Method and system for dynamically increasing capacity of shagor transmission line based on weather correction

Abstract

The present application relates to the technical field of dynamic capacity increasing of power transmission lines, and provides a dynamic capacity increasing method and system for Shaguo barren power transmission lines based on meteorological correction, which comprises: obtaining meteorological data and terrain factor data, and performing time-space alignment and interpolation; correcting the interpolated meteorological data based on a physical model to obtain physical prediction values; constructing a feature vector based on the physical prediction values and the obtained grid meteorological data, and predicting residual prediction values through an XGBoost regression residual model; adding the physical prediction values and the residual prediction values to obtain the final prediction values of the target meteorological parameters; and determining the capacity increasing margin of the current time step of the power transmission line based on the final prediction values of the target meteorological parameters. The method adopts a double-driving mechanism of prior fusion of a physical model and XGBoost residual learning to realize high-precision correction of meteorological parameters, and calculates the dynamic line rating based on the corrected meteorological parameters, thereby realizing accurate capacity increasing evaluation of the power transmission line.

Description

Technical Field

[0001] This invention relates to the technical field of dynamic capacity expansion of transmission lines, specifically to a method and system for dynamic capacity expansion of transmission lines in desert areas based on weather correction. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the continuous expansion of the power grid and the large-scale integration of new energy sources such as wind power and photovoltaics, the need to increase the transmission capacity of transmission lines while ensuring safe operation is becoming increasingly urgent. Existing transmission line capacity assessments typically employ the static line rating method. This method sets fixed current-carrying limits based on conservative meteorological conditions such as high ambient temperature and low wind speed. While this meets safety requirements, it fails to reflect the dynamic changes in meteorological conditions during actual line operation, resulting in underutilization of the transmission capacity under most operating conditions. Especially in desert and barren areas, where the scale of new energy transmission is large and the line operating environment is complex, traditional static assessment methods are no longer sufficient to meet the actual needs of refined line operation and dynamic capacity expansion. Therefore, it is necessary to improve the accuracy of acquiring and predicting key meteorological parameters along transmission lines, and on this basis, enhance the accuracy and reliability of dynamic capacity expansion assessments.

[0004] The existing technology has the following main problems: First, existing methods are unable to accurately obtain key meteorological parameters of the transmission line location, resulting in inaccurate basic data for dynamic capacity expansion calculations. Existing meteorological data mainly comes from grid-based meteorological forecast data and ground monitoring data. Among them, grid-based meteorological forecast data has low spatial resolution and is difficult to reflect the local micro-meteorological changes such as wind speed and ambient temperature along the transmission line, especially at the tower locations, caused by terrain undulations and differences in underlying surface. Although ground monitoring data has higher accuracy, the number of monitoring points is limited, making it difficult to cover the entire line, and there are also problems such as data loss and communication delays. Secondly, the Dynamic Line Rating (DLR) is calculated dynamically under current meteorological conditions by real-time monitoring of environmental parameters (such as wind speed, ambient temperature, and solar radiation intensity) and a thermal balance model, achieving a balance between safety and efficiency. The accuracy of dynamic capacity expansion technology is highly dependent on the precision of meteorological parameters. There are methods based on physical models or data-driven methods for predicting meteorological parameters. Existing physical models are mainly based on empirical formulas or standard thermal balance models, which are not adaptable to complex terrain, abnormal weather, and special environments such as high altitude, low humidity, strong winds and sandstorms, and large temperature differences in desert areas, and are prone to calculation errors. Purely data-driven machine learning or statistical models can use historical data to mine local meteorological patterns, but if physical constraints are lacking, the model may be inaccurate when extrapolating. Existing methods cannot meet the accuracy and reliability requirements of dynamic capacity expansion assessment for transmission lines. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes a dynamic capacity expansion method and system for desert transmission lines based on meteorological correction. This method employs a dual-drive mechanism that integrates physical model priors with XGBoost residual learning to achieve high-precision correction of meteorological parameters. Based on the corrected meteorological parameters, the dynamic line rating is calculated, thereby achieving accurate capacity expansion assessment of the transmission line.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: One or more embodiments provide a method for dynamic capacity expansion of the Shagohuang transmission line based on weather correction, including the following steps: Acquire grid meteorological data, tower monitoring meteorological data, and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; Based on a physical model that includes height and terrain correction, the interpolated grid meteorological data is corrected and predicted to obtain the physical prediction values ​​of the target meteorological parameters. Based on the physical prediction value at the current moment and the acquired grid meteorological data, a feature vector at the current moment is constructed and input into the trained XGBoost regression residual model to perform residual prediction and obtain the residual prediction value; the physical prediction value and the residual prediction value are added together to obtain the final prediction value of the target meteorological parameter. Based on the final predicted values ​​of the target meteorological parameters, heat balance calculations are performed to determine the capacity margin of the transmission line at the current time step.

[0007] One or more embodiments provide a weather-corrected dynamic capacity expansion system for the Shagolang transmission line, including: The data acquisition and alignment module is configured to acquire grid meteorological data, tower monitoring meteorological data and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; The physical prior prediction module is configured to correct and predict the interpolated grid meteorological data based on a physical model that includes height and terrain correction, and obtain the physical prediction value of the target meteorological parameter. The residual learning module is configured to construct a feature vector based on the physical prediction value at the current time and the acquired grid meteorological data, and input it into the trained XGBoost regression residual model to perform residual prediction and obtain the residual prediction value; the physical prediction value and the residual prediction value are added together to obtain the final prediction value of the target meteorological parameter. The capacity margin calculation module is configured to perform heat balance calculations based on the final predicted values ​​of the target meteorological parameters to determine the capacity margin of the transmission line at the current time step.

[0008] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the steps in the above-described method for dynamic capacity expansion of the desert transmission line based on weather correction.

