Power transmission section line overload correction method, system and device and storage medium

Abstract

The invention discloses a power transmission section line overload correction method, system and device and a storage medium, and the method comprises the steps: obtaining historical generator output information and node load data of a power grid under different operation conditions, and synchronously recording the active power of all lines in a key power transmission section under corresponding conditions; constructing a data set for target model training; training a target model according to the data set, and calculating the sensitivity of the active output of each generator to the active power of each line in the key power transmission section based on the trained target model; and establishing a correction model in combination with the power sensitivity and the pre-calculated section transfer coefficient, and solving the correction model to obtain an optimal generator output adjustment scheme for eliminating the line overload risk. The circuit overload problem of the power transmission section can be effectively identified and corrected under the complex power grid operation condition, the stability and reliability of power grid operation are improved, the power failure risk caused by the circuit overload is reduced, and meanwhile the power grid operation strategy is optimized.

Description

Technical Field

[0001] This invention relates to the technical field of power transmission line safety testing, and in particular to a method, system, equipment, and storage medium for overload correction of power transmission line sections. Background Technology

[0002] With the continuous expansion and increasing complexity of power systems, the safe operation of transmission lines faces numerous challenges. In particular, when some lines are disconnected or shut down, transmission line overload becomes a significant factor affecting the stable operation of the power system. Traditional overload correction methods typically rely on a single correction strategy or a fixed model, making it difficult to adapt to the complex and ever-changing operating conditions of the power grid. Furthermore, existing methods suffer from low computational efficiency, insufficient correction accuracy, and poor adaptability when dealing with large-scale power grids and various fault scenarios.

[0003] Existing research has the following shortcomings: 1. It lacks a method that can flexibly select the optimal correction strategy based on different fault scenarios and operating conditions; 2. Traditional correction models usually ignore the dynamic changes in the power grid's operating state and fail to fully utilize the real-time information of the power grid; 3. When facing complex power grid structures and multiple fault combinations, the solution efficiency and correction accuracy of existing methods are difficult to meet practical needs. Therefore, there is an urgent need for a transmission line overload correction method that systematically considers the optimal correction strategy under different fault scenarios, balancing correction accuracy and computational efficiency. Summary of the Invention

[0004] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a method, system, device, and storage medium for overload correction of transmission line sections, addressing the problems of poor adaptability, low computational efficiency, and insufficient accuracy of traditional transmission line overload correction methods under complex power grid operating conditions.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for overload correction of power transmission line sections, comprising: Acquire historical generator output information and node load data of the power grid under different operating conditions, and simultaneously record the active power of all lines in the key transmission sections under the corresponding operating conditions; Using the generator output and node load as input features and the line active power as output labels, a dataset for training the target model is constructed. The target model is trained based on the dataset, and the sensitivity of the active power output of each generator to the active power of each line in the key transmission section is calculated based on the trained target model. A correction model is established by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and the correction model is solved to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

[0006] As a preferred embodiment of the overload correction method for transmission line sections according to the present invention, the step of training the target model based on the dataset includes: Establish an end-to-end nonlinear mapping relationship from system operating status to line power at cross sections; Test the prediction accuracy of the target model, and adjust the model parameters according to the test results until the prediction accuracy of the target model reaches the preset standard.

[0007] As a preferred embodiment of the overload correction method for transmission line sections according to the present invention, the step of establishing an end-to-end nonlinear mapping relationship from the system operating state to the power of the transmission line section includes: representing the dataset as an input matrix. and output matrix ,in The number of input features, The amount of data in the model training set. The number of output features, It is the set of real numbers; Input weights and biases are generated using a sparse autoencoder, and a nonlinear activation function is used to apply the input. Perform feature mapping to obtain the output of the feature mapping layer; The output of the feature mapping layer is used as input, and the enhanced node layer output is obtained by randomly connecting the weight matrix and bias and then calculating through a nonlinear activation function. The output of the feature mapping layer and the output of the enhancement node layer are combined to form the final model input matrix. The connection weight matrix is ​​then solved to obtain the model prediction output.

[0008] As a preferred embodiment of the overload correction method for transmission line sections according to the present invention, the following steps are included: testing the prediction accuracy of the target model and adjusting the model parameters according to the test results until the prediction accuracy of the target model reaches a preset standard: calculating the root mean square error, mean absolute percentage error, and coefficient of determination of the target model; and adjusting the number of feature nodes, the number of enhancement nodes, the incremental learning parameters, and the regularization parameters of the target model according to the evaluation results of the root mean square error, mean absolute percentage error, and coefficient of determination until the prediction accuracy of the model reaches a preset standard.

