Power transmission line risk early warning method, device and equipment based on intelligent grid correction

Abstract

The invention provides a power transmission line risk early warning method, device and equipment based on intelligent grid correction. The method comprises the steps of obtaining multi-mode historical meteorological prediction data and historical meteorological observation data corresponding to a target area; based on the historical meteorological observation data, performing space-time interpolation and multi-mode fusion on the historical meteorological prediction data to obtain a plurality of meteorological element prediction values after fusion of each grid in the target area; and training a machine learning model by using the meteorological element prediction values before and after fusion and historical meteorological observation data to obtain a meteorological element correction model set, the meteorological element correction model set being used for correcting the multi-mode meteorological prediction data of the target area in severe convective weather. The meteorological element target correction value is used for carrying out power transmission line risk early warning for the target area. According to the invention, the precision and accuracy of the power transmission line risk early warning result can be significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid disaster prevention and mitigation, and in particular to a transmission line risk early warning method, device and equipment based on intelligent grid correction. Background Art

[0002] With the development of power systems and the improvement of their intelligence, the requirements for the safety and reliability of power grid operation are increasing. As an important part of the power system, the safe operation of transmission lines directly affects the stable power supply of the power grid. However, transmission lines are often affected by severe convective weather. When strong winds or heavy rains act on the wind pressure surface of conductors and insulator strings, the conductors will deflect and displace to a certain extent. When the air gap between the conductors and tower components or surrounding objects is less than the power frequency voltage breakdown distance, air breakdown occurs, resulting in wind-induced discharge tripping, which in turn causes equipment damage or even line failure, posing serious safety hazards to the power grid. Therefore, it is necessary to study risk assessment and early warning methods for transmission lines, especially to take advance protective measures for transmission lines that may experience wind-induced discharge in severe convective weather.

[0003] At present, the identification and early warning management of meteorological disaster characteristics for transmission lines can be divided into the following according to the means of data acquisition: ① Identification of severe convective weather such as hail, thunderstorms, tornadoes, etc. based on Doppler weather radar, and 0-2h near-term extrapolation of precipitation. ② Diagnosis of the location of severe convective weather such as hail, short-term heavy precipitation, thunderstorms, etc. based on numerical forecasts, and improvement of temporal and spatial resolution through downscaling.

[0004] The above methods have improved the level of refinement of severe convective weather forecasts for transmission lines to a certain extent, but they have certain limitations when dealing with complex and changeable meteorological conditions and the influence of power grid topology. The spatial range of Doppler weather radar cannot cover all transmission line areas, and the forecast time is only 2 hours; there are systematic errors in the regional downscaling numerical forecast results. Therefore, it is very necessary to provide a more accurate transmission line risk warning method under severe convective weather. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a transmission line risk warning method, device and equipment based on intelligent grid correction, which can significantly improve the precision and accuracy of transmission line risk warning results.

[0006] In a first aspect, the present invention provides a transmission line risk early warning method based on intelligent grid correction, comprising: Obtain multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area, the historical meteorological forecast data includes forecast values ​​of multiple meteorological elements under different modes, and the historical meteorological observation data includes observation values ​​of multiple meteorological elements corresponding to each grid in the target area; Based on historical meteorological observation data, the historical meteorological forecast data is temporally and spatially interpolated and multi-mode fused to obtain the forecast values ​​of multiple meteorological elements after fusion for each grid in the target area; The machine learning model is trained using the meteorological element prediction values ​​and historical meteorological observation data before and after fusion to obtain a meteorological element correction model set. The meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather. The obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each grid in the target area. The multiple meteorological element target correction values ​​are used to provide transmission line risk warning for the target area.

[0007] In one embodiment, based on historical meteorological observation data, historical meteorological forecast data is subjected to spatiotemporal interpolation and multi-mode fusion to obtain multiple meteorological element forecast values ​​after fusion for each grid in the target area, including: Perform spatiotemporal interpolation of historical meteorological forecast data under each mode; Based on historical meteorological observation data and historical meteorological forecast data after spatiotemporal interpolation, the corresponding mode fusion weights of the same meteorological element under different modes are determined; For each grid in the target area, the meteorological element prediction values ​​of the same meteorological element in the grid under different modes are fused by multi-mode using the mode fusion weight to obtain the fused meteorological element prediction value of the grid.

[0008] In one implementation, based on historical meteorological observation data and historical meteorological forecast data after spatiotemporal interpolation, the mode fusion weights corresponding to the same meteorological element in different modes are determined, including: The following processing is performed on the historical weather forecast data in each mode: The historical meteorological forecast data after spatiotemporal interpolation is normalized to obtain the normalized forecast values ​​of multiple meteorological elements for each grid in the target area under this mode; And, performing residual processing on the historical meteorological forecast data after spatiotemporal interpolation according to the historical meteorological observation data, so as to obtain residual data of multiple meteorological elements for each grid in the target area; Based on the normalized meteorological element prediction values ​​and meteorological element residual data corresponding to the same meteorological element, the mode fusion weight corresponding to the same meteorological element under this mode is determined.

