Line current-carrying capacity prediction and risk assessment method and system

Abstract

The invention relates to the technical field of power transmission line state prediction and risk assessment, and discloses a line current-carrying capacity prediction and risk assessment method and system.The method comprises the steps that firstly, a current-carrying capacity prediction model is dynamically constructed by fusing actually-measured weather data and thermal characteristic basic parameters, and the limitation that a traditional BP neural network prediction model only depends on historical data is broken through; then, predicted weather data are input into a current-carrying capacity prediction model for current-carrying capacity prediction, quantile processing is performed according to a current-carrying capacity prediction result, and a current-carrying capacity prediction probability sequence of multiple probability intervals is determined, so that the problem that a static rated value method only outputs a single current-carrying capacity value and cannot quantify a risk level is solved; and finally, performing risk assessment by adopting a plurality of current-carrying capacity prediction probability sequences, and ensuring that the output of the current-carrying capacity probability prediction model is credible through statistical verification, so that the blank that a traditional method lacks system verification on a prediction result is filled, the problem that the traditional method lacks reliability verification is solved, and finally, the safe operation stability of the power transmission line is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of transmission line state prediction and risk assessment, and in particular to a method and system for line current carrying capacity prediction and risk assessment. Background Art

[0002] In modern power system operations and maintenance, precise control of transmission line current capacity is crucial for ensuring reliable power transmission. With the continued growth of renewable energy grid integration and increasingly complex load characteristics, transmission line current capacity is significantly affected by environmental factors and dynamic operating conditions. Accurately predicting current capacity and effectively assessing its risks have become core technical requirements for supporting power dispatch and maintenance decisions.

[0003] At present, the existing current carrying capacity prediction method based on BP neural network relies only on historical data to build models, which cannot adapt to the complex dynamic characteristics of meteorological factors. In addition, the model lacks robustness and generalization, resulting in large prediction errors, making it difficult to support the accurate needs of power system risk assessment. In addition, the existing risk assessment adopts the static rating method, which only outputs a single current carrying capacity value and cannot quantify different risk levels, and cannot effectively ensure the safe operation of transmission lines. Summary of the Invention

[0004] The present invention provides a line current carrying capacity prediction and risk assessment method and system, which solves the technical problem of how to improve the operational stability of power transmission lines.

[0005] A first aspect of the present invention provides a line current carrying capacity prediction and risk assessment method, comprising:

[0006] Obtain measured weather data, forecasted weather data, and basic thermal characteristic parameters of the target line;

[0007] Constructing a current carrying capacity prediction model based on the measured weather data and the basic thermal characteristic parameters;

[0008] The predicted weather data is input into the capacity prediction model to perform capacity prediction, and quantile processing is performed according to the capacity prediction result to determine multiple capacity prediction probability sequences;

[0009] A plurality of the current carrying capacity prediction probability sequences are used to perform risk assessment to obtain a risk assessment result.

[0010] Optionally, the measured weather data is a plurality of measured weather data at current and historical moments, and the constructing of the current carrying capacity prediction model based on the measured weather data and the basic thermal characteristic parameters includes:

[0011] Using each of the measured weather data and the associated thermal characteristic basic parameters to perform observed current carrying capacity calculations to determine each current carrying capacity observation value;

[0012] Using each of the current carrying capacity observation values ​​and the associated measured weather data to input into a preset linear regression model for solution to obtain a plurality of regression coefficients;

[0013] The prediction error of each regression coefficient is calculated, and the regression coefficient with the smallest prediction error is selected to construct a current carrying capacity prediction model.

[0014] Optionally, the measured weather data includes sunshine intensity, ambient temperature and wind speed; the basic thermal characteristic parameters include conductor surface heat absorption coefficient, conductor diameter, conductor surface emissivity, Stephen Boltzmann constant, conductor temperature, air viscosity, conductor surface air density, air thermal conductivity and conductor resistance; and the observed current carrying capacity calculation is specifically as follows:

[0015] Determining solar radiation heat absorption data using the sunlight intensity, the conductor surface heat absorption coefficient, and the conductor diameter;

[0016] Determining radiation heat dissipation data using the wire diameter, the wire surface emissivity, the Stephen-Boltzmann constant, the ambient temperature, and the wire temperature;

[0017] Determining a Reynolds number using the conductor diameter, the wind speed, the air viscosity coefficient, and the air density on the conductor surface;

[0018] comparing the wind speed with a preset wind speed test threshold;

[0019] Based on the comparison result, the Reynolds number, the ambient temperature, the wire temperature and the air thermal conductivity are used to determine forced convection heat dissipation data;

[0020] The solar radiation heat absorption data, the radiation heat dissipation data, the forced convection heat dissipation data and the conductor resistance value are used to determine an observed current carrying capacity value.

[0021] Optionally, determining the forced convection heat dissipation data based on the comparison result by using the Reynolds number, the ambient temperature, the wire temperature, and the air thermal conductivity includes:

[0022] When the wind speed is less than a preset wind speed test threshold, the Reynolds number, the ambient temperature, the conductor temperature, and the air thermal conductivity are used to input a first preset forced convection heat dissipation function to obtain forced convection heat dissipation data;

[0023] When the wind speed is greater than or equal to the preset wind speed test threshold, the Reynolds number, the ambient temperature, the wire temperature and the air thermal conductivity are used to input a second preset forced convection heat dissipation function to obtain forced convection heat dissipation data.