[0009] As a preferred embodiment of the overload correction method for transmission sections described in this invention, the sensitivity of the active power output of each generator to the active power of each line within the key transmission section, based on the trained target model, includes: For the The generator and the first For each line, the input feature values ​​of the generator output under the current operating state are obtained, and the trained target model is used to make a prediction to obtain the line. The ground-state active power; In the input features of the target model, the first... The active power output of the generator is increased by a positive perturbation while keeping other input characteristics unchanged; Using the target model for secondary prediction, we obtain the result at the generator. After the output changes, the line New active power; calculation of the line The change in active power; The line The change in active power divided by the increase in positive perturbation yields the generator output. The active power output of the line Sensitivity of active power.

[0010] As a preferred embodiment of the overload correction method for transmission line sections described in this invention, the pre-calculated section transfer coefficient includes: The initial active power of each line in the key section is predicted using the target model; When any line is disconnected, its power will be transferred to the remaining lines according to the cross-sectional transfer coefficient; at the same time or before the line is disconnected, the output of the selected generator is adjusted to obtain the final power of the remaining lines.

[0011] The beneficial effects of this preferred technical solution are that by pre-calculating and storing the cross-sectional transfer coefficient, there is no need to re-perform complex full-network power flow calculations when a fault occurs, which shortens the time from fault identification to risk assessment and improves online calculation efficiency and real-time performance.

[0012] As a preferred embodiment of the overload correction method for transmission section lines described in this invention, the correction model is established by combining power sensitivity and pre-calculated section transfer coefficient, wherein the correction model takes minimizing the total adjustment as the objective function and generator output constraint and line safety constraint as the constraint conditions. The correction model is expressed as follows: in, For the first The active power of the generator. For the first The maximum active power of the generator. For the first The minimum active power of the generator. and These are the upper and lower power limits for each overloaded line. This represents the remaining line power value after the components within the cross-section have been disconnected.

[0013] The beneficial effects of this preferred technical solution are that by constructing optimized correction models for different cross sections, the overload correction effect and system operating cost of the line are comprehensively considered; and by using a condition selection mechanism, the rapid selection and application of efficient correction strategies are achieved, thereby improving the accuracy and efficiency of overload correction.

[0014] Secondly, the present invention provides a power transmission section line overload correction system, comprising: The data acquisition module is used to acquire historical generator output information and node load data of the power grid under different operating conditions, and simultaneously record the active power of all lines in the key transmission section under the corresponding operating conditions. The dataset construction module is used to construct a dataset for training the target model, using the generator output and node load as input features and the line active power as output labels. The sensitivity calculation module is used to train the target model based on the dataset, and calculate the sensitivity of the active power output of each generator to the active power of each line in the key transmission section based on the trained target model. The correction module is used to establish a correction model by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and solve the correction model to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

[0015] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the overload correction method for the power transmission section.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the transmission section line overload correction method.

[0017] Compared with existing technologies, the beneficial effects of this invention are as follows: First, this invention establishes a conditional selection model based on the power grid operating status and fault scenarios, and dynamically selects the optimal target model; it constructs optimized correction models for different cross sections, comprehensively considering the overload correction effect of lines and the system operating cost; through the conditional selection mechanism, it realizes the rapid selection and application of efficient correction strategies, improves the accuracy and efficiency of overload correction, and can also effectively reduce the system operating cost. It has broad practical prospects in complex power grid operating scenarios, and can provide power grid operators with a more intelligent and reliable overload correction decision support tool, which helps to build a safe, stable, and economical new power system and promotes the development of power system operation towards intelligence and efficiency. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic flowchart of a method for overload correction of a power transmission section line according to an embodiment of the present invention; Figure 2 This is a flowchart of the overload correction process for a transmission line section using a BLS model, as described in one embodiment of the present invention. Figure 3 This is a graph showing the accuracy verification of a prediction model after training for a power transmission section line overload correction method according to an embodiment of the present invention. Figure 4 This is a topology diagram of an IEEE 30-bus system before adjustment of a transmission line overload correction method according to an embodiment of the present invention. Figure 5 This refers to the generator set adjustment amount in a power transmission section line overload correction method according to one embodiment of the present invention; Figure 6 This is a line graph showing the generator sensitivity of a power transmission section line overload correction method according to an embodiment of the present invention; Figure 7 This is a topology diagram of an IEEE 30-node system after adjustment, illustrating a transmission line overload correction method according to an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0020] Example 1, referring to Figures 1-3 As one embodiment of the present invention, this embodiment provides a method for overload correction of power transmission line sections, such as... Figure 1 As shown, it includes: S100: Acquire historical generator output information and node load data of the power grid under different operating conditions, and synchronously record the active power of all lines in the key transmission section under the corresponding operating conditions. S200: Using generator output and node load as input features and line active power as output labels, a dataset is constructed for training the target model. S300: Train the target model based on the dataset, and calculate the sensitivity of the active power output of each generator to the active power of each line in the key transmission section based on the trained target model. S400: A correction model is established by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and the correction model is solved to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