[0009] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps in the above-described method for dynamic capacity expansion of the Shagohuang transmission line based on weather correction.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention acquires grid meteorological data, tower monitoring meteorological data, and terrain factor data, and performs spatiotemporal alignment and tower location interpolation to enable coarse-resolution grid data to more accurately reflect the local meteorological characteristics along transmission lines, especially at tower locations. Based on this, a physical model incorporating height and terrain corrections is used to correct the interpolation results, reducing the systematic bias of meteorological parameters under complex terrain conditions. Furthermore, an XGBoost regression residual model is combined to compensate for the residuals of the physical prediction results, integrating physical mechanism constraints with data-driven corrections to improve the prediction accuracy of target meteorological parameters. Finally, based on the corrected meteorological parameters, a heat balance calculation is performed to determine the capacity expansion margin of the transmission line at the current time step, thereby improving the accuracy and reliability of dynamic capacity expansion assessment results.

[0011] The advantages of the present invention, as well as its additional advantages, will be described in detail in the following specific embodiments. Attached Figure Description

[0012] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute a limitation thereof.

[0013] Figure 1 This is a general framework diagram of the dynamic capacity expansion method for transmission lines according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of the training process of the XGBoost residual learning model in Embodiment 1 of the present invention; Figure 3 This is a flowchart of the dynamic capacity expansion assessment in Embodiment 1 of the present invention. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0016] It should be noted that the terminology used herein is for describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof. It should be noted that, without conflict, the various embodiments and features within those embodiments can be combined with each other. The embodiments will now be described in detail with reference to the accompanying drawings.

[0017] This invention provides a two-stage fusion process for meteorological parameter correction and dynamic capacity expansion assessment. The core idea is to first use a physical model to correct and predict the original meteorological data, then employ an XGBoost machine learning model to refine the residuals of the physical predictions, obtaining high-precision meteorological parameters. These parameters are then applied to the calculation of dynamic line ratings based on the IEEE 738-2023 standard, ultimately achieving accurate capacity expansion assessment of transmission lines. The entire method ensures that the prediction results conform to physical laws while fully utilizing historical data to improve accuracy, making it suitable for the high-precision prediction of meteorological parameters required in dynamic capacity expansion assessments of transmission lines. Specific embodiments are described below.

[0018] Example 1 In one or more of the technical solutions disclosed in the embodiments, such as Figures 1 to 3 As shown, a dynamic capacity expansion method for the Shagohuang transmission line based on weather correction includes the following steps: Step 1: Obtain grid meteorological data, tower monitoring meteorological data, and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; Step 2: Based on the physical model that includes height and terrain corrections, correct and predict the interpolated gridded meteorological data to obtain the physical prediction values ​​of the target meteorological parameters. ; Step 3: Based on the physical prediction value at the current time Using the acquired grid meteorological data, construct the feature vector for the current time, input it into the trained XGBoost regression residual model for residual prediction, and obtain the residual prediction value; add the physical prediction value and the residual prediction value to obtain the final prediction value of the target meteorological parameter; Step 4: Based on the final predicted values ​​of the target meteorological parameters, perform heat balance calculations to determine the capacity margin of the transmission line at the current time step; In the above implementation, grid meteorological data, tower monitoring meteorological data, and terrain factor data are first acquired and spatiotemporally aligned. Then, the grid meteorological data is interpolated to the tower locations of the transmission line. This allows the coarse-resolution grid data, which originally could not directly reflect the local meteorological differences at the tower locations, to establish a correspondence with the actual spatial distribution of the line. This improves the problem of low spatial resolution of grid meteorological data in the prior art, which makes it difficult to accurately characterize the micro-meteorological characteristics of the tower locations. It also provides basic data that is closer to the actual location of the line for subsequent dynamic capacity expansion calculations.

[0019] By employing a physical model that incorporates altitude and terrain corrections, the interpolated gridded meteorological data is modified for forecasting. This allows the target meteorological parameters to comprehensively account for the impacts of altitude changes and complex terrain disturbances on temperature and wind speed distributions, thereby reducing the systematic biases that arise when coarse-scale meteorological data is directly applied to desert and Gobi regions. Compared to methods that rely solely on raw grid forecast results for evaluation, this approach improves the rationality and stability of local meteorological parameter forecasts.

[0020] By constructing a feature vector at the current time and inputting it into a trained XGBoost regression residual model for residual prediction, the final predicted value of the target meteorological parameter is obtained by superimposing the residual prediction value with the physical prediction value, thus forming a collaborative correction mechanism between the physical model and the data-driven model. This approach utilizes both the mechanistic constraints provided by the physical model and the residual learning to compensate for nonlinear errors that are difficult to explicitly model in complex environments. Therefore, it can improve the problems of insufficient adaptability when relying solely on the physical model and insufficient extrapolation reliability when relying solely on the data-driven model, thereby improving the prediction accuracy of the target meteorological parameter.

[0021] Thermal balance calculations are performed based on the final predicted values ​​of target meteorological parameters to determine the capacity expansion margin of the transmission line at the current time step. This allows the dynamic capacity expansion assessment to be based on the corrected meteorological parameters, thereby improving the accuracy and reliability of the dynamic line current carrying capacity assessment results. It also alleviates the problem that existing technologies may lead to conservative or deviated-from-actual-operational-capacity results due to inaccurate basic meteorological data, thus making it more conducive to the refined assessment of the transmission line operating capacity.

[0022] In step 1, for the transmission line area to be regulated, grid meteorological data, tower monitoring meteorological data, and terrain factor data are acquired, spatiotemporally aligned, and the grid meteorological data is interpolated to the tower locations of the transmission line, including the following steps: Step 11, Data Acquisition and Spatiotemporal Alignment: Align the acquired grid meteorological data, tower monitoring meteorological data, and topographic factor data in terms of temporal and spatial scales; Specifically, multi-source data is organized according to the predetermined tower numbers and time series on the transmission line. The multi-source data includes at least: grid meteorological data obtained from meteorological forecasting systems or reanalysis databases, monitoring meteorological data deployed at the transmission line towers, and topographic factor data along the transmission line.