[0024] Optionally, the predicted weather data is forecast weather data for multiple moments in a preset future time period obtained by predicting a preset mesoscale weather forecast model. The predicted weather data is input into the capacity prediction model to perform capacity prediction, and quantile processing is performed based on the capacity prediction result to determine multiple capacity prediction probability sequences, including:

[0025] Using each of the forecast weather data to input the corresponding capacity prediction model to obtain a plurality of capacity prediction values;

[0026] Performing interval quantile operation on the plurality of the current carrying capacity prediction values ​​to obtain a plurality of quantile current carrying capacity prediction value sequences;

[0027] Each of the quantile current carrying capacity prediction value sequences is input into a preset probability prediction model for solution to obtain a plurality of current carrying capacity prediction probability sequences.

[0028] Optionally, performing interval quantile operation on the plurality of the current carrying capacity prediction values ​​to obtain a plurality of quantile current carrying capacity prediction value sequences includes:

[0029] Selecting a maximum value and a minimum value from the plurality of predicted current carrying capacity values, and determining boundary values ​​of a preset width interval based on the maximum value and the minimum value to obtain a target width interval;

[0030] Dividing the target width interval into intervals to obtain a plurality of fixed width intervals;

[0031] Dividing the plurality of predicted current carrying capacity values ​​into corresponding fixed width intervals, and sorting the predicted current carrying capacity values ​​within each fixed width interval in ascending order;

[0032] Calculate the predicted quantile current carrying capacity values ​​corresponding to multiple quantiles in each fixed width interval after ascending sorting;

[0033] A plurality of the quantile current carrying capacity prediction values ​​are counted to obtain a quantile current carrying capacity prediction value sequence corresponding to each of the sorted fixed-width intervals.

[0034] Optionally, the capacity prediction probability sequence includes a capacity prediction probability value corresponding to each quantile, and the risk assessment is performed using multiple capacity prediction probability sequences to obtain a risk assessment result, including:

[0035] Performing a reliability assessment using each of the current carrying capacity prediction probability values ​​to obtain a reliability assessment result;

[0036] Performing line capacity utilization analysis using each of the current carrying capacity prediction probability values ​​to obtain a line capacity utilization analysis result;

[0037] The risk assessment result includes the reliability assessment result and the line capacity utilization analysis result.

[0038] Optionally, the reliability assessment is performed using each of the current carrying capacity prediction probability values ​​to obtain a reliability assessment result, including:

[0039] Extracting the current carrying capacity prediction probability value corresponding to the same quantile from the plurality of current carrying capacity prediction probability sequences, and obtaining target prediction probability sequences corresponding to a plurality of different quantile types;

[0040] The target prediction probability sequence includes a plurality of current carrying capacity prediction probability values ​​corresponding to the same quantile;

[0041] Calculating the reliability index corresponding to each target prediction probability sequence;

[0042] Performing difference calculations between each of the reliability indicators and a preset over-limit frequency coefficient to obtain a plurality of over-limit frequencies;

[0043] performing difference operations on each of the over-limit frequencies and the associated quantiles to obtain a plurality of target deviations;

[0044] Comparing each of the target deviations with a preset tolerance threshold;

[0045] When the target deviation is less than or equal to the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is reliable;

[0046] When the target deviation is greater than the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is unreliable.

[0047] Optionally, performing line capacity utilization analysis using each of the current carrying capacity prediction probability values ​​to obtain a line capacity utilization analysis result includes:

[0048] Select the largest quantile as the target quantile;

[0049] Extracting the current carrying capacity prediction probability value corresponding to the target quantile from the plurality of current carrying capacity prediction probability sequences as the target current carrying capacity prediction probability value;

[0050] Calculating a prediction ratio of each target current carrying capacity prediction probability value;

[0051] ranking each of the predicted ratios and determining a median predicted ratio;

[0052] comparing the median predicted ratio with a preset standard predicted ratio threshold;

[0053] When the median prediction ratio is greater than or equal to the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model meets the line capacity utilization qualification requirement;

[0054] When the median prediction ratio is less than the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model does not meet the line capacity utilization qualification requirement.

[0055] A second aspect of the present invention provides a line current carrying capacity prediction and risk assessment system, comprising:

[0056] An acquisition module is used to obtain measured weather data, predicted weather data, and basic thermal characteristic parameters of the target line;

[0057] A construction module, configured to construct a current carrying capacity prediction model based on the measured weather data and the basic thermal characteristic parameters;

[0058] A prediction module, configured to use the predicted weather data to input the capacity prediction model to perform capacity prediction, and perform quantile processing based on the capacity prediction results to determine multiple capacity prediction probability sequences;

[0059] An evaluation module is used to perform risk evaluation using a plurality of the current carrying capacity prediction probability sequences to obtain a risk evaluation result.