[0021] It should be noted that traditional overload correction methods typically rely on a single correction strategy or a fixed model, making it difficult to adapt to the complex and ever-changing operating conditions of power grids. Furthermore, existing methods suffer from low computational efficiency, insufficient correction accuracy, and poor adaptability when dealing with large-scale power grids and various fault scenarios. Current research lacks a method that can flexibly select the optimal correction strategy based on different fault scenarios and operating conditions; traditional correction models often ignore the dynamic changes in the power grid's operating state, failing to fully utilize real-time information; and when facing complex power grid structures and multiple fault combinations, the solution efficiency and correction accuracy of existing methods are insufficient to meet practical requirements. This invention first establishes a condition selection model based on the power grid operating status and fault scenarios, dynamically selecting the optimal target model; it then constructs optimized correction models for different cross-sections, comprehensively considering the overload correction effect of lines and the system operating cost; through the condition selection mechanism, it achieves rapid selection and application of efficient correction strategies, improving the accuracy and efficiency of overload correction, and effectively reducing system operating costs. It has broad practical prospects in complex power grid operating scenarios, providing power grid operators with a more intelligent and reliable overload correction decision support tool, contributing to the construction of a safe, stable, and economical new power system, and promoting the development of power system operation towards intelligence and efficiency.

[0022] In this embodiment of the invention, step S100 involves obtaining historical generator output information and node load data of the power grid under different operating conditions, and simultaneously recording the active power of all lines in the key transmission section under the corresponding operating conditions. This is the power grid topology corresponding to the initial state of the section of interest under the main operating mode of the power grid. The load and generator output information of historical data under this topology are collected, and the active power of the lines in the key section corresponding to the load and generator output is recorded. The main operating mode of the power grid is a typical operating scenario with many usage scenarios. The critical section consists of several lines. When a selected line is disconnected, it will cause overload of the remaining lines. The state before disconnection is called the initial state or ground state. These sections can be obtained from NK scans or historical power grid operation and are usually of concern to dispatchers. The topology of these sections before disconnecting the first line may be inconsistent. For example, section 1 may be generated under the initial topology, while section 2 may be generated with some lines already out of service. Therefore, these topologies are trained separately into models. The model predicts the initial active power of all lines within the section, including the disconnected line and the remaining lines.

[0023] In this embodiment of the invention, step S200 uses generator output and node load as input features and line active power as output labels to construct a dataset for training the target model; wherein, the input to the target model is: in, In the first Generator active power under various operating conditions In the first Generator reactive power under various operating conditions In the first Active power of load under various operating conditions In the first Reactive power of load under various operating conditions In the first The voltage amplitude set for the generator under these operating conditions; The output of the target model is: in, For the node under the k-th working condition To the node The line power.

[0024] In an optional embodiment, the target model can be a Broad Learning System (BLS), and the flowchart of the transmission line overload correction for the BLS model is shown below. Figure 2As shown, the BLS model, compared to traditional linear models, can more accurately capture the complex nonlinear relationships in power systems. In this embodiment of the invention, the sensitivity calculated based on the BLS model using the perturbation method is dynamic, reflecting the current actual operating state and is more accurate than a fixed sensitivity. The BLS model has a fast training speed, and once training is complete, prediction and sensitivity calculation are very rapid, making it suitable for online applications.

[0025] In this embodiment of the invention, step S300, training the target model based on the dataset, includes: Establish an end-to-end nonlinear mapping relationship from system operating status to line power at cross sections; Test the prediction accuracy of the target model and adjust the model parameters based on the test results until the prediction accuracy of the target model reaches the preset standard.