[0023] Optionally, the grid meteorological data includes predicted values ​​of wind speed, wind direction, temperature, humidity, solar radiation, etc., at a certain spatial resolution within the transmission line area. Meteorological data monitoring can include wind speed, temperature, etc., and is the wind speed and temperature data measured by sensors installed at the transmission line towers; Topographic data along the transmission line, including tower elevation, surface slope, underlying surface roughness, and slope aspect; Alignment of time and space scales: Specifically, unifying the coordinate reference system, height reference, and timestamp reference for data from different sources, so that grid meteorological data, tower monitoring meteorological data, and topographic factor data correspond to the same geographic coordinate system and height reference in space, and correspond to a unified time scale in time. Step 12: Based on the geographical coordinates of each tower, interpolate and map the grid meteorological data to the corresponding tower locations to obtain the initial meteorological element values ​​at the tower locations. Optionally, spatial interpolation can be performed using nearest neighbor, bilinear, or distance-weighted interpolation methods. Step 13: Time-align the interpolated grid meteorological data mapped to the corresponding tower location with the monitoring meteorological data of the corresponding tower. Specifically, multi-source data is resampled to a unified time granularity, and missing values ​​are marked or filled in for communication delays or missing data in the meteorological data of pole monitoring, so as to obtain complete and aligned grid meteorological data and pole monitoring meteorological data.

[0024] Optionally, a uniform time granularity can be set as compensation for the collected data, such as converting all data into time series with a step size of 10 minutes or 1 hour. Furthermore, historical meteorological data are standardized, anomaly detection and correction are performed based on DBSCAN, and missing value imputation is performed based on KNNImputer, thereby obtaining high-quality input data for subsequent calculations. When studying historical data, problems such as missing data, small data volume, and poor data quality often arise, hindering computation. Therefore, machine learning algorithms are introduced to clean and impute historical data, optimizing data quality while ensuring reliability, thus facilitating subsequent calculations. The specific operational process is as follows: First, all historical meteorological data is standardized. Then, DBSCAN detects and processes outliers, identifying dense and sparse regions in the data; outliers are typically located in sparse regions. These values ​​are marked as noise points and corrected using smoothing methods. Finally, KNNImputer imputes missing values ​​by calculating the Euclidean distance between the missing value sample and the complete sample, selecting the k nearest neighbors, and filling the missing values ​​with their mean or weighted average.

[0025] In step 2, based on the obtained alignment data, the atmospheric physics model is used to make basic predictions of key meteorological parameters at the tower points, generating physical prior values. This includes physical predictions of ambient temperature and wind speed. The physical model incorporates atmospheric boundary layer principles, considering the variations of meteorological parameters with altitude and topography, and corrects the gridded meteorological data accordingly. The physical model corrects the grid data based on atmospheric physical laws to obtain the baseline prediction value of the tower location; In step 2, based on a physical model that includes height and terrain corrections, the interpolated grid meteorological data is corrected and predicted to obtain the physical prediction values ​​of the target meteorological parameters at each tower location. ; The achievable physical model is a modified model of the target meteorological parameters set according to the laws of atmospheric physics, including an altitude correction model and a terrain correction model; the target meteorological parameters include meteorological parameters such as wind speed and temperature. In some embodiments, the altitude correction model includes an altitude-based temperature correction model and an altitude-based wind speed correction model; For ambient temperature, considering the empirical law that air temperature decreases with increasing altitude, a temperature lapse rate is introduced. The temperature is corrected for the grid temperature based on altitude. The formula for the altitude-based temperature correction model can be expressed as: ; in, This refers to the reference temperature in the gridded meteorological data, and the corresponding altitude for the reference temperature data measurement. ; This represents the actual altitude of the tower. Using this linear, altitude-based temperature correction model, the original temperature estimate can be converted to the tower height, compensating for the lower temperatures observed at high altitudes.

[0026] A height-based wind speed correction model is used, employing the atmospheric boundary layer wind profile exponential law to correct for wind speed at height. It is assumed that the wind speed given in the grid data corresponds to a height... (e.g., wind height of 10m), the tower conductor height is Therefore, according to the power law relationship, the wind speed correction model based on altitude is as follows: ; in, It is an exponential factor for wind speed profile, determined by surface roughness and atmospheric stability, and is usually in the range of 0.1 to 0.3 in flat and open areas; The reference wind speed in the gridded meteorological data, and the corresponding height for wind speed and temperature data measurements. This formula extrapolates low-altitude wind speeds to tower height, making the predicted wind speeds more consistent with the atmospheric flow at the actual height.

[0027] In some embodiments, the terrain correction model is a terrain-based wind speed correction model; The terrain-based wind speed correction model considers the disturbance effect of complex terrain on the wind field and further corrects the predicted wind speed at the tower. When the terrain is significantly undulating, wind flowing along the slope will produce effects such as acceleration on the windward slope or deceleration on the leeward slope. This embodiment introduces a slope correction factor. To characterize the influence of topography, the slope correction factor The calculation formula is: ; Where slope is the terrain gradient (the angle between the slope and the horizontal plane), θ is the angle between the wind direction and the slope normal, and k is an empirical coefficient. Based on this factor, when the tower is on the windward slope (i.e., wind blows from a lower elevation to the upper slope), cos(θ) > 0, and the local wind speed increases due to slope uplift. Conversely, when on the leeward slope, the airflow forms vortices downhill, causing wind speed to decrease. In practical applications, the acceleration effect of the windward slope and the deceleration effect of the leeward slope can be considered separately according to the slope, and the above effects can be combined by selecting an appropriate k.

[0028] Finally, multiply the height-corrected wind speed by... The comprehensive corrected physical wind speed prediction is obtained using the terrain-based wind speed correction model formula: ; In this embodiment, through dual corrections based on altitude and terrain, the physical model can output benchmark predictions of meteorological parameters that more closely match the actual location of the tower. For example, the physical prediction value of the temperature at the tower point can be obtained through the above calculations. and wind speed physical prediction value Together they constitute the physical prediction quantity .