[0060] It can be seen from the above technical solutions that the present invention has the following advantages:

[0061] In the present invention, firstly, a capacity prediction model is dynamically constructed by integrating measured weather data and basic thermal characteristic parameters, breaking through the limitation of traditional BP neural network prediction model that only relies on historical data. Then, the predicted weather data is input into the capacity prediction model to predict the capacity. The capacity prediction result is subjected to quantile processing to determine the capacity prediction probability sequence of multiple probability intervals, which solves the problem that the static rating method only outputs a single capacity value and cannot quantify the risk level. Finally, multiple capacity prediction probability sequences are used for risk assessment, and the output of the capacity probability prediction model is ensured to be credible through statistical verification, which fills the gap of the traditional method in lacking systematic verification of the prediction results and solves the problem of lack of reliability verification in the traditional method, thereby ultimately improving the safe operation stability of the transmission line. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 A flowchart of the steps of a line current carrying capacity prediction and risk assessment method provided in Example 1 of the present invention;

[0064] Figure 2 A flowchart of the steps of a line current carrying capacity prediction and risk assessment method provided in the second embodiment of the present invention;

[0065] Figure 3 This is a structural block diagram of a line current carrying capacity prediction and risk assessment system provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0066] The embodiments of the present invention provide a line current carrying capacity prediction and risk assessment method and system for solving the technical problem of how to improve the operational stability of power transmission lines.

[0067] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only 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 making creative work are within the scope of protection of the present invention.

[0068] In many power grids, the original design capacity of lines is no longer sufficient to meet current transmission demands. Thermal limitations are particularly prominent on short-distance, high-load lines. The thermal capacity of overhead lines is significantly affected by weather conditions (such as ambient temperature, solar radiation, and wind speed). When weather conditions are unfavorable, line conductor temperatures may exceed their maximum allowable conductor temperature (MACT), leading to line overload and even safety incidents. To address these challenges, traditional power grid management methods often use conservative static line ratings (SLRs) and point prediction methods to assess line current carrying capacity. The static line rating method is a traditional method for assessing line current carrying capacity. It primarily sets a fixed current carrying capacity limit based on line design parameters such as conductor type and maximum allowable conductor temperature. This limit represents the maximum current that the line can carry under safe operating conditions. The point prediction method is a more refined forecasting method that uses specific values ​​of weather variables (such as wind speed and temperature) to predict the current carrying capacity of overhead lines. This method is usually based on historical data and statistical models, which provide a single numerical prediction result by analyzing the relationship between weather variables and line current carrying capacity.

[0069] The static load rating (SLR) method sets a fixed ampacity limit based on the line's design parameters. This method is overly conservative, fails to consider the impact of real-time weather conditions on the line's thermal capacity, and fails to fully utilize the line's ampacity potential under favorable weather conditions. Point prediction methods predict ampacity by predicting specific values ​​of weather variables, but they typically only provide a single numerical prediction and cannot quantify the uncertainty of the prediction.

[0070] This invention proposes a method and system for predicting and assessing line current carrying capacity. This method combines predicted weather data from a mesoscale weather forecast model with basic thermal characteristic parameters, and uses statistical methods to adapt the predicted weather data to the line span scale, thereby accurately predicting the current carrying capacity of overhead lines. This method also provides a reliability and line capacity utilization assessment scheme, helping to more effectively utilize line resources while ensuring safety, thereby improving the transmission efficiency and reliability of overhead transmission lines.

[0071] See also Figure 1 , Figure 1 This is a flowchart of the steps of a line current carrying capacity prediction and risk assessment method provided in Example 1 of the present invention.

[0072] The present invention provides a line current carrying capacity prediction and risk assessment method, comprising:

[0073] Step 101: Obtain measured weather data, predicted weather data, and basic thermal characteristic parameters of the target line.

[0074] The target line refers to the overhead transmission line for current carrying capacity prediction and risk assessment.

[0075] Measured weather data refers to meteorological parameters collected in real time or historically recorded from meteorological monitoring equipment (such as weather stations and sensors) around the target line. Measured weather data consists of multiple weather data points measured at current and historical times. Measured weather data refers to environmental parameters of the target line collected by sensors, including but not limited to sunshine intensity, ambient temperature, and wind speed. These data serve as dynamic inputs for current carrying capacity calculations.

[0076] Forecasted weather data, which is forecasted weather data for multiple moments in a preset future time period, obtained through predictions by a preset mesoscale weather forecast model. The multiple moments in the preset time period can be the 24 hours (moments) of tomorrow (time period), or 9:00 (moments) every day in the next 30 days (time period), etc., and there is no restriction here.

[0077] Basic thermal characteristic parameters refer to the inherent parameters that describe the thermal physical properties of conductors. They include but are not limited to the conductor surface heat absorption coefficient, diameter, emissivity, Stefan-Boltzmann constant, conductor temperature, air viscosity, surface air density, air thermal conductivity, and conductor resistance. They are the physical constraints for current-carrying capacity calculations.

[0078] In an embodiment of the present invention, measured weather data, predicted weather data, and basic thermal characteristic parameters of an overhead transmission line for current carrying capacity prediction and risk assessment are obtained.

[0079] Step 102: Construct a current carrying capacity prediction model based on measured weather data and basic thermal characteristic parameters.