[0009] In one implementation, the meteorological element forecast values ​​and historical meteorological observation data before and after fusion are used to train the machine learning model to obtain a meteorological element correction model set, including: Based on the meteorological element forecast values ​​and historical meteorological observation data before and after fusion, a preliminary sample data set is constructed; The predicted values ​​of meteorological elements before and after fusion in the preliminary sample data set are used as input, and the historical meteorological observation data are used as training labels to train the XGBoost model to obtain the target XGBoost model. The SHAP method is used to interpret the model, select key meteorological elements, and construct the target sample data set based on the predicted values ​​of meteorological elements before and after fusion and the historical meteorological observation data corresponding to the key meteorological elements. Based on the target sample data set, the meteorological element prediction values ​​before and after fusion corresponding to each key meteorological element are taken as input, and the meteorological element observation values ​​corresponding to the key meteorological elements in the historical meteorological observation data are used as training labels. The machine learning model is trained to obtain the meteorological element correction model set.

[0010] In one embodiment, the method further comprises: Obtain multi-model meteorological forecast data for the target area under severe convective weather; The multi-mode meteorological forecast data is corrected by using the meteorological element correction model set to obtain the initial meteorological correction data output by each meteorological element correction model in the meteorological element correction model set. The initial meteorological correction data includes the initial correction values ​​of multiple meteorological elements corresponding to each grid in the target area. Perform multi-model fusion on the initial meteorological correction data to obtain the target meteorological correction data; Based on the location information of the transmission lines and towers corresponding to the target area, multiple target correction values ​​of meteorological elements corresponding to the target grids to which the transmission lines and towers belong are extracted from the target meteorological correction data; According to the target correction values ​​of multiple meteorological elements corresponding to the target grid, a transmission line risk warning is carried out for the target area, and the transmission line risk warning result corresponding to the target area is obtained.

[0011] In one embodiment, the target meteorological correction data includes a wind speed correction value; according to the target correction values ​​of multiple meteorological elements corresponding to the target grid, a transmission line risk warning is performed for the target area, and a transmission line risk warning result corresponding to the target area is obtained, including: According to the wind speed correction value corresponding to the target grid, determine the wind load borne by the tower under the wind speed at the tower height in the target area; Determine the wind deflection angle of the insulator string based on the wind load on the tower and the gravity load on the transmission line, and determine the gap distance between the transmission line and the tower according to the wind deflection angle of the insulator string; Based on the comparison between the gap distance and the preset gap distance threshold, the transmission line risk warning result corresponding to the target area is determined.

[0012] In one embodiment, performing multi-model fusion on initial meteorological correction data to obtain target meteorological correction data includes: For each meteorological element, the initial correction value of the meteorological element output by each meteorological element correction model is subjected to multi-model fusion according to the dynamically adjustable model fusion weight to obtain the meteorological element target correction value corresponding to the meteorological element.

[0013] In a second aspect, the present invention further provides a transmission line risk warning device based on intelligent grid correction, comprising: A data acquisition module is used to acquire multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area, the historical meteorological forecast data includes multiple meteorological element forecast values ​​under different modes, and the historical meteorological observation data includes multiple meteorological element observation values ​​corresponding to each grid in the target area; The data processing module is used to perform spatiotemporal interpolation and multi-mode fusion on the historical meteorological forecast data based on the historical meteorological observation data, so as to obtain the fused forecast values ​​of multiple meteorological elements for each grid in the target area; The model training module is used to train the machine learning model using the meteorological element prediction values ​​and historical meteorological observation data before and after fusion to obtain a meteorological element correction model set. The meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather. The obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each grid in the target area. The multiple meteorological element target correction values ​​are used to provide transmission line risk warning for the target area.

[0014] In a third aspect, the present invention further provides an electronic device, comprising a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement any one of the methods provided in the first aspect.

[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement any one of the methods provided in the first aspect.

[0016] The present invention provides a transmission line risk warning method, device and equipment based on intelligent grid correction. First, multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to a target area are obtained, the historical meteorological forecast data include multiple meteorological element prediction values ​​in different modes, and the historical meteorological observation data include multiple meteorological element observation values ​​corresponding to each grid in the target area; then, based on the historical meteorological observation data, the historical meteorological forecast data are subjected to spatiotemporal interpolation and multi-mode fusion to obtain multiple meteorological element prediction values ​​after fusion for each grid in the target area; finally, the meteorological element prediction values ​​before and after fusion and the historical meteorological observation data are used to train a machine learning model to obtain a meteorological element correction model set, the meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather, the obtained target meteorological correction data include multiple meteorological element target correction values ​​corresponding to each grid in the target area, and the multiple meteorological element target correction values ​​are used to perform transmission line risk warning for the target area. The above method uses historical meteorological observation data to perform spatiotemporal interpolation and multi-mode fusion on historical meteorological forecast data, and on this basis, trains the machine learning model to obtain a meteorological element correction model set. The meteorological element correction model set can effectively correct the systematic deviations and local errors in the multi-mode meteorological forecast data, thereby improving the accuracy of meteorological forecasts. In addition, the meteorological element correction model set can provide grid forecast products with high spatiotemporal resolution and long forecast time, achieving seamless coverage from short-term to extended periods, and providing continuous and high-resolution meteorological information for transmission line risk warning, which helps to improve the precision and accuracy of transmission line risk warning results.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of a process flow of a transmission line risk early warning method based on intelligent grid correction provided by an embodiment of the present invention; Figure 2 An overall flow chart of a transmission line risk warning method based on intelligent grid correction provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a transmission line risk warning device based on intelligent grid correction provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] At present, the related technologies have the following problems: the spatial range of Doppler weather radar cannot cover all transmission line areas, and the forecast time is only 2 hours; the regional downscaling numerical forecast results have systematic errors. Based on this, the present invention implements a transmission line risk warning method, device and equipment based on intelligent grid correction, which can significantly improve the precision and accuracy of transmission line risk warning results.