[0026] In this embodiment of the invention, step S300, establishing the end-to-end nonlinear mapping relationship from the system operating state to the cross-sectional line power, includes: representing the dataset as an input matrix. and output matrix ,in The number of input features, The amount of data in the model training set. The number of output features, It is the set of real numbers; Input weights and biases are generated using a sparse autoencoder, and a nonlinear activation function is used to apply the input. Perform feature mapping to obtain the output of the feature mapping layer; Specifically, the expression for the feature mapping layer is: in, For activation function, The input weights are generated by the sparse autoencoder. For the first The bias matrix of the feature nodes in the group mapping. Input to the model, The output of the feature mapping layer, For the first The output feature matrix of each mapping node group This is the output matrix of the feature mapping layer.

[0027] The output of the feature mapping layer is taken as input, and the weight matrix and bias are randomly connected. The output of the enhanced node layer is obtained by calculation through a nonlinear activation function. Specifically, the expression for enhancing the node layer is: in, For the first The output value of the group enhancement node, It is a non-linear activation function. For the first The random connection weight matrix of the group-enhanced nodes, The bias matrix, To increase the number of groups in the node layer.

[0028] The output of the feature mapping layer and the output of the enhancement node layer are combined to form the final model input matrix. The connection weight matrix is ​​then solved to obtain the model prediction output. Specifically, the expression for the model's predicted output is: in, The input matrix after model training. To solve for the connection weight matrix obtained from ridge regression.

[0029] Furthermore, the expression for the connection weight matrix is: in, Input to the model, For regularization parameters, It is an identity matrix.

[0030] Reference Figure 3 In this embodiment of the invention, step S300, which tests the prediction accuracy of the target model and adjusts the model parameters according to the test results until the prediction accuracy of the target model reaches the preset standard, includes: calculating the root mean square error, mean absolute percentage error, and coefficient of determination of the target model; and adjusting the number of feature nodes, number of augmentation nodes, incremental learning parameters, and regularization parameters of the target model according to the evaluation results of the root mean square error, mean absolute percentage error, and coefficient of determination until the prediction accuracy of the model reaches the preset standard.

[0031] It should be noted that the prediction accuracy in the embodiments of the present invention is less than 1%.

[0032] Specifically, the root mean square error (RMSE) is expressed as: in, For the true value, This is a predicted value; Mean Absolute Percentage Error (MAPE) is expressed as: Coefficient of determination Represented as: in, This is the average value.

[0033] In this embodiment of the invention, step S300, which calculates the sensitivity of the active power output of each generator to the active power of each line within the key transmission section based on the trained target model, includes: For the The generator and the first For each line, the input feature values ​​of the generator output under the current operating state are obtained, and the trained target model is used to make a prediction to obtain the line. The ground-state active power; In the input features of the target model, the first... The active power output of the generator is increased by a positive perturbation while keeping other input characteristics unchanged; Using the target model for secondary prediction, we obtain the result at the generator. After the output changes, the line New active power; calculation of the line The change in active power; Line The change in active power divided by the increase in positive perturbation yields the generator output. The active power output of the line Sensitivity of active power.

[0034] In an optional embodiment, the line power is predicted using a BLS model, and the sensitivity of the generator output to the active power of each line within the cross section is obtained by perturbing the generator's active power output. The formula for obtaining this sensitivity is as follows: in, For the line The ground state change For generator Change in active power This refers to power sensitivity.

[0035] The specific method involves adding a very small amount to the characteristic position of the BLS generator output. Calculate the predicted route at this time. Change Calculate the sensitivity factor for the two variables using the formula.

[0036] In this embodiment of the invention, the pre-calculated section transfer coefficient in step S400 includes: The initial active power of each line in the key section is predicted using the target model; When any line is disconnected, its power will be transferred to the remaining lines according to the cross-sectional transfer coefficient; at the same time or before the line is disconnected, the output of the selected generator is adjusted to obtain the final power of the remaining lines.

[0037] Furthermore, the final power of the remaining lines consists of three parts: the original ground-state power, the power transferred due to line disconnection, which is determined by the transfer coefficient and the power of the disconnected lines, and the power change caused by generator output adjustment, which is jointly determined by generator sensitivity and output adjustment amount.

[0038] Furthermore, for the lines within the section of interest, the power relationship between the cut-off lines and the remaining overloaded lines is as follows: in, This represents the initial power of the remaining lines. This represents the remaining line power value after the components within the cross-section have been disconnected. For the first The cross-sectional transfer coefficient corresponding to the cut-off line. For the first The ground-state power flow of the cut-off line.

[0039] It should be noted that the cross-sectional transfer coefficient is pre-calculated and directly saved data, which can be directly read when used.