[0029] Step 3, Residual Construction and Feature Engineering: Construct residual samples using the difference between the measured values ​​and the physical prediction values ​​at the same time, train the XGBoost regression residual model for residual learning, construct feature vector X using the grid meteorological data of the current time t and the historical time, and the physical prediction values ​​predicted in Step 2, and input it into the trained XGBoost regression residual model for residual prediction to obtain the residual prediction values. Furthermore, the training process of the XGBoost regression residual model includes the following steps: Step S31, Residual Sample Construction: Based on the physical prediction value obtained in Step 2, calculate the difference with the actual meteorological value of the tower monitoring at the corresponding time to obtain the residual sample used for model training, which is used as the output of the XGBoost regression residual model; Specifically, obtain physical prediction values Then, it is compared with the actual meteorological values ​​at the corresponding time, and the difference is calculated to obtain the residual. The residual is defined as the actual monitored value minus the physical prediction value in the meteorological data, i.e.: ; in, This represents the measured value of meteorological data for the corresponding tower location; The residual reflects the amount by which the physical model's prediction deviates from reality. For each meteorological parameter, such as wind speed and temperature, a series of residual samples can be obtained.

[0030] Step S32: Based on the physical prediction values ​​and the interpolated grid meteorological data, construct the input feature vector for the XGBoost regression residual model used for residual learning. The input feature vector includes the original meteorological features, topographic features, time features, lag features, physical prior features, and difference features. To fit the residuals using a machine learning model, an input feature vector needs to be constructed for each residual sample. In this embodiment, features are designed from multiple perspectives to fully characterize the factors affecting the residuals, including but not limited to: Raw meteorological features include forecast values ​​from gridded meteorological data, such as wind speed, ambient temperature, relative humidity, solar radiation intensity, and wind direction. These features provide basic meteorological and environmental information.

[0031] Topographic features, which are the geographical attributes of the route points, such as tower elevation, slope, aspect, wind direction, slope angle, and surface roughness type, help the model learn the degree of influence of topographic factors on meteorological errors.

[0032] Time characteristics are introduced, such as time codes representing periodic changes, like the sine and cosine values ​​calculated using the number of hours in the day. , (This is used to capture systematic biases caused by diurnal variations and seasonal cyclical factors.)

[0033] The lag characteristic, i.e., gridded meteorological data from historical moments, is used to reflect the continuity of atmospheric conditions. It allows us to consider the previous moment (t...) as a reference point. 1) Values ​​such as grid wind speed, grid temperature, and solar radiation from several hours ago can be used as lag inputs, and monitoring values ​​from the previous moment can be introduced as needed for updates in online rolling mode. Lag features help the model learn the evolution trends and short-term correlations of atmospheric parameters.

[0034] Physical prior features, i.e. physical predictions output by the physical model, are used to characterize the model's basic estimates for the current moment and help the model automatically adjust the correction strength according to the magnitude of the deviation.

[0035] Differential features, such as the difference between actual monitoring data and gridded meteorological data, are derived features. For example, the difference between the monitored wind speed at the previous moment and the gridded forecast wind speed can indicate the direction and magnitude of the deviation of the model forecast from the actual measurement, providing prior information about the residuals. This feature is relevant to the location of transmission lines that can be obtained from actual monitoring values ​​during actual operation. Through the feature engineering described above, the various feature codes are concatenated to obtain a feature vector, forming training samples for the residual learning model. It should be noted that feature selection is flexible and can be adjusted according to the application scenario and data availability; however, the above six types of features collectively provide a comprehensive information foundation for the generation of residuals.

[0036] Step S33: Take the input feature vector as input and the residual sample as output, and input them into the XGBoost regression residual model for training to obtain the trained XGBoost regression residual model. XGBoost is a gradient boosting algorithm based on decision tree ensembles, which approximates the objective function by iteratively training a series of decision trees. In this embodiment, independent XGBoost regression residual models can be trained separately for different meteorological parameter residuals (such as wind speed residuals and temperature residuals).

[0037] Furthermore, such as Figure 2 As shown, before training, the samples are split into training and validation sets, based on the constructed input feature vector sample dataset, for model training and performance evaluation. To ensure a reasonable splitting order, a time-sequential method or a combination of random shuffle and cross-validation can be used to improve the accuracy of the model's generalization ability evaluation.

[0038] It is feasible to use the loss function of the XGBoost regression residual model as the optimization objective, i.e., minimizing the mean squared error (MSE) of the residual prediction error. During training, the model reduces this error by progressively fitting the residuals of the residuals.

[0039] During training, the hyperparameters of the XGBoost model are adjusted using methods such as grid search or Bayesian optimization on the validation set. These parameters include, but are not limited to, the maximum depth of the decision tree, the learning rate, the subsampling ratio, and the number of decision trees. Cross-validation is used to determine the parameter combination that enables the model to achieve the best performance on the validation set, thereby preventing overfitting and improving generalization performance.

[0040] Using a predetermined optimal parameter configuration, the XGBoost model is iteratively trained on the training set, with a new regression tree added in each iteration to fit the current residuals. As iterations proceed, the training error gradually decreases, and training stops when the preset upper limit for the number of trees is reached or the error converges. Finally, evaluation metrics such as the coefficient of determination R² and root mean square error are calculated on the validation set to confirm that the model has a good fitting ability for the residuals.

[0041] Through this step, wind speed residual prediction models and temperature residual prediction models can be obtained, and their output functions can be collectively referred to as... That is, input feature vector X, output predicted value of residual.

[0042] After the above training process, two parts, the physical model and the residual model, have been obtained. The next step is the online application stage, where the physical prior predictions and XGBoost residual correction results are fused in real time to generate the final prediction of the target meteorological parameters.