[0023] To facilitate understanding of this embodiment, firstly, a transmission line risk warning method based on intelligent grid correction disclosed in an embodiment of the present invention is described in detail, see Figure 1 The flowchart of a transmission line risk early warning method based on intelligent grid correction is shown in FIG. 1 , and the method mainly includes the following steps S102 to S106: Step S102, obtaining multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area.

[0024] The historical meteorological forecast data includes the forecast values ​​of multiple meteorological elements in different modes, and the historical meteorological observation data includes the observation values ​​of multiple meteorological elements corresponding to each grid in the target area. For example, the multi-mode historical meteorological forecast data can be 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature in the EC mode, and 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature in the GRAPES-GFS mode; the historical meteorological observation data can be 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature in the HRCLDAS land surface fusion real-time analysis product.

[0025] Step S104, based on the historical meteorological observation data, performing spatiotemporal interpolation and multi-mode fusion on the historical meteorological forecast data to obtain a plurality of fused meteorological element forecast values ​​for each grid in the target area.

[0026] In one example, historical meteorological forecast data can be subjected to spatiotemporal interpolation to unify the spatiotemporal scales of historical meteorological forecast data of different modes, and obtain the meteorological element prediction value corresponding to each grid in the study area; based on the historical meteorological forecast data and historical meteorological observation data after spatiotemporal interpolation processing, the prediction accuracy of the same meteorological element in different modes can be determined, and the mode fusion weight can be obtained using the prediction accuracy, and then the meteorological element prediction values ​​corresponding to the same meteorological element in each grid under different modes can be subjected to multi-mode fusion, so as to obtain multiple fused meteorological element prediction values ​​corresponding to each grid.

[0027] Step S106, using the meteorological element prediction values ​​and historical meteorological observation data before and after the fusion, the machine learning model is trained to obtain a meteorological element correction model set.

[0028] In one example, the meteorological element prediction values ​​before and after fusion can be used as input, and the meteorological element observation values ​​can be used as labels to train the machine learning model to obtain a preset number of meteorological element correction models to form a meteorological element correction model set. In another example, the meteorological elements can be further screened, and the meteorological element prediction values ​​before and after fusion corresponding to the screened key meteorological elements can be used as input, and the observation values ​​corresponding to the key meteorological elements can be used as labels to train the machine learning model to obtain a preset number of meteorological element correction models to form a meteorological element correction model set.

[0029] Among them, the meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather, and the obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each grid in the target area, and multiple meteorological element target correction values ​​are used to carry out transmission line risk warning for the target area. Exemplarily, after obtaining the multi-mode meteorological forecast data of the target area under severe convective weather, the multi-mode meteorological forecast data can be temporally interpolated and multi-mode fused, and the processed data can be input into the meteorological element correction model set to obtain multiple meteorological element target correction values ​​corresponding to each grid in the target area, extract multiple meteorological element target correction values ​​corresponding to the grid where the transmission line and the tower are located, and carry out transmission line risk warning on this basis.

[0030] The transmission line risk warning method based on intelligent grid correction provided by the embodiment of the present invention uses historical meteorological observation data to perform spatiotemporal interpolation and multi-mode fusion on historical meteorological forecast data, and on this basis trains the machine learning model to obtain a meteorological element correction model set. The meteorological element correction model set can effectively correct the systematic deviations and local errors in the multi-mode meteorological forecast data, thereby improving the accuracy of meteorological forecasts. In addition, the meteorological element correction model set can provide grid forecast products with high spatiotemporal resolution and long forecast time, achieving seamless coverage from short-term to extended periods, providing continuous and high-resolution meteorological information for transmission line risk warning, and helping to improve the precision and accuracy of transmission line risk warning results.

[0031] For ease of understanding, the present invention provides a specific implementation of a transmission line risk warning method based on intelligent grid correction, see Figure 2 The overall flow chart of a transmission line risk early warning method based on intelligent grid correction is shown, including: 1. Obtain multi-mode historical meteorological forecast data and historical meteorological observation data. Specifically: (1.1) The multi-model historical meteorological forecast data can be 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature of the EC model, and 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature of the GRAPES-GFS model; (1.2) Historical meteorological observation data can provide the 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature for the HRCLDAS land surface fusion real-time analysis product.