[0040] It should be noted that establishing and solving a correction model combining the BLS model and power sensitivity involves first considering the selection of generators participating in the correction and the magnitude of their corresponding sensitivity factors; a larger absolute value of the sensitivity indicates a greater ability to influence the line's power. Simultaneously, factors such as generator reserve capacity must be considered. Ultimately, dispatchers select several suitable regulating units. The relationship between the regulation output of these generators and the initial power of the line is then as follows: ; At this point, considering the sensitivity and adjustment of the cut-off lines and the power relationship between the remaining overloaded lines, the following is true: in, For generator For the line The power sensitivity represents the impact of changes in generator output on power flow in the line.

[0041] The focus here is on regulating the un-disconnected overloaded lines L. This method still applies when disconnecting lines leads to simultaneous overload of multiple lines. A similar power relationship can be established for each overloaded line L. In the correction model, corresponding safety constraints need to be added for each overloaded line, and the transfer coefficients are different for different overloaded lines.

[0042] In this embodiment of the invention, step S400, which combines power sensitivity and pre-calculated section transfer coefficient to establish a correction model, includes: the correction model takes minimizing the total adjustment as the objective function and generator output constraint and line safety constraint as the constraint conditions. The calibration model is represented as: in, For the first The active power of the generator. For the first The maximum active power of the generator. For the first The minimum active power of the generator. and These are the upper and lower power limits for each overloaded line. This represents the remaining line power value after the components within the cross-section have been disconnected.

[0043] It should be noted that this invention reads historical load and output data under the main operating topology and records the active power values ​​of several lines on key sections. The collected data is organized into a dataset and imported into a BLS model for training, enabling prediction of active power on the lines at the section. The prediction accuracy of the BLS model is tested, and model parameters are adjusted. The switching transfer factor for several branches within the section is calculated. By predicting line power using the BLS model, the sensitivity factor of generator output to line active power is obtained using the perturbation method. A correction model combining the BLS model and power sensitivity is established and solved. This model can effectively identify and correct line overload problems at transmission sections under complex power grid operating conditions, improving the stability and reliability of power grid operation, reducing the risk of power outages due to line overload, and optimizing power grid operation strategies. It has good engineering application value and promising prospects for promotion.

[0044] Example 2: The above example is an illustrative scheme of a transmission line overload correction method. It should be noted that the technical solution of this transmission line overload correction system belongs to the same concept as the above-described transmission line overload correction method. Details not described in detail in this example can be found in the description of the above-described transmission line overload correction method.

[0045] This embodiment provides a power transmission section line overload correction system, comprising: The data acquisition module is used to acquire historical generator output information and node load data of the power grid under different operating conditions, and simultaneously record the active power of all lines in the key transmission section under the corresponding operating conditions. The dataset building module is used to construct a dataset for training the target model, using generator output and node load as input features and line active power as output labels. The sensitivity calculation module is used to train the target model based on the dataset, and calculate the sensitivity of the active power output of each generator to the active power of each line in the key transmission section based on the trained target model. The correction module is used to establish a correction model by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and solve the correction model to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

[0046] This embodiment also provides an electronic device applicable to the overload correction method for power transmission sections, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the overload correction method for power transmission sections as proposed in the above embodiments.

[0047] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the overload correction method for power transmission sections as proposed in the above embodiments.

[0048] The storage medium proposed in this embodiment and the method for overload correction of transmission line sections proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0049] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0050] Example 3, referring to Figures 4-7 This is an embodiment of the present invention, which takes the IEEE 30-node system as an example to verify the beneficial effects of the present invention.

[0051] Reference Figure 4 In this embodiment, when Line 1-2 of the IEEE 30-node system is disconnected, Line 1-3, Line 3-4, and Line 4-6 are overloaded, forming a critical section. The BLS model is trained to predict the power of the four lines in the ground state, and the solution is obtained according to the defined solution model to obtain the generator output adjustment scheme and correct the active power of the overloaded line.

[0052] When Line 1-2 is disconnected, the information for each line is shown in Table 1: Table 1 Information on each line after Line 1-2 disconnection

[0053] The generator output adjustment scheme is shown in Table 2: Table 2 Generator Output Adjustment Scheme

[0054] By combining Table 1 and Table 2 Figures 5 to 6 It can be seen that this invention successfully predicted the ground-state power and solved the generator output adjustment scheme, such as the output adjustment amounts for nodes 2, 5, 8, 11, and 13, as shown in the reference. Figure 7 After the adjustment, the line overload was effectively eliminated.