[0080] The current carrying capacity prediction model refers to a mathematical model constructed through a linear regression algorithm based on measured weather data and basic thermal characteristic parameters. It is used to deduce the mapping relationship between input (predicted weather data) and output (predicted current carrying capacity value).

[0081] In an embodiment of the present invention, a current carrying capacity prediction model is constructed based on measured weather data and basic thermal characteristic parameters.

[0082] Step 103: Use the predicted weather data to input the capacity prediction model to perform capacity prediction, and perform quantile processing based on the capacity prediction results to determine multiple capacity prediction probability sequences.

[0083] Current carrying capacity prediction refers to the process of using a constructed current carrying capacity prediction model, taking predicted weather data as input, and using the model's algorithm to deduce and output the current value that the target line may carry within a specific period in the future.

[0084] Quantile processing refers to the statistical analysis of the capacity prediction results (capacity prediction values). By setting different quantiles (such as 0.5%, 1%, 2.5%, ..., 50%, ..., 90%, etc.), the capacity prediction values ​​are divided into multiple intervals according to the probability distribution, and solved through a preset probability prediction model to obtain a capacity prediction probability sequence reflecting different probability levels.

[0085] The current carrying capacity prediction result refers to the current carrying capacity prediction value output by the current carrying capacity prediction model.

[0086] The capacity prediction probability sequence refers to the sequence of capacity prediction results corresponding to different quantiles after the capacity prediction results (i.e., the capacity prediction values) are quantile processed, reflecting the probability distribution of the capacity falling within a certain range.

[0087] In an embodiment of the present invention, predicted weather data is input into a capacity prediction model to perform capacity prediction, and the capacity prediction results (capacity prediction values) are statistically analyzed. By setting different quantiles, the capacity prediction values ​​are divided into multiple intervals according to the probability distribution, and are solved by a preset probability prediction model to obtain a capacity prediction probability sequence reflecting different probability levels.

[0088] Step 104: Perform risk assessment using multiple current carrying capacity prediction probability sequences to obtain risk assessment results.

[0089] Risk assessment refers to the process of quantitatively evaluating the safety risks of the target line in the future period through reliability verification and capacity utilization analysis, based on multiple current-carrying capacity prediction probability sequences, combined with line operation safety standards (such as maximum allowable current-carrying capacity, preset risk thresholds, etc.).

[0090] The risk assessment result refers to the quantitative conclusion on the future operation risk of the target line, including the reliability assessment result and capacity utilization analysis result. The reliability assessment result refers to whether the current carrying capacity prediction value output by the model is credible, and the capacity utilization analysis result refers to whether the line capacity utilization is qualified.

[0091] In the embodiment of the present invention, multiple current carrying capacity prediction probability sequences are used to perform reliability assessment and capacity utilization analysis results to obtain risk assessment results.

[0092] In the present invention, firstly, a capacity prediction model is dynamically constructed by integrating measured weather data and basic thermal characteristic parameters, breaking through the limitation of traditional BP neural network prediction model that only relies on historical data. Then, the predicted weather data is input into the capacity prediction model to predict the capacity. The capacity prediction result is subjected to quantile processing to determine the capacity prediction probability sequence of multiple probability intervals, which solves the problem that the static rating method only outputs a single capacity value and cannot quantify the risk level. Finally, multiple capacity prediction probability sequences are used for risk assessment, and the output of the capacity probability prediction model is ensured to be credible through statistical verification, which fills the gap of the traditional method in lacking systematic verification of the prediction results and solves the problem of lack of reliability verification in the traditional method, thereby ultimately improving the safe operation stability of the transmission line.

[0093] See also Figure 2 , Figure 2 This is a flowchart of the steps of a line current carrying capacity prediction and risk assessment method provided in Example 2 of the present invention.

[0094] The present invention provides a line current carrying capacity prediction and risk assessment method, comprising:

[0095] Step 201: Obtain measured weather data, predicted weather data, and basic thermal characteristic parameters of the target line.

[0096] In an embodiment of the present invention, measured weather data and basic thermal characteristic parameters along the target route, as well as predicted weather data obtained by predicting a preset mesoscale weather forecast model, are collected. The collected measured weather data are preprocessed. In the data cleaning stage, outliers (such as outliers caused by sensor failures) need to be eliminated, and missing data are filled by linear interpolation.

[0097] Step 202: Construct a current carrying capacity prediction model based on measured weather data and basic thermal characteristic parameters.

[0098] It is worth mentioning that the present invention constructs a current carrying capacity prediction model by combining the heat balance equation with the linear regression model, as follows:

[0099] The measured weather data is substituted into the heat balance equation, combined with the basic parameters of thermal characteristics, to calculate the current carrying capacity observation value, and the historical current carrying capacity observation value and the corresponding measured weather data are input into the linear regression model to solve the regression coefficient. By comparing the prediction errors of different regression coefficient combinations (such as the root mean square error (MSE)), the coefficient group with the smallest error is selected to finally determine the current carrying capacity prediction model. Compared with the traditional static rating method that does not consider real-time meteorological changes, the current carrying capacity model of the present invention is based on the physical laws of the heat balance equation and fits the actual data through linear regression, avoiding the parameter errors of pure physical models and the lack of generalization of pure data models. At the same time, the model is updated in real time using measured weather data to adapt to the differences in meteorological characteristics in different regions and seasons, thereby improving prediction accuracy.