[0032] (II) Based on historical meteorological observation data, perform spatiotemporal interpolation and multi-mode fusion on historical meteorological forecast data to obtain the fused forecast values ​​of multiple meteorological elements for each grid in the target area. For details, please refer to (2.1) to (2.3) below: (2.1) Perform spatiotemporal interpolation on the historical meteorological forecast data in each mode. In a specific implementation, the historical meteorological forecast data at a 1-hour time interval can be obtained by linear interpolation, and the historical meteorological forecast data at a 0.01°×0.01° grid spacing can be obtained by inverse distance weighted interpolation.

[0033] (2.2) Based on the historical meteorological observation data and the historical meteorological forecast data after spatiotemporal interpolation, the mode fusion weights corresponding to the same meteorological element in different modes are determined. In a specific implementation, the following processing can be performed for the historical meteorological forecast data in each mode: (a) The historical meteorological forecast data after spatiotemporal interpolation is normalized to obtain the normalized forecast values ​​of multiple meteorological elements for each grid in the target area under the mode. In one example, the maximum forecast value and the minimum forecast value corresponding to each meteorological element in the historical meteorological forecast data after spatiotemporal interpolation under the mode can be determined, and the forecast values ​​of each meteorological element are normalized to obtain the normalized forecast values ​​of multiple meteorological elements for each grid under the mode.

[0034] (b) Perform residual processing on the historical meteorological forecast data after spatiotemporal interpolation according to the historical meteorological observation data to obtain the residual data of multiple meteorological elements for each grid in the target area. Specifically, perform residual processing on the historical meteorological EC forecast data after spatiotemporal interpolation according to the historical meteorological observation data to obtain the residual data of multiple meteorological elements for each grid in the target area under the EC mode. The residual processing process is as follows: the observed value of the same meteorological element in each grid is subtracted from the predicted value under the EC mode to obtain the residual data of each meteorological element in each grid under this mode (i.e., the residual data of the meteorological element).

[0035] (c) Based on the normalized meteorological element prediction value and meteorological element residual data corresponding to the same meteorological element, determine the model fusion weight corresponding to the same meteorological element in the mode. In one example, the accuracy of the historical meteorological forecast data can be determined based on the normalized meteorological element prediction value and meteorological element residual data corresponding to the same meteorological element. For example, RMSE is used as the accuracy. By determining the RMSE of EC and GFS, the inverse of the accuracy is used as the model fusion weight corresponding to the meteorological element in the mode. The higher the accuracy, the larger the inverse, indicating that the model performance is better.

[0036] (2.3) For each grid in the target area, the meteorological element forecast values ​​of the same meteorological element in the grid under different modes are fused by using the mode fusion weight to obtain the fused meteorological element forecast value corresponding to the grid. Taking wind speed as an example, the 10-meter wind speed of the EC mode and the 10-meter wind speed of the GRAPES-GFS mode can be weighted summed according to the EC mode fusion weight and the GRAPES-GFS mode fusion weight corresponding to the wind speed to obtain the fused 10-meter wind speed forecast value corresponding to each grid.

[0037] (III) Using the meteorological element forecast values ​​and historical meteorological observation data before and after fusion, the machine learning model is trained to obtain a meteorological element correction model set. For details, please refer to (3.1) to (3.3) as follows: (3.1) Based on the meteorological element prediction values ​​before and after fusion and the historical meteorological observation data, a preliminary sample data set is constructed. The meteorological element prediction values ​​before and after fusion are the model input, and the historical meteorological observation data are the training labels.

[0038] (3.2) The predicted values ​​of meteorological elements before and after fusion of the preliminary sample data set are used as input, and the historical meteorological observation data are used as training labels to train the XGBoost model to obtain the target XGBoost model. The SHAP method is used to interpret the model, and the key meteorological elements that have a greater contribution to the historical meteorological observation data are screened (for example, the contribution value (SHAP value) is used to screen the top ten meteorological elements as key meteorological elements), and the target sample data set is constructed based on the predicted values ​​of meteorological elements before and after fusion corresponding to the key meteorological elements and the historical meteorological observation data as the target sample data set for machine learning model training. For example, the 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature of the EC model, the 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature of the GRAPES-GFS model, and the 10-meter wind direction, 10-meter wind speed, 2-meter relative humidity, and 2-meter temperature predicted values ​​after the fusion of the EC model and the GRAPES-GFS model are used as input, and the 10-meter wind speed of the historical meteorological observation data is used as the training label to train the XGBoost model to obtain the target XGBoost model. The SHAP method is used to interpret the model, and the key meteorological elements that have a greater contribution to the 10-meter wind speed of historical meteorological observation data are screened out from the above meteorological elements. For example, the 10-meter wind direction and 10-meter wind speed of the EC model, the 10-meter wind direction and 10-meter wind speed of the GRAPES-GFS model, and the 10-meter wind direction and 10-meter wind speed after the fusion of the EC model and the GRAPES-GFS model can be used as key meteorological elements.