[0043] In step 3, the acquired grid meteorological data at the current time t and historical times, along with the physical prediction values ​​from step 2, are used to construct a feature vector X. This feature vector X is then input into the trained XGBoost regression residual model for residual learning and prediction to obtain the residual prediction values. The process includes the following steps: Step 31: At any given time t, obtain the grid meteorological data of the current time t and the historical time, and the physical prediction value predicted in Step 2. Construct the feature vector X of the current time. The input feature vector includes the original meteorological features, terrain features, time features, lag features, physical prior features and difference features. Step 32: Input the current feature vector X into the trained XGBoost regression residual model to calculate the residual prediction value. ; Furthermore, the residual predicted values ​​are added back to the physical predicted values ​​to obtain the corrected meteorological parameter predicted values, i.e.: ; This output This refers to the prediction results of meteorological parameters such as wind speed or ambient temperature at the tower location, which have higher accuracy compared to the original physical prediction values.

[0044] Furthermore, this embodiment predicts residuals based on the XGBoost regression residual model, including rolling prediction mode and pure meteorological prediction mode; Rolling forecast mode: If the monitored meteorological values ​​at the current or previous time are available, the rolling forecast mode is entered. Specifically: after the forecast is completed at time t, if the monitored values ​​at that time can be obtained, the monitored values ​​are fed back to construct the input feature vector for the next time t+1; the monitored wind speed, monitored temperature, or historical residuals calculated based on the monitored values ​​at time t are used as the lag features of time t+1 and input into the XGBoost regression residual model to obtain the residual forecast result for the next time. In the above implementation, by introducing the latest real observation information, the residual model can dynamically adjust the residual prediction results for the next moment according to the latest local meteorological conditions, thereby suppressing error accumulation and improving the prediction accuracy and stability in continuous prediction scenarios.

[0045] Specifically, for real-time applications, calculation is performed at each time t. Simultaneously, the monitored measured value at that moment (if it exists) is fed back to update the input feature vector at the next moment t+1. For example, the latest monitored wind speed at moment t is used as the monitoring lag feature at moment t+1, thereby introducing the latest real information into the model and achieving iterative prediction. This rolling prediction mode ensures that the model can adaptively correct its own biases: once there is an error in the prediction at a certain moment, the residual model will readjust its output based on the corrected lag feature by introducing the real value at the next moment, so that the error will not accumulate in the long term.

[0046] Pure meteorological forecast mode: When monitoring data is temporarily unavailable, the system switches to pure meteorological forecast mode. In pure meteorological forecast mode, the system does not rely on real-time monitoring data at the current moment, but instead makes short-term forecasts based on grid meteorological data, topographic factor data, physical prediction values, and existing historical data. The lag features that originally relied on the actual monitoring values ​​can be generated using at least one of the following alternative methods: The final predicted value from the previous moment is used to replace the monitored value; The physical prior prediction value from the previous moment is used to replace the monitored value; The lag characteristics are estimated based on the recent grid forecast trends. Use historical statistical features or sliding window mean to construct alternatives to the lagged input.

[0047] In certain application scenarios, monitoring data may be temporarily unavailable (e.g., communication failures or sensor maintenance). To address this, this embodiment provides a pure meteorological forecasting mode: when real-time monitoring updates are lacking, the model can still make short-term predictions relying solely on physical forecasts and existing historical data. In this case, lag characteristics can be replaced by previous model predictions or estimated using recent grid forecast trends, allowing the model to operate independently for a period. The two modes can be seamlessly switched based on the availability of monitoring data, ensuring uninterrupted forecasting of meteorological parameters for transmission lines. When monitoring data is restored, a rolling mode is immediately activated to maximize prediction accuracy.

[0048] The processing framework described in this embodiment supports the rolling integration and forecast updates of real-time data. When new monitoring data arrives, the meteorological correction model updates the lagging input in a timely manner, enabling the forecast results to correct deviations from previous moments and maintain sensitivity to environmental changes. This online learning-based forecasting mode ensures that the model still has accurate response capabilities under sudden weather changes or extreme events, improving the timeliness and reliability of forecasts, and thus enhancing the real-time performance and accuracy of dynamic capacity expansion assessments.

[0049] The corrected meteorological parameters (corrected wind speed) obtained in step 3 and ambient temperature Step 4 enters the dynamic capacity expansion assessment stage based on modified meteorological parameters. A thermal balance model is constructed based on the IEEE 738-2023 standard, and the line current carrying capacity is accurately calculated in combination with the special climate characteristics of the Shagohuang area.

[0050] This embodiment employs a dual-drive mechanism combining a prior physical model with XGBoost residual learning. The physical model integrates altitude correction and micro-topographic wind field correction, fully considering the vertical decay of meteorological elements and topographic disturbance effects in complex underlying surface areas such as deserts, Gobi, and high altitudes. By applying altitude correction to air temperature using the temperature lapse rate Γ and correcting wind speed using the wind speed profile exponent α combined with the slope factor, the systematic bias of coarse-scale weather forecasts in local applications is significantly reduced. Simultaneously, an XGBoost machine learning model is introduced to learn and correct the physical prediction residuals, automatically capturing nonlinear relationships and random errors that are difficult to explicitly quantify in the physical model using historical monitoring data, significantly reducing the final prediction error. Compared to methods relying solely on physical models or single statistical models, this approach provides higher accuracy in wind speed and temperature predictions.

[0051] In step 4, based on the final predicted values ​​of the target meteorological parameters, a heat balance calculation is performed to determine the capacity margin of the transmission line at the current time step, including the following steps: Step 41: Obtain the key environmental parameters affecting the transmission line capacity under extreme conditions from historical meteorological data, and use the pre-built steady-state thermal balance calculation model to calculate the static rated values ​​based on the key environmental parameters under extreme conditions. Step 42: Take the final predicted value of the target meteorological parameter output in Step 3 as the real-time key environmental parameter. Based on the pre-built transient thermal balance calculation model, calculate the conductor temperature change under the real-time key environmental parameter. Based on the conductor temperature change, determine the dynamic rated value at the current time step. Step 43: Use the dynamic and static ratings to determine the capacity margin for the current time step.

