basic model is stopped when the training set error is less than or equal to the corresponding threshold. The test set is input into the trained base model to verify the accuracy of the control command output. If the error is greater than a preset threshold, the input layer, hidden layer and output layer of the base model are readjusted until the accuracy of the control command output meets the standard, thus obtaining the multi-parameter control model. The qualified multi-parameter control model is fixedly distributed on each edge control node, and a parameter update interface for the multi-parameter control model is set on each edge control node. The multi-parameter control model supports periodic iterative optimization based on the running data through the update interface.

3. The power transmission data control method for a smart grid according to claim 2, characterized in that: The basic model is built based on a backpropagation (BP) neural network. The specific construction method of this basic model, including its input layer, hidden layer, and output layer, includes: The number and function of neurons in the input layer, hidden layer, and output layer are determined to match the application scenario requirements, specifically as follows: The input layer to the hidden layer, and the output of the hidden layer neurons are: ； in, For input parameters, For the input layer The first neuron is connected to the hidden layer. The connection weights of each neuron For the hidden layer number Bias of each neuron The activation function used in the hidden layer is n, which is the number of neurons in the input layer. The number of neurons in the input layer is set to 4, which correspond to the four input parameters of the transmission line: current, voltage, temperature and power factor. m is the total number of neurons in the hidden layer. There are 2 hidden layers with 12 neurons in each layer. From the hidden layer to the output layer, the predicted value of the output layer is: ； in, For the output layer The predicted value of each neuron. For the hidden layer number The output value of each neuron, that is, the output result of the hidden layer neurons mentioned above. For the hidden layer number The nth neuron to the output layer The connection weights of each neuron For the output layer Bias of each neuron The activation function used for the output layer.

4. The power transmission data control method for a smart grid according to claim 2, characterized in that: The training set is used to train the base model, and the weight optimization, specifically the iterative optimization of the weight parameters, is as follows: The output layer weights are optimized as follows: ； in, For the hidden layer number The first neuron to the output layer The weight update amount of each neuron, if A positive value indicates that the current weight is too small and needs to be increased. A negative value indicates that the current weight is too high and needs to be reduced. For learning rate, For the output layer The true target value of each neuron. For the output layer The predicted value of each neuron. The output layer activation function The derivative, For the hidden layer number The output value of each neuron; The input layer weights are optimized as follows: ； in, For the input layer The first neuron is connected to the hidden layer. The weight update amount of each neuron. For the hidden layer number The first neuron to the output layer The weights of each neuron, This is an approximation of the derivative of the ReLU activation function in the hidden layer. For the input layer The input data of each neuron is p, which is the number of neurons in the output layer. The number of neurons in the output layer is set to 2, which correspond to the control command type and the control parameter value, respectively.

5. The power transmission data control method for a smart grid according to claim 1, characterized in that: The data transmission and interaction between the edge control nodes and the control center for both conventional and statistical transmission standard data is encrypted. Specifically: The control center generates a root key for the national cryptographic SM4 algorithm and an SM9 identifier key. It distributes the root key of the SM4 algorithm to each edge control node through an offline secure channel, deploys the public key of the SM9 identifier key to the edge control nodes, and stores the private key of the SM9 identifier key in the control center. The edge control nodes generate local session keys based on the root key. When the edge control node and the control center transmit the request command, the control command, and the power transmission standard data, the content is encrypted with the local session key to generate first encrypted data. The first encrypted data is transmitted through the interaction gateway and then encrypted again using the AES-256 algorithm to generate second encrypted data, which is then transmitted to the control center. After receiving the second encrypted data, the control center first decrypts it using the AES-256 algorithm to obtain the first encrypted data, and then decrypts the first encrypted data using the local session key corresponding to the edge control node to obtain the original data. After the control center generates the control command, it encrypts the original data using the private key of the SM9 identifier key to generate encrypted data for transmission back. The encrypted data is transmitted to the edge control node via an encrypted link. The edge control node uses the public key of the SM9 identifier key to decrypt the encrypted data to obtain the control command.

6. The power transmission data control method for a smart grid according to claim 1, characterized in that: The steps for transmitting statistical transmission standard data in batches to the control center for storage and analysis at preset intervals, and generating prediction models corresponding to the transmission lines, include: The edge control node transmits the statistical transmission standard data in batches to the control center at preset intervals. The control center processes the power transmission standard data and extracts the daily / weekly / monthly load peak characteristics of the power transmission standard data to form a feature dataset. The feature dataset is divided into a training set and a validation set in an 8:2 ratio. The prediction model is constructed based on an LSTM network. The prediction model uses the root mean square error (RMSE) as the loss function. The prediction model is trained using the Adam optimizer with gradient descent optimization algorithm until the RMSE of the validation set is less than a preset function threshold. The control center calls the prediction model daily to predict the operating parameters of the transmission line at the corresponding time in the future. When the predicted value of the operating parameter is higher than the preset safety threshold, an early warning information is pushed. The prediction model is trained monthly with new operating data increments to adapt to the load changes of the transmission line.

7. A power transmission data control system for a smart grid, applied to the power transmission data control method for a smart grid as described in any one of claims 1-6, characterized in that: The power transmission data control system includes: an edge control node module, an execution terminal module, a data type classification module, and a data result comparison module; The edge control node module is deployed in each of the functionally divided grid areas and is used to perform data processing, data classification, data transmission and command execution on the transmission lines in each of the grid areas. The execution terminal module is used to receive the control command issued by the edge control node module, execute the corresponding control operation of the transmission line, and feed back the result of the control operation to the edge control node module; The data type classification module is integrated into the edge control node module. It is used to receive the power transmission standard data output by the edge control node module, classify the power transmission standard data into emergency power transmission standard data, regular power transmission standard data and statistical power transmission standard data, and feed the classification data back to the edge control node module. The data result comparison module, integrated within the edge control node module, is used to receive the results of the control operation from the execution terminal module, retrieve the power transmission standard data of the edge control node module before and after the control operation, compare them according to a preset adjustment standard, and feed the comparison results back to the edge control node module to confirm the validity of the control operation from the execution terminal module.

8. The power transmission data control system for a smart grid according to claim 7, characterized in that: The power transmission data control system further includes: a data interaction module and a data acquisition module; The data interaction module is used for transmitting request instructions and control instructions for conventional power transmission standard data between the edge control node module and the control center, and for transmitting power transmission standard data for statistical power transmission standard data from the edge control node module to the control center. It is also used for data transmission between the edge control node module and the execution terminal module. The data acquisition module is deployed on the transmission lines in each of the grid areas to collect data on the current, voltage, line temperature, and power factor of the transmission lines, and transmit the collected data to the edge control node module.

9. A power transmission data control system for a smart grid according to claim 8, characterized in that: The data acquisition module includes a current transformer, a voltage transformer, and a fiber optic temperature sensor. The current transformer, voltage transformer, and fiber optic temperature sensor are distributed and installed on the transmission towers of each transmission line and are connected to the edge control node module through a general data communication interface.

Images