[0039] (3.3) Based on the target sample data set, the meteorological element prediction values ​​before and after fusion corresponding to each key meteorological element are used as input, and the meteorological element observation values ​​corresponding to the key meteorological elements in the historical meteorological observation data are used as training labels to train the machine learning model to obtain the meteorological element correction model set. Please continue to refer to the above example, the 10-meter wind direction and 10-meter wind speed of the EC model, the 10-meter wind direction and 10-meter wind speed of the GRAPES-GFS model, and the 10-meter wind direction and 10-meter wind speed prediction values ​​after the fusion of the EC model and the GRAPES-GFS model can be used as inputs of the machine learning model, and the 10-meter wind speed observation values ​​are used as training labels to train and interpret the machine learning model, and a preset number of meteorological element correction models are obtained through cross-validation. The preset number of meteorological element correction models will constitute the meteorological element correction model set.

[0040] (IV) Obtain multi-mode meteorological forecast data for the transmission line area where severe convection will occur in the future, and obtain the ground 10m wind speed correction value in the transmission line area under severe convective weather through the meteorological element correction model set. For details, please refer to (4.1) to (4.3): (4.1) Obtain multi-model meteorological forecast data for the target area under severe convective weather. In one example, the meteorological forecast data of the EC model and the meteorological forecast data of the GRAPES-GFS model for the transmission line area where severe convection will occur in the future can be obtained. Optionally, only the 10m wind speed forecast value of the EC model and only the 10m wind speed forecast value of the GRAPES-GFS model can be obtained.

[0041] (4.2) The multi-mode meteorological forecast data is corrected by the meteorological element correction model set to obtain the initial meteorological correction data output by each meteorological element correction model in the meteorological element correction model set. The initial meteorological correction data includes multiple initial correction values ​​of meteorological elements corresponding to each grid in the target area. In one example, the multi-mode meteorological forecast data can be subjected to spatiotemporal interpolation processing, and the multi-mode meteorological forecast data can be subjected to multi-mode fusion based on the mode fusion weight corresponding to each meteorological element. The mode fusion weight is determined based on historical meteorological observation data and historical meteorological forecast data after spatiotemporal interpolation. For details, please refer to the above (2.2), which will not be elaborated in the embodiment of the present invention. The meteorological forecast data before and after the fusion is input into each meteorological element correction model in the meteorological element correction model set to obtain the initial meteorological correction data output by each meteorological element correction model. Optionally, each meteorological element correction model can output 10 initial correction values ​​of wind direction.

[0042] (4.3) The initial meteorological correction data is fused with multiple models to obtain the target meteorological correction data.

[0043] In one example, for each meteorological element, the initial meteorological correction data output by each meteorological element correction model for the meteorological element may be averaged to obtain the target meteorological correction data corresponding to the meteorological element.

[0044] In another example, for each meteorological element, the initial correction value of the meteorological element output by each meteorological element correction model for the meteorological element is subjected to multi-model fusion according to the dynamically adjustable model fusion weight, so as to obtain the target correction value of the meteorological element corresponding to the meteorological element. Exemplarily, the meteorological element correction model set is used as the primary model, the weight adjustment model is used as the secondary model, and the initial meteorological correction data output by each meteorological element correction model in the meteorological element correction model set for each meteorological element is input into the weight adjustment model. The weight adjustment model can output the weight value corresponding to each meteorological element correction model (that is, the model fusion weight). The initial correction value of the meteorological element output by each meteorological element correction model for the same meteorological element is fused using the weight value, so as to obtain the target correction value of the meteorological element corresponding to each meteorological element.

[0045] Optionally, the target meteorological correction data includes wind speed correction values, that is, 10 wind direction target correction values ​​can be obtained after multi-model fusion.

[0046] (V) Obtain the transmission line parameters and tower coordinates, calculate the wind deviation discharge gap distance of the transmission line according to the target correction values ​​of multiple meteorological elements corresponding to the target grid, and evaluate and publish the transmission line risk warning results. For details, please refer to (5.1) to (5.4): (5.1) Based on the transmission line and tower location information corresponding to the target area, multiple meteorological element target correction values ​​corresponding to the target grid to which the transmission line and tower belong are extracted from the target meteorological correction data. In one example, the transmission line and tower location information can describe the specific location information of the transmission line and tower through longitude and latitude, and then the location information can be used to spatially match the transmission line and tower with the grid to determine the target grid to which the transmission line and tower belong, and then the 10 wind direction target correction values ​​corresponding to the target grid are screened out from the target meteorological correction data.

[0047] (5.2) Based on the wind speed correction value corresponding to the target grid, determine the wind load on the tower under the wind speed at the tower height within the target area.

[0048] In the specific implementation, the 10 wind direction target correction value is corrected to the wind speed value at the tower height, and the wind load borne by the tower under the wind speed at the tower height is calculated: Wind load on insulator string: ; Wind load on conductor: ; in, , are the wind loads on the insulator string and conductor, respectively, in kN; is the wind pressure height variation coefficient; B is the wind load increase coefficient when covered with ice; is the wind-exposed area of ​​the insulator string, in m2; is the wind speed value at the tower height, in m / s; α is the wind pressure unevenness coefficient; is the conductor shape coefficient; is the wind load adjustment factor; is the outer diameter of the conductor, in m; Horizontal span of tower, in m; It is the angle between the wind direction the conductor is subject to and the conductor.