[0100] Furthermore, the measured weather data is a plurality of measured weather data at the current and historical moments, and step 202 may include the following sub-steps:

[0101] S11. Calculate the observed current carrying capacity using the measured weather data and the associated thermal characteristic basic parameters to determine the observed current carrying capacity values.

[0102] Furthermore, the measured weather data includes sunshine intensity, ambient temperature and wind speed. The basic thermal characteristic parameters include the conductor surface heat absorption coefficient, conductor diameter, conductor surface emissivity, Stephen Boltzmann constant, conductor temperature, air viscosity, conductor surface air density, air thermal conductivity and conductor resistance. The specific calculation of the observed current carrying capacity is as follows:

[0103] A11. Determine solar radiation heat absorption data using sunlight intensity, conductor surface heat absorption coefficient, and conductor diameter.

[0104] In a specific implementation, in order to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of solar radiation heat absorption data can be as follows:

[0105]

[0106] Where, represents solar radiation heat absorption data, Indicates the wire diameter, Indicates the heat absorption coefficient of the conductor surface, specifically the heat absorption coefficient of the conductor surface, which is generally equal to the radiation coefficient of the conductor surface, with a typical value of 0.5. Indicates the intensity of sunlight, .

[0107] A12. Determine radiation heat dissipation data using conductor diameter, conductor surface emissivity, Stephen-Boltzmann constant, ambient temperature, and conductor temperature.

[0108] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the radiation heat dissipation data can be as follows:

[0109]

[0110] Where, Represents radiation heat dissipation data, It represents the surface radiation coefficient of the conductor. The value of new bright conductor is 0.23~0.43, and the value of old conductor or conductor coated with black preservative is 0.90~0.95. represents the Stephen Boltzmann constant, , Indicates the conductor temperature, specifically the conductor temperature itself, Indicates the ambient temperature, specifically the temperature of the environment where the conductor is located.

[0111] It should be noted that the surface of a shiny new conductor is smooth and contains fewer impurities, with a surface emissivity ranging from 0.23 to 0.43. Old conductors, however, may be oxidized and contaminated due to long-term use, or coated with black preservatives, and have a surface emissivity ranging from 0.90 to 0.95. Since shiny and old conductors have different radiation and heat dissipation characteristics, not distinguishing between their emissivity values ​​when constructing a current-carrying capacity prediction model can increase the model's prediction error. Therefore, distinguishing between the emissivity values ​​of shiny and old conductors can improve the accuracy of the current-carrying capacity prediction model, ensuring the safe and reliable operation of transmission lines.

[0112] A13. Determine the Reynolds number using the conductor diameter, wind speed, air viscosity, and air density on the conductor surface.

[0113] It should be noted that convective heat loss is related to air flow conditions and can be represented by the Reynolds number, a dimensionless number used to characterize fluid flow conditions.

[0114] In a specific implementation, in order to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the Reynolds number can be as follows:

[0115]

[0116] Where, represents the Reynolds number, Indicates wind speed, represents the air viscosity coefficient, Indicates the air density on the surface of the conductor, specifically the actual density of the air around the conductor.

[0117] A14. Compare wind speed with preset wind speed test threshold;

[0118] It should be noted that at low wind speeds, air flowing over the conductor surface tends to form a laminar boundary layer (ordered flow of fluid particles), leading to slow diffusion as heat transfer, a low convective heat transfer coefficient, and stable heat dissipation efficiency. At high wind speeds, airflow disturbances intensify, and the boundary layer becomes turbulent (disordered mixing of particles), allowing heat to be rapidly "entrained," significantly increasing the convective heat transfer coefficient and significantly enhancing heat dissipation efficiency. This indicates that wind speed dominates flow patterns and influences heat dissipation. Using a single function to cover all wind speeds would result in significant deviations in heat dissipation calculations due to mismatched flow patterns. The core logic of current carrying capacity is "thermal balance" (heat generation = heat dissipation), and accurate heat dissipation calculation is fundamental. Therefore, the present invention selects appropriate forced convection heat dissipation functions based on different wind speed scenarios, enabling accurate calculation of forced convection heat dissipation data. This improves the accuracy of current carrying capacity predictions, provides a more reliable foundation for risk assessment, and represents a key step in moving current carrying capacity predictions from "rough estimation" to "precise quantification."

[0119] The preset wind speed test threshold refers to a judgment value used to distinguish whether the current scene of the target line is a low wind speed scene or a high wind speed scene. It can be determined through simulation experiments or engineering experience, and is preferably 10m / s.

[0120] In an embodiment of the present invention, the wind speed is compared with a preset wind speed test threshold.

[0121] A15. Based on the comparison results, the Reynolds number, ambient temperature, conductor temperature, and air thermal conductivity are used to determine the forced convection heat dissipation data.