[0052] In step 41, based on the line design standards and historical extreme weather data, the key environmental parameters affecting the line capacity are determined, including: Step 411: Based on the traditional heat balance equation, the relationship between each environmental factor and the line's current carrying capacity is determined using meteorological simulation data and the controlled variable method, thereby obtaining the key environmental parameters affecting the line's capacity, including ambient temperature, wind speed, solar radiation intensity, and wind direction.

[0053] Step 412: Select key environmental parameters for extreme conditions based on historical meteorological data, including the highest ambient temperature, lowest wind speed, strongest solar radiation intensity, and conservative conditions for unfavorable wind direction.

[0054] Furthermore, considering the climatic characteristics of the desert region, a steady-state thermal balance calculation model is constructed based on the principle of thermal balance. Specifically, through the balance relationship of four power components—convective heat dissipation, radiative heat dissipation, solar radiation heat absorption, and resistive heating—a steady-state thermal balance calculation model is established, as follows: ; in, It is the heating power of the wire's resistance. It is the solar radiation heat absorption power of the conductor. It is the convective heat dissipation power of the conductor. This is the radiative heat dissipation power of the conductor; the equation shows that the sum of the convective heat dissipation power and the radiative heat dissipation power of the transmission line conductor is equal to the sum of its resistive heating power and the solar radiation heat absorption power. When the conductor reaches thermal equilibrium, its temperature remains stable.

[0055] Furthermore, the power term in the steady-state thermal equilibrium model is calculated, including: The calculation of convective heat dissipation power specifically considers both natural convection and forced convection. The natural convection heat dissipation power and the heat dissipation power corresponding to different forced convection formulas are calculated separately, and the maximum value is taken as the conductor convective heat dissipation power. In the calculation process, an effective wind speed coefficient is introduced to correct the wind speed. At the same time, a low humidity correction factor and a humidity correction term are introduced to correct the convective heat dissipation capacity. Convection cooling is an important method for heat dissipation in power transmission lines. The calculation of convection cooling power includes natural convection and forced convection, and the maximum value of the two is ultimately taken. Natural convection occurs under still air conditions (wind speed of 0), where heat is dissipated through the natural flow of air. Forced convection is caused by wind, and the results of two formulas need to be calculated and the larger value taken, then compared with natural convection, and the maximum value is finally taken as the convection cooling power.

[0056] Considering that the inertia of sand particles in the desert region consumes the kinetic energy of the wind, resulting in the effective wind speed acting on the conductor being lower than the measured wind speed, an effective wind speed coefficient is introduced. The higher the concentration of dust in the environment, The smaller the value, the better. Additionally, due to the low humidity and high altitude characteristics of the desert region, the aerodynamic viscosity is lower than the standard value; therefore, a low humidity correction factor is introduced. The thermal conductivity of the air is lower than the standard value, so a humidity correction term is introduced.

[0057] The calculation of radiative heat dissipation power is specifically based on the Stefan-Boltzmann law, combined with the emissivity of the conductor surface, the conductor temperature, and the ambient temperature. Radiative heat dissipation is calculated based on the Stefan-Boltzmann law. The solar radiation heat absorption power needs to consider the sun's position (elevation angle, azimuth angle), atmospheric attenuation, and altitude correction. The altitude-corrected solar radiation intensity takes into account the influence of altitude on solar radiation. Resistance heat loss is the heat loss rate caused by current flowing in the conductor. The conductor resistance is not a fixed value but varies with temperature, and is calculated using linear interpolation.

[0058] In step 41, based on the key environmental parameters under extreme operating conditions, the static rated value of the transmission line capacity under extreme operating conditions is calculated using a steady-state thermal balance calculation model. The calculation formula is: ; in, This represents the equivalent resistance of the transmission line. The core principle of this formula is to balance the heat absorption and dissipation power of the conductor to ensure that the conductor temperature does not exceed a safe threshold, thereby calculating the maximum allowable current under extreme weather conditions, i.e., the static rated value. ; When calculating the static ratings of transmission lines, key environmental parameters under extreme operating conditions are determined based on a steady-state thermal balance calculation model and historical meteorological data. Specifically, the calculations utilize key environmental parameters under extreme conditions in the measurement area. The static ratings calculated under these conditions ensure the safe operation of the line even in the most severe weather environments.

[0059] The probability of the worst weather conditions occurring is very small, and the maximum current carrying capacity of transmission lines is affected by various weather conditions, which are constantly changing over time. Therefore, the maximum current carrying capacity of a transmission line is not a static straight line like the static rated value. In order to reduce the wasted capacity expansion margin of the static rated value, a transient thermal balance model of the transmission line is established in step 42 of this embodiment to calculate the change of the transmission line's current carrying capacity with environmental changes.

[0060] Step 42: Dynamic Rated Value Calculation; Using the final predicted value of the target meteorological parameters output in Step 3 as the real-time key environmental parameters, and based on the pre-built transient thermal balance calculation model, the conductor temperature change under the real-time key environmental parameters is calculated. Based on the conductor temperature change, the dynamic rated value of the transmission line capacity at the current time step is determined. The method for determining the dynamic rated value includes the following: Step 421: The final predicted value of the target meteorological parameters output in Step 3 ( and As a key real-time environmental parameter, the initial dynamic rated value is obtained by calculating using a steady-state thermal balance calculation model. ; Optionally, real-time key environmental parameters may include: real-time ambient temperature prediction, real-time wind speed prediction, real-time solar radiation intensity prediction, and real-time wind direction prediction. In this embodiment, real-time ambient temperature prediction and real-time wind speed prediction are used as key environmental parameters. Step 422: Divide the measurement time range into multiple time intervals. Based on the current and key environmental parameters in each time interval, as well as the pre-built transient thermal balance calculation model, calculate the conductor temperature change in each time step within the time interval. Specifically, assuming that the current and key environmental parameters remain constant within each time interval, the average value of that time period is taken; based on the pre-built transient thermal balance calculation model, the conductor temperature change within each time step of the time interval is calculated; The transient thermal balance calculation model is based on the relationship between the conductor's heat capacity and heat power balance, and is used to describe the dynamic change of conductor temperature between adjacent time steps. The conductor temperature calculation formula is as follows: ; The formula for calculating the temperature change of the conductor described above is transformed as follows: ; in, It is the mass per unit length of the conductor. It is the specific heat capacity of the conductor material. It is the first Each time step. For the temperature of the conductor; Step 423: Based on the conductor temperature change in each time step, calculate the balance relationship of the four power components: convective heat dissipation, radiative heat dissipation, solar radiation heat absorption, and resistive heating in each time step. Specifically, if the conductor temperature change does not exceed the maximum allowable temperature, the balance relationship of the four power components—convective heat dissipation, radiative heat dissipation, solar radiation heat absorption, and resistive heating—is obtained based on the conductor temperature change in each time step.