[0049] Consider the worst case, the wind direction and the insulator string are in the same plane, and the wind on the conductor is perpendicular to the conductor, that is, =90°, wind load , Can be abbreviated as: ; ; in, and It is related to its own parameters such as line structure and terrain.

[0050] Wind load F on insulator string: +2 ; (5.3) The wind deflection angle of the insulator string is determined based on the wind load on the tower and the gravity load on the transmission line, and the clearance distance between the transmission line and the tower is determined based on the wind deflection angle of the insulator string.

[0051] First, the insulator string force analysis is performed based on the wind load and the transmission line's own gravity load, and the insulator string wind deflection angle is calculated: ; Among them, G The combination of the string deadweight load and the conductor deadweight load acting on the end of the insulator string.

[0052] Then, based on the geometric relationship, two types of "tower-line" air gaps where windage discharge may occur are calculated, that is, the gap between the end conductor of the insulator string and the tower body. and the clearance from the crossarm

[0053] ; = ; in, , There are two kinds of “tower-line” gaps, the unit is m; , is the tower structure parameter; is the length of the insulator string, in meters.

[0054] The clearance distance between the line and the tower is: d= .

[0055] (5.4) Based on the comparison between the gap distance and the preset gap distance threshold, the transmission line risk warning result corresponding to the target area is determined.

[0056] In one embodiment, the minimum allowable discharge gap between the live parts and the tower components under different nominal power frequency voltages can be , determine whether wind-induced discharge will occur in the transmission line: ; In the event of wind-induced discharge on a transmission line, the transmission line risk can be published as shown in Table 1 below.

[0057] Table 1 Risk warning levels and warning conditions

[0058] In summary, the embodiments of the present invention have at least the following features: (I) Smart grid correction technology, through the application of deep learning models, can effectively correct systematic deviations and local errors in meteorological forecasts compared to traditional methods, thereby improving forecast accuracy.

[0059] (II) Intelligent grid correction technology can provide national grid forecast products with high spatial and temporal resolution and long forecast period, achieving seamless coverage from short-term to extended periods, and providing continuous and high-resolution meteorological information for transmission line risk warning.

[0060] (III) By combining advanced machine learning technology with the needs of power system operation and management, and by calculating the wind-induced discharge gap distance of transmission lines under severe convective weather, the risk warning of transmission lines under severe convective weather is abstracted into the prediction accuracy of corresponding meteorological elements, which improves the assessment and warning capabilities of wind-induced discharge risks of transmission lines under severe convective weather conditions. This can assist power departments in timely assessing the wind-induced discharge risks of transmission lines so as to take necessary measures to reduce the losses caused by line tripping.

[0061] Based on the above embodiments, the present invention provides a transmission line risk warning device based on intelligent grid correction, see Figure 3 The structure diagram of a transmission line risk early warning device based on intelligent grid correction is shown in FIG. The device mainly includes the following parts: A data acquisition module 302 is used to acquire multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area, the historical meteorological forecast data includes multiple meteorological element forecast values ​​under different modes, and the historical meteorological observation data includes multiple meteorological element observation values ​​corresponding to each grid in the target area; The data processing module 304 is used to perform spatiotemporal interpolation and multi-mode fusion on the historical meteorological forecast data based on the historical meteorological observation data, so as to obtain the fused forecast values ​​of multiple meteorological elements for each grid in the target area; The model training module 306 is used to train the machine learning model using the meteorological element prediction values ​​and historical meteorological observation data before and after fusion to obtain a meteorological element correction model set. The meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather. The obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each grid in the target area. The multiple meteorological element target correction values ​​are used to provide transmission line risk warning for the target area.

[0062] The transmission line risk warning device based on intelligent grid correction provided by the embodiment of the present invention uses historical meteorological observation data to perform spatiotemporal interpolation and multi-mode fusion on historical meteorological forecast data, and on this basis trains the machine learning model to obtain a meteorological element correction model set. The meteorological element correction model set can effectively correct the systematic deviations and local errors in the multi-mode meteorological forecast data, thereby improving the accuracy of meteorological forecasts. In addition, the meteorological element correction model set can provide grid forecast products with high spatiotemporal resolution and long forecast time, achieving seamless coverage from short-term to extended periods, providing continuous and high-resolution meteorological information for transmission line risk warnings, and helping to improve the precision and accuracy of transmission line risk warning results.

[0063] In one implementation, the data processing module 304 is specifically configured to: Perform spatiotemporal interpolation of historical meteorological forecast data under each mode; Based on historical meteorological observation data and historical meteorological forecast data after spatiotemporal interpolation, the corresponding mode fusion weights of the same meteorological element under different modes are determined; For each grid in the target area, the meteorological element prediction values ​​of the same meteorological element in the grid under different modes are fused by multi-mode using the mode fusion weight to obtain the fused meteorological element prediction value of the grid.

[0064] In one implementation, the data processing module 304 is specifically configured to: The following processing is performed on the historical weather forecast data in each mode: The historical meteorological forecast data after spatiotemporal interpolation is normalized to obtain the normalized forecast values ​​of multiple meteorological elements for each grid in the target area under this mode; And, performing residual processing on the historical meteorological forecast data after spatiotemporal interpolation according to the historical meteorological observation data, so as to obtain residual data of multiple meteorological elements for each grid in the target area; Based on the normalized meteorological element prediction values ​​and meteorological element residual data corresponding to the same meteorological element, the mode fusion weight corresponding to the same meteorological element under this mode is determined.