[0122] Furthermore, A15 may include the following sub-steps:

[0123] A151. When the wind speed is less than a preset wind speed test threshold, the Reynolds number, the ambient temperature, the conductor temperature, and the air thermal conductivity are used to input a first preset forced convection heat dissipation function to obtain forced convection heat dissipation data.

[0124] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, wherein the expression of the first preset forced convection heat dissipation function can be as follows:

[0125]

[0126] Where, Indicates the forced convection heat dissipation data calculated for low wind speed scenarios. Indicates the thermal conductivity of air, which is related to the temperature of the conductor itself and the ambient temperature.

[0127] A152. When the wind speed is greater than or equal to the preset wind speed test threshold, the Reynolds number, ambient temperature, conductor temperature and air thermal conductivity are used to input the second preset forced convection heat dissipation function to obtain forced convection heat dissipation data.

[0128] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, wherein the expression of the second preset forced convection heat dissipation function can be as follows:

[0129]

[0130] Where, Indicates the forced convection heat dissipation data calculated for high wind speed scenarios.

[0131] A16. Use solar radiation heat absorption data, radiation heat dissipation data, forced convection heat dissipation data and conductor resistance values ​​to determine the observed current carrying capacity.

[0132] In a specific implementation, in order to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the current carrying capacity observation value can be as follows:

[0133]

[0134] Where, Indicates the observed value of current carrying capacity, Indicates the resistance value of the wire, specifically the resistance value corresponding to the wire temperature.

[0135] It should be noted that steps A11-A16 are the process of performing observation current-carrying calculations on a single measured weather data and an associated basic thermal characteristic parameter to determine the observed current-carrying value. Each time the observed current-carrying value calculation is performed, only one measured weather data and one associated basic thermal characteristic parameter need to be used to perform the calculation to obtain the observed current-carrying value.

[0136] S12. Input each current carrying capacity observation value and the associated measured weather data into a preset linear regression model to solve the problem and obtain multiple regression coefficients.

[0137] The preset linear regression model refers to a pre-set linear relationship model that fits the "capacity observation value-measured weather data" through the least squares algorithm.

[0138] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the preset linear regression model can be as follows:

[0139]

[0140] Where, represents the regression coefficient, Represents the input features, which are measured weather data in this step, specifically The measured weather data at the time environmental parameters, Indicates the total number of environment parameters.

[0141] In an embodiment of the present invention, each set of current carrying capacity observation values ​​and the associated measured weather data are input into a preset linear regression model, and the least squares method is used to solve the model to obtain multiple sets of regression coefficients.

[0142] S13. Calculate the prediction error of each regression coefficient, and select the regression coefficient with the smallest prediction error to construct a current carrying capacity prediction model.

[0143] In an embodiment of the present invention, the prediction error of each regression coefficient is calculated. The prediction error can use the root mean square error (MSE) and adopt ten-fold cross validation with the goal of minimizing the root mean square error. The regression coefficient with the smallest root mean square error is selected as the basic parameter for constructing the current carrying capacity prediction model.

[0144] In a specific implementation, in order to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, wherein the expression of the current carrying capacity prediction model can be as follows:

[0145]

[0146] Where, Indicates that at the current moment For the future The predicted current carrying capacity, Represents input features, specifically predicted weather data, when performing subsequent current carrying capacity prediction.

[0147] Step 203: Use the predicted weather data to input the capacity prediction model to perform capacity prediction, and perform quantile processing based on the capacity prediction results to determine multiple capacity prediction probability sequences.

[0148] Furthermore, the predicted weather data is the forecast weather data for multiple moments in a preset future time period predicted by a preset mesoscale weather forecast model. Step 203 may include the following sub-steps:

[0149] The preset mesoscale weather forecast model refers to a pre-set type of numerical forecast tool used to simulate and predict mesoscale weather systems. Its core is to simulate atmospheric movement and weather evolution by solving the atmospheric dynamic-thermodynamic equations and combining physical parameterization schemes, and can output forecasts of meteorological elements for the next few days.

[0150] In its specific implementation, the mesoscale weather forecast model first divides the Earth's atmosphere into three-dimensional grids according to longitude, latitude and altitude. Each grid stores initial meteorological variables (temperature, humidity, wind speed, air pressure, etc.). Then, numerical methods are used to solve the nonlinear partial differential equations describing atmospheric motion at each grid point, including equations such as conservation of mass, conservation of momentum, conservation of thermodynamic energy and changes in the physical state of water. Combined with parameterization schemes (such as the cumulus convection parameterization scheme that uses empirical formulas to simulate the impact of thunderstorms and convective clouds on the weather, the radiation transfer parameterization scheme that calculates the transmission of solar radiation and long-wave radiation in the atmosphere, and the boundary layer parameterization that simulates the energy and momentum exchange between the near-ground atmosphere and the surface), the convection, radiation and surface boundary processes are modeled to output meteorological prediction variables such as wind speed, temperature, and radiation for multiple time periods in the future.

[0151] S21. Using each forecast weather data input corresponding to the current carrying capacity prediction model to obtain multiple current carrying capacity prediction values.

[0152] In the embodiment of the present invention, each forecast weather data is input into a corresponding current carrying capacity prediction model to obtain a plurality of current carrying capacity prediction values.