[0061] After each time step, the conductor resistance is recalculated based on the new conductor temperature. Convection heat dissipation Radiative heat dissipation Solar heat increase It is typically updated at each time interval.

[0062] Step 424: Utilize the balance relationship of the four power components—convective heat dissipation, radiative heat dissipation, solar radiation heat absorption, and resistive heating—within each time step to determine the change in dynamic rated value for each time step. The calculation formula is: ; in, They represent Temperature at any moment and Temperature at any moment; Step 425: Add the change in the rated value at each time step to the initial dynamic rated value to obtain the dynamic rated value I at each time step. DLR(t) .

[0063] In step 43, the dynamic rating I is used. DLR(t) With static rating I SLR The process for determining the capacity margin at the current time step is as follows: Step 431: Calculate the difference between the dynamic rating and the static rating, and use it as the first difference. The calculation formula is as follows: ; Among them, I SLR It is the static rated value, I DLR(t) It is the dynamic rating, and ΔI is the difference between the dynamic rating and the static rating.

[0064] Step 432: Calculate the ratio of the first difference to the static rated value to determine the capacity margin for the current time step. ; Transmission line capacity margin refers to the difference between the maximum load capacity that a line can currently accommodate and the existing load, provided that the line operates safely and stably. It is a crucial indicator of a line's carrying capacity potential and is of great significance for the planning, operation, and upgrading of power systems. Each line has its maximum carrying capacity limit; blindly increasing the load may lead to conductor overheating, accelerated insulation aging, and even safety accidents such as short circuits and fires. In this embodiment, capacity margin calculation clarifies the upper limit of the line's load capacity, preventing overload from affecting system safety. Accurately calculating the capacity margin and then reasonably increasing the load within that range can fully utilize the line's carrying capacity, avoid idle and wasted line resources, and improve the economic efficiency of the power system.

[0065] Through the above implementation steps, this embodiment realizes a complete process from high-precision correction of meteorological parameters to accurate assessment of dynamic capacity expansion. The physical model provides reliable basic estimates, while XGBoost residual learning specifically compensates for the systematic bias of the basic estimates, making the final meteorological forecast results closer to actual measurements. Applying high-precision meteorological parameters to a heat balance model based on the IEEE 738-2023 standard, and making adaptive corrections for the special environment of the desert region, effectively reduces forecast errors caused by complex terrain and local special climates, providing more accurate meteorological input and capacity expansion decision support for the dynamic capacity assessment and safety monitoring of transmission lines.

[0066] It should be noted that the method steps in this embodiment can be implemented through a combination of hardware and software. For example, the above method can be deployed as a meteorological parameter correction and dynamic capacity expansion assessment system, whose module division is as follows: Figure 1 As shown, the system includes modules for data acquisition, spatiotemporal alignment, physical prediction, residual learning, fusion output, static analysis, dynamic analysis, and capacity expansion. Each module corresponds to the function described in the aforementioned methods, achieving full automation from data acquisition and preprocessing to model calculation and result output. This system can be implemented using a computer or embedded device. The relevant software program is stored on a computer-readable medium, and the processor executes the program to complete the functions of each step.

[0067] In practical implementation, the algorithm details can be adjusted as needed, such as changing the machine learning model or adding other feature variables, all of which fall within the scope of protection of this invention. The technical solution provided in this embodiment can effectively improve the accuracy and reliability of predicting key meteorological parameters of transmission lines, and apply high-precision meteorological parameters to dynamic capacity expansion assessment, realizing a complete technical chain from meteorological correction to capacity expansion decision-making.

[0068] Example 2 Based on Example 1, this example provides a dynamic capacity expansion system for the Shagohuang transmission line based on weather correction, including: The data acquisition and alignment module is configured to acquire grid meteorological data, tower monitoring meteorological data and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; The physical prior prediction module is configured to correct and predict the interpolated grid meteorological data based on a physical model that includes height and terrain correction, and obtain the physical prediction value of the target meteorological parameter. The residual learning module is configured to construct a feature vector based on the physical prediction value at the current time and the acquired grid meteorological data, and input it into the trained XGBoost regression residual model to perform residual prediction and obtain the residual prediction value; the physical prediction value and the residual prediction value are added together to obtain the final prediction value of the target meteorological parameter. The capacity margin calculation module is configured to perform heat balance calculations based on the final predicted values ​​of the target meteorological parameters to determine the capacity margin of the transmission line at the current time step.

[0069] Furthermore, the capacity expansion margin calculation module includes: The static module is configured to acquire key environmental parameters affecting the transmission line capacity under extreme conditions from historical meteorological data, and use a pre-built steady-state thermal balance calculation model to calculate the static rated value of the transmission line capacity under extreme conditions based on the key environmental parameters. The dynamic module is configured to use the final predicted value of the target meteorological parameters as the real-time key environmental parameters, calculate the conductor temperature change under the real-time key environmental parameters based on the pre-built transient thermal balance calculation model, and determine the dynamic rated value of the transmission line capacity at the current time step based on the conductor temperature change. The capacity expansion module is configured to determine the capacity expansion margin for the current time step using dynamic and static ratings.

[0070] It should be noted that each module in this embodiment corresponds one-to-one with each step in embodiment 1, and their specific implementation process is the same, so it will not be repeated here.