[0065] In one embodiment, the model training module 306 is specifically used to: Based on the meteorological element forecast values ​​and historical meteorological observation data before and after fusion, a preliminary sample data set is constructed; The predicted values ​​of meteorological elements before and after fusion in the preliminary sample data set are used as input, and the historical meteorological observation data are used as training labels to train the XGBoost model to obtain the target XGBoost model. The SHAP method is used to interpret the model, select key meteorological elements, and construct the target sample data set based on the predicted values ​​of meteorological elements before and after fusion and the historical meteorological observation data corresponding to the key meteorological elements. Based on the target sample data set, the meteorological element prediction values ​​before and after fusion corresponding to each key meteorological element are taken as input, and the meteorological element observation values ​​corresponding to the key meteorological elements in the historical meteorological observation data are used as training labels. The machine learning model is trained to obtain the meteorological element correction model set.

[0066] In one embodiment, a transmission line risk early warning module is also included, which is used to: Obtain multi-model meteorological forecast data for the target area under severe convective weather; The multi-mode meteorological forecast data is corrected by using the meteorological element correction model set to obtain the initial meteorological correction data output by each meteorological element correction model in the meteorological element correction model set. The initial meteorological correction data includes the initial correction values ​​of multiple meteorological elements corresponding to each grid in the target area. Perform multi-model fusion on the initial meteorological correction data to obtain the target meteorological correction data; Based on the location information of the transmission lines and towers corresponding to the target area, multiple target correction values ​​of meteorological elements corresponding to the target grids to which the transmission lines and towers belong are extracted from the target meteorological correction data; According to the target correction values ​​of multiple meteorological elements corresponding to the target grid, a transmission line risk warning is carried out for the target area, and the transmission line risk warning result corresponding to the target area is obtained.

[0067] In one embodiment, the target meteorological correction data includes a wind speed correction value; the transmission line risk warning module is specifically used for: According to the wind speed correction value corresponding to the target grid, determine the wind load borne by the tower under the wind speed at the tower height in the target area; Determine the wind deflection angle of the insulator string based on the wind load on the tower and the gravity load on the transmission line, and determine the gap distance between the transmission line and the tower according to the wind deflection angle of the insulator string; Based on the comparison between the gap distance and the preset gap distance threshold, the transmission line risk warning result corresponding to the target area is determined.

[0068] In one implementation, the transmission line risk warning module is specifically used to: For each meteorological element, the initial correction value of the meteorological element output by each meteorological element correction model is subjected to multi-model fusion according to the dynamically adjustable model fusion weight to obtain the meteorological element target correction value corresponding to the meteorological element.

[0069] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0070] An embodiment of the present invention provides an electronic device. Specifically, the electronic device includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above-mentioned implementation methods.

[0071] Figure 4 A structural diagram of an electronic device provided in an embodiment of the present invention, the electronic device 100 includes: a processor 40, a memory 41, a bus 42 and a communication interface 43, wherein the processor 40, the communication interface 43 and the memory 41 are connected via the bus 42; the processor 40 is used to execute an executable module stored in the memory 41, such as a computer program.

[0072] The memory 41 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 43 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0073] The bus 42 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0074] Among them, the memory 41 is used to store programs, and the processor 40 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0075] The processor 40 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 40. The above processor 40 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and completes the steps of the above method in combination with its hardware.

[0076] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be referred to the previous method embodiments, which will not be repeated here.

[0077] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0078] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A transmission line risk early warning method based on intelligent grid correction, characterized in that: include: Acquire multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area, wherein the historical meteorological forecast data includes multiple meteorological element forecast values ​​under different modes, and the historical meteorological observation data includes multiple meteorological element observation values ​​corresponding to each grid in the target area; Based on the historical meteorological observation data, the historical meteorological forecast data is subjected to spatiotemporal interpolation and multi-mode fusion to obtain a plurality of meteorological element forecast values ​​after fusion for each of the grids in the target area; The machine learning model is trained using the meteorological element prediction values ​​before and after fusion and the historical meteorological observation data to obtain a meteorological element correction model set. The meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather. The obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each of the grids in the target area. The multiple meteorological element target correction values ​​are used to provide transmission line risk warning for the target area.

2. The transmission line risk early warning method based on intelligent grid correction according to claim 1 is characterized in that: Based on the historical meteorological observation data, the historical meteorological forecast data is subjected to spatiotemporal interpolation and multi-mode fusion to obtain a plurality of meteorological element forecast values ​​after fusion for each grid in the target area, including: Performing spatiotemporal interpolation on the historical meteorological forecast data under each mode; Determine the mode fusion weights corresponding to the same meteorological element under different modes based on the historical meteorological observation data and the historical meteorological forecast data after spatiotemporal interpolation; For each of the grids in the target area, the meteorological element prediction values ​​of the same meteorological element in the grid under different modes are multi-mode fused using the mode fusion weight to obtain the fused meteorological element prediction value of the grid.