[0153] S22. Perform interval quantile operation on the multiple predicted current carrying capacity values ​​to obtain multiple quantile current carrying capacity prediction value sequences.

[0154] The interval quantile operation refers to the process of "interval division-sorting-quantile calculation" for the carrying capacity prediction value, which converts the carrying capacity prediction values ​​at consecutive moments into quantile carrying capacity prediction values.

[0155] Furthermore, S22 may include the following sub-steps:

[0156] S221. Select a maximum value and a minimum value from a plurality of current carrying capacity prediction values, and determine boundary values ​​of a preset width interval based on the maximum value and the minimum value to obtain a target width interval.

[0157] In an embodiment of the present invention, for example, the maximum value of the predicted current carrying capacity is 93A, and the minimum value of the predicted current carrying capacity is 25A. Then the upper boundary value of the preset width interval is 100A, and the lower boundary value of the preset width interval is 20A. Therefore, the target width interval obtained is [20, 100].

[0158] S222: Divide the target width interval into intervals to obtain multiple fixed width intervals.

[0159] In an embodiment of the present invention, the target width interval is divided into intervals according to a preset division interval (eg, 10A) to obtain a plurality of fixed width intervals, such as [20, 30), [30, 40), and the like.

[0160] S223 , dividing the multiple current carrying capacity prediction values ​​into corresponding fixed-width intervals, and sorting the current carrying capacity prediction values ​​in each fixed-width interval in ascending order.

[0161] In an embodiment of the present invention, the predicted current carrying capacity values ​​are divided into fixed-width intervals, and the predicted current carrying capacity values ​​in each fixed-width interval are sorted from small to large, for example, the predicted current carrying capacity values ​​of the interval [20, 30) A (for example, 10: 20.5, 21.2, 22.1, 23.5, 24.7, 25.3, 26.8, 27.4, 28.9, 29.1).

[0162] S224. Calculate the predicted quantile current carrying capacity values ​​corresponding to multiple quantiles in the fixed-width intervals after ascending sorting.

[0163] In the embodiment of the present invention, each fixed width interval after ascending sorting is correspondingly provided with a plurality of quantiles (such as =0.5%, 1%, 2.5%, ..., 50%, ..., 90%, etc.), for example, calculate = 1%, a total of 10 (i.e. n) current carrying capacity prediction values, position , substituted into the calculation to get 1.09, indicating that the quantile is between the first and second predicted values ​​of current carrying capacity, and is calculated by linear interpolation. , where Indicates the first current carrying capacity prediction value of the calculated sorting position. The first current carrying capacity prediction value of the calculated sorting position is obtained. =1% of the predicted value of the quantile current carrying capacity .

[0164] It should be noted that when the position is calculated as an integer, It represents the predicted current carrying capacity value corresponding to the calculated row position. When the calculated position is not an integer, Indicates the calculated first current carrying capacity prediction value of the preceding sort position.

[0165] S225. Count multiple quantile current carrying capacity prediction values ​​to obtain a quantile current carrying capacity prediction value sequence corresponding to each sorted fixed-width interval.

[0166] In the embodiment of the present invention, a plurality of quantile current carrying capacity prediction values ​​are counted to obtain a quantile current carrying capacity prediction value sequence corresponding to each sorted fixed width interval, and each sorted fixed width interval corresponds to a quantile current carrying capacity prediction value sequence. .

[0167] S23. Input each quantile current carrying capacity prediction value sequence into a preset probability prediction model for solving, thereby obtaining multiple current carrying capacity prediction probability sequences.

[0168] The preset probability prediction model refers to a pre-set sequence of quantile capacity prediction values, which is again fitted through linear regression to obtain the mapping equation of "quantile capacity prediction value - capacity prediction probability value".

[0169] The preset probability prediction model is specifically:

[0170]

[0171] Where, Indicates the predicted probability value of current carrying capacity, represents the quantile regression coefficient, which can be determined by minimizing the quantile loss.

[0172] In the embodiment of the present invention, each quantile current carrying capacity prediction value sequence is input into the preset probability prediction model for solution to obtain multiple current carrying capacity prediction probability sequences. Each sorted fixed width interval corresponds to a current carrying capacity prediction probability sequence. .

[0173] It should be noted that, assuming A is 4 and B is 1, Substitute into the preset probability prediction model and get , which means that at time t+1 in the future, there is a 1% probability that the actual current carrying capacity will be less than or equal to 25A, and a 99% probability that the actual current carrying capacity will be greater than 25A. Substitute into the preset probability prediction model and get , which indicates that at time t+1 in the future, there is a 90% probability that the actual current carrying capacity will be less than or equal to 35A, and a 10% probability that the actual current carrying capacity will be greater than 35A.

[0174] It's worth noting that in step 203, through interval partitioning and quantile calculation, current capacity predictions are fitted under different probabilities, covering both low-probability quantiles (e.g., 1% and 2.5%) corresponding to "extremely conservative scenarios" and high-probability quantiles (e.g., 90%) corresponding to "high-frequency current-carrying scenarios." Accurately predicting high-probability quantiles (scenarios nearing overload) can provide early warning of line overheating risks. This invention, through conditional quantiles and regression models, can more accurately predict low-probability quantiles, thereby reducing the risk of line overheating.