[0055] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for overload correction of a power transmission line section, characterized in that, include: Acquire historical generator output information and node load data of the power grid under different operating conditions, and simultaneously record the active power of all lines in the key transmission sections under the corresponding operating conditions; Using the generator output and node load as input features and the line active power as output labels, a dataset for training the target model is constructed. The target model is trained based on the dataset, and the sensitivity of the active power output of each generator to the active power of each line in the key transmission section is calculated based on the trained target model. A correction model is established by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and the correction model is solved to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

2. The overload correction method for transmission line sections as described in claim 1, characterized in that, The step of training the target model based on the dataset includes: Establish an end-to-end nonlinear mapping relationship from system operating status to line power at cross sections; Test the prediction accuracy of the target model, and adjust the model parameters according to the test results until the prediction accuracy of the target model reaches the preset standard.

3. The overload correction method for transmission line sections as described in claim 2, characterized in that, The establishment of the end-to-end nonlinear mapping relationship from system operating state to cross-sectional line power includes: representing the dataset as an input matrix. and output matrix ,in The number of input features, The amount of data in the model training set. The number of output features, It is the set of real numbers; Input weights and biases are generated using a sparse autoencoder, and a nonlinear activation function is used to apply the input. Perform feature mapping to obtain the output of the feature mapping layer; The output of the feature mapping layer is used as input, and the enhanced node layer output is obtained by randomly connecting the weight matrix and bias and then calculating through a nonlinear activation function. The output of the feature mapping layer and the output of the enhancement node layer are combined to form the final model input matrix. The connection weight matrix is ​​then solved to obtain the model prediction output.

4. The overload correction method for transmission line sections as described in claim 3, characterized in that, Testing the prediction accuracy of the target model and adjusting the model parameters based on the test results until the prediction accuracy of the target model reaches the preset standard includes: calculating the root mean square error, mean absolute percentage error, and coefficient of determination of the target model; and adjusting the number of feature nodes, number of augmentation nodes, incremental learning parameters, and regularization parameters of the target model based on the evaluation results of the root mean square error, mean absolute percentage error, and coefficient of determination until the model prediction accuracy reaches the preset standard.

5. The overload correction method for transmission line sections as described in claim 4, characterized in that, The sensitivity of the active power output of each generator to the active power of each line within the key transmission section is calculated based on the trained target model, including: For the The generator and the first For each line, the input feature values ​​of the generator output under the current operating state are obtained, and the trained target model is used to make a prediction to obtain the line. The ground-state active power; In the input features of the target model, the first... The active power output of the generator is increased by a positive perturbation while keeping other input characteristics unchanged; Using the target model for secondary prediction, we obtain the result at the generator. After the output changes, the line New active power; calculation of the line Changes in active power; The line The change in active power divided by the increase in positive perturbation yields the generator output. The active power output of the line Sensitivity of active power.

6. The overload correction method for transmission line sections as described in claim 5, characterized in that, The pre-calculated section transfer coefficients include: The initial active power of each line in the key section is predicted using the target model; When any line is disconnected, its power will be transferred to the remaining lines according to the cross-sectional transfer coefficient; at the same time or before the line is disconnected, the output of the selected generator is adjusted to obtain the final power of the remaining lines.

7. The overload correction method for transmission line sections as described in claim 6, characterized in that, The correction model is established by combining power sensitivity and pre-calculated section transfer coefficient. The correction model takes minimizing the total regulation as the objective function and generator output constraint and line safety constraint as the constraint conditions. The correction model is expressed as follows: in, For the first The active power of the generator. For the first The maximum active power of the generator. For the first The minimum active power of the generator. and These are the upper and lower power limits for each overloaded line. This represents the remaining line power value after the components within the cross-section have been disconnected.

8. A transmission line overload correction system, applied to the method described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire historical generator output information and node load data of the power grid under different operating conditions, and simultaneously record the active power of all lines in the key transmission section under the corresponding operating conditions. The dataset construction module is used to construct a dataset for training the target model, using the generator output and node load as input features and the line active power as output labels. The sensitivity calculation module is used to train the target model based on the dataset, and calculate the sensitivity of the active power output of each generator to the active power of each line in the key transmission section based on the trained target model. The correction module is used to establish a correction model by combining power sensitivity and pre-calculated cross-sectional transfer coefficient, and solve the correction model to obtain the optimal generator output adjustment scheme to eliminate the risk of line overload.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the overload correction method for transmission line sections according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the overload correction method for a transmission line section according to any one of claims 1 to 7.

Images