3. The transmission line risk early warning method based on intelligent grid correction according to claim 2 is characterized in that: Based on the historical meteorological observation data and the historical meteorological forecast data after spatiotemporal interpolation, the mode fusion weights corresponding to the same meteorological element in different modes are determined, including: The following processing is performed for the historical weather forecast data in each mode: Normalizing the historical meteorological forecast data after spatiotemporal interpolation to obtain a plurality of meteorological element forecast values ​​normalized in this mode for each grid in the target area; And, performing residual processing on the historical meteorological forecast data after spatiotemporal interpolation according to the historical meteorological observation data to obtain residual data of multiple meteorological elements of each of the grids in the target area; Based on the normalized meteorological element prediction value and the meteorological element residual data corresponding to the same meteorological element, the mode fusion weight corresponding to the same meteorological element in this mode is determined.

4. The transmission line risk early warning method based on intelligent grid correction according to claim 3 is characterized in that: Using the meteorological element forecast values ​​before and after fusion and the historical meteorological observation data, the machine learning model is trained to obtain a meteorological element correction model set, including: Constructing a preliminary sample data set based on the meteorological element forecast values ​​before and after fusion and the historical meteorological observation data; Taking the predicted values ​​of the meteorological elements before and after fusion in the preliminary sample data set as input, and taking the historical meteorological observation data as training labels, the XGBoost model is trained to obtain a target XGBoost model, the SHAP method is used to interpret the model, key meteorological elements are screened, and a target sample data set is constructed based on the predicted values ​​of the meteorological elements before and after fusion corresponding to the key meteorological elements and the historical meteorological observation data; Based on the target sample data set, the meteorological element prediction values ​​before and after fusion corresponding to each key meteorological element are taken as input, and the meteorological element observation values ​​corresponding to the key meteorological elements in the historical meteorological observation data are taken as training labels, and the machine learning model is trained to obtain a meteorological element correction model set.

5. The transmission line risk early warning method based on intelligent grid correction according to claim 1 is characterized in that: The method further comprises: Acquiring multi-mode meteorological forecast data for the target area under severe convective weather; The multi-mode meteorological forecast data is corrected by using the meteorological element correction model set to obtain initial meteorological correction data output by each meteorological element correction model in the meteorological element correction model set, wherein the initial meteorological correction data includes initial correction values ​​of multiple meteorological elements corresponding to each of the grids in the target area; Performing multi-model fusion on the initial meteorological correction data to obtain target meteorological correction data; Based on the transmission line and tower location information corresponding to the target area, extracting multiple meteorological element target correction values ​​corresponding to the target grid to which the transmission line and tower belong from the target meteorological correction data; According to the target correction values ​​of multiple meteorological elements corresponding to the target grid, a transmission line risk warning is performed for the target area to obtain a transmission line risk warning result corresponding to the target area.

6. The transmission line risk early warning method based on intelligent grid correction according to claim 5 is characterized in that: The target meteorological correction data includes a wind speed correction value; According to the target correction values ​​of multiple meteorological elements corresponding to the target grid, a transmission line risk warning is performed for the target area to obtain a transmission line risk warning result corresponding to the target area, including: Determine the wind load borne by the tower under the wind speed at the tower height in the target area according to the wind speed correction value corresponding to the target grid; Determine the wind deflection angle of the insulator string based on the wind load borne by the tower and the gravity load borne by the transmission line, and determine the gap distance between the transmission line and the tower according to the wind deflection angle of the insulator string; Based on the comparison between the gap distance and the preset gap distance threshold, the transmission line risk warning result corresponding to the target area is determined.

7. The transmission line risk early warning method based on intelligent grid correction according to claim 5 is characterized in that: The initial meteorological correction data is subjected to multi-model fusion to obtain target meteorological correction data, including: For each of the meteorological elements, multi-model fusion is performed on the initial correction value of the meteorological element output by each meteorological element correction model for the meteorological element according to the dynamically adjustable model fusion weight to obtain the meteorological element target correction value corresponding to the meteorological element.

8. A transmission line risk warning device based on intelligent grid correction, characterized in that: include: A data acquisition module, used to acquire multi-mode historical meteorological forecast data and historical meteorological observation data corresponding to the target area, wherein the historical meteorological forecast data includes multiple meteorological element forecast values ​​under different modes, and the historical meteorological observation data includes multiple meteorological element observation values ​​corresponding to each grid in the target area; A data processing module is used to perform spatiotemporal interpolation and multi-mode fusion on the historical meteorological forecast data based on the historical meteorological observation data to obtain a plurality of meteorological element forecast values ​​after fusion for each of the grids in the target area; A model training module is used to train a machine learning model using the meteorological element prediction values ​​before and after fusion and the historical meteorological observation data to obtain a meteorological element correction model set, wherein the meteorological element correction model set is used to correct the multi-mode meteorological forecast data of the target area under severe convective weather, and the obtained target meteorological correction data includes multiple meteorological element target correction values ​​corresponding to each of the grids in the target area, and the multiple meteorological element target correction values ​​are used to provide transmission line risk warning for the target area.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the method according to any one of claims 1 to 7.

Images