[0175] It is worth mentioning that the quantile processing obtains multiple probability sequences of current carrying capacity prediction, covering different probability scenarios in the long future span (such as the next 24 hours, 72 hours, etc.), supporting dynamic scheduling decisions.

[0176] Furthermore, the current carrying capacity prediction probability sequence includes the current carrying capacity prediction probability value corresponding to each quantile,

[0177] The risk assessment results include reliability assessment results and line capacity utilization analysis results.

[0178] Step 204: Perform reliability assessment using each current carrying capacity prediction probability value to obtain a reliability assessment result.

[0179] Furthermore, step 204 may include the following sub-steps:

[0180] S31. Extracting the current carrying capacity prediction probability value corresponding to the same quantile from multiple current carrying capacity prediction probability sequences to obtain target prediction probability sequences corresponding to multiple different quantile types;

[0181] The target prediction probability sequence includes multiple current carrying capacity prediction probability values ​​corresponding to the same quantile.

[0182] In the embodiment of the present invention, since each sorted fixed-width interval corresponds to a capacity prediction probability sequence, each quantile capacity prediction value sequence is input into the preset probability prediction model for solution to obtain multiple capacity prediction probability sequences, and the capacity prediction probability value corresponding to the same quantile is extracted for each capacity prediction probability sequence. For example, taking the quantile 0.5% as an example, the capacity prediction probability value associated with the quantile 0.5% in each capacity prediction probability sequence is extracted. And integrate to obtain the target prediction probability sequence with a quantile of 0.5%, and the quantiles of 1%, 2.5%, ..., 50%, ..., 90% are also extracted accordingly, so that the target prediction probability sequence of all quantiles can be obtained.

[0183] S32. Calculate the reliability index corresponding to each target prediction probability sequence.

[0184] In an embodiment of the present invention, a binary indicator function is first used to determine whether a single predicted probability value causes the temperature to exceed the standard. If the predicted probability value is greater than the associated preset standard threshold, it is considered to exceed the standard and is recorded as 1; otherwise, it is considered to not exceed the standard and is recorded as 0.

[0185] The binary indicator function is specifically:

[0186]

[0187] Where, represents the binary indicator function value, Indicates the preset standard threshold.

[0188] Then, multiple binary indicator function values ​​in each target prediction probability sequence are counted, and the reliability index corresponding to each target prediction probability sequence is calculated through the reliability index function;

[0189] The reliability index function is specifically:

[0190]

[0191] Where, Represents the number of binary indicator function values.

[0192] It should be noted that for each constructed quantile (e.g., 0.5%, 1%, 2.5%, ..., 50%, ..., 90%), the corresponding reliability index is independently evaluated. The reliability index is used to determine whether the model accurately covers the specified risk level.

[0193] S33. Perform difference calculations on each reliability index and a preset over-limit frequency coefficient to obtain a plurality of over-limit frequencies.

[0194] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the over-limit frequency can be as follows:

[0195]

[0196] S34. Perform difference calculations on each over-limit frequency and the associated quantile to obtain multiple target deviations.

[0197] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the over-limit frequency can be as follows:

[0198]

[0199] S35. Compare each target deviation with a preset allowable deviation threshold.

[0200] In the embodiment of the present invention, each target deviation is compared with a preset allowable deviation threshold.

[0201] S36. When the target deviation is less than or equal to the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is reliable.

[0202] In this embodiment of the present invention, if the target deviation does not exceed the preset allowable deviation threshold, , then it is considered that the preset probability prediction model is at the quantile The predicted probability value of the current carrying capacity under the condition is reliable, which can be understood as the preset probability prediction model at the quantile The following is reliable.

[0203] S37. When the target deviation is greater than the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is unreliable.

[0204] In an embodiment of the present invention, if the target deviation exceeds the preset allowable deviation threshold, , then it is considered that the preset probability prediction model is at the quantile The predicted probability value of current carrying capacity under the condition is unreliable, which can be understood as the preset probability prediction model is not reliable at the quantile point. The following is unreliable.

[0205] Step 205: Use the predicted probability values ​​of each current carrying capacity to perform line capacity utilization analysis to obtain a line capacity utilization analysis result.

[0206] It should be noted that the line capacity utilization rate refers to the "ratio of actual current carrying capacity to the maximum allowable current carrying capacity of the line", but directly using "actual current carrying capacity / maximum allowable current carrying capacity" for evaluation is flawed, because the actual current carrying capacity is ex post data and cannot be decided in advance. Therefore, the "probability distribution of predicted values" is introduced to assist in judgment, and the "probability distribution of predicted values" is used to indirectly evaluate the "rationality of actual capacity utilization rate". The more accurate the model prediction (the higher the prediction ratio), the more controllable the line capacity utilization.

[0207] Furthermore, step 205 may include the following sub-steps:

[0208] S41. Select the largest quantile as the target quantile.

[0209] In an embodiment of the present invention, among the quantile sets of 0.5%, 1%, 2.5%, ..., 50%, ..., 90%, the 90% quantile is preferentially selected as the core analysis object. The 90% quantile corresponds to