[0071] Example 3 Based on Embodiment 1, this embodiment provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps in the dynamic capacity expansion method for the desert transmission line based on weather correction described in Embodiment 1.

[0072] Example 4 Based on Embodiment 1, this embodiment provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, they complete the steps in the dynamic capacity expansion method for the desert transmission line based on weather correction described in Embodiment 1.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0074] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A dynamic capacity expansion method for desert transmission lines based on meteorological correction, characterized in that, Includes the following steps: Acquire grid meteorological data, tower monitoring meteorological data, and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; Based on a physical model that includes height and terrain correction, the interpolated grid meteorological data is corrected and predicted to obtain the physical prediction values ​​of the target meteorological parameters. Based on the physical prediction value at the current moment and the acquired grid meteorological data, a feature vector at the current moment is constructed and input into the trained XGBoost regression residual model to perform residual prediction and obtain the residual prediction value; the physical prediction value and the residual prediction value are added together to obtain the final prediction value of the target meteorological parameter. Based on the final predicted values ​​of the target meteorological parameters, heat balance calculations are performed to determine the capacity margin of the transmission line at the current time step.

2. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 1, characterized in that, For the transmission line area to be regulated, grid meteorological data, tower monitoring meteorological data, and terrain factor data are acquired, spatiotemporally aligned, and the grid meteorological data is interpolated to the tower locations of the transmission line, including the following steps: Align the acquired grid meteorological data, tower monitoring meteorological data, and topographic factor data in terms of time and spatial scale; Based on the geographical coordinates of each tower, the grid meteorological data is interpolated and mapped to the corresponding tower location to obtain the initial meteorological element values ​​at the tower location; The interpolated meteorological data mapped to the corresponding tower locations are time-aligned with the monitoring meteorological data of the corresponding towers.

3. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 1, characterized in that, The physical model is a modified model of the target meteorological parameters set according to the laws of atmospheric physics, including an altitude correction model and a terrain correction model; the target meteorological parameters include wind speed and temperature parameters.

4. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 3, characterized in that, The altitude correction model includes an altitude-based temperature correction model and an altitude-based wind speed correction model; The temperature correction model formula based on altitude can be expressed as: ； in, This refers to the reference temperature in the gridded meteorological data, and the corresponding altitude for the reference temperature data measurement. ; This refers to the actual altitude of the tower. This represents the temperature lapse rate.

5. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 4, characterized in that, The height-based wind speed correction model uses the atmospheric boundary layer wind profile exponential law to correct for wind speed at height, and the formula is as follows: ； in, It is an exponential factor for the wind speed profile, determined by surface roughness and atmospheric stability; The reference wind speed in the grid meteorological data. This indicates the corresponding height at which wind speed and temperature data were measured.

6. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 1, characterized in that, The training process of the XGBoost regression residual model includes the following steps: Based on the obtained physical prediction value, the difference between it and the actual meteorological value measured by the tower monitoring at the corresponding time is calculated to obtain the residual sample used for model training, which is used as the output of the XGBoost regression residual model; Based on the physical predictions and interpolated gridded meteorological data, an input feature vector for the XGBoost regression residual model for residual learning is constructed. The input feature vector includes the original meteorological features, topographic features, temporal features, lag features, physical prior features, and difference features. The input feature vector is used as input, and the residual sample is used as output. The input is fed into the XGBoost regression residual model for training, and the trained XGBoost regression residual model is obtained. Alternatively, the process of constructing a feature vector X from the acquired grid meteorological data and the predicted physical values, and then inputting it into the trained XGBoost regression residual model for residual learning and prediction to obtain the residual prediction value includes the following steps: At any given time t, obtain the grid meteorological data and predicted physical values ​​of the current time t and historical times, and construct the feature vector X of the current time. The input feature vector includes the original meteorological features, terrain features, time features, lag features, physical prior features and difference features. The current feature vector X is input into the trained XGBoost regression residual model to calculate the residual prediction value. ; Alternatively, residual prediction can be performed based on the XGBoost regression residual model, including rolling forecast mode and pure weather forecast mode.

7. The dynamic capacity expansion method for desert transmission lines based on meteorological correction as described in claim 1, characterized in that, Based on the final predicted values ​​of the target meteorological parameters, a heat balance calculation is performed to determine the capacity margin of the transmission line at the current time step, including the following steps: Key environmental parameters affecting transmission line capacity under extreme conditions are obtained from historical meteorological data. Using a pre-built steady-state thermal balance calculation model, the static rated value of transmission line capacity under extreme conditions based on key environmental parameters is calculated. The final predicted value of the target meteorological parameters is used as the real-time key environmental parameter. Based on the pre-built transient thermal balance calculation model, the conductor temperature change under the real-time key environmental parameter is calculated. Based on the conductor temperature change, the dynamic rated value of the transmission line capacity at the current time step is determined. The capacity margin for the current time step is determined by using dynamic and static ratings.

8. A dynamic capacity expansion system for desert transmission lines based on meteorological correction, characterized in that, include: The data acquisition and alignment module is configured to acquire grid meteorological data, tower monitoring meteorological data and terrain factor data, perform spatiotemporal alignment, and interpolate the grid meteorological data to the tower locations of the transmission line; The physical prior prediction module is configured to correct and predict the interpolated grid meteorological data based on a physical model that includes height and terrain correction, and obtain the physical prediction value of the target meteorological parameter. The residual learning module is configured to construct a feature vector based on the physical prediction value at the current moment and the acquired grid meteorological data, and input it into the trained XGBoost regression residual model to perform residual prediction and obtain the residual prediction value. The final predicted value of the target meteorological parameter is obtained by adding the physical prediction value and the residual prediction value. The capacity margin calculation module is configured to perform heat balance calculations based on the final predicted values ​​of the target meteorological parameters to determine the capacity margin of the transmission line at the current time step.

9. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the steps in the dynamic capacity expansion method for the Shagohuang transmission line based on weather correction as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the steps in the dynamic capacity expansion method for the Shagohuang transmission line based on weather correction as described in any one of claims 1-7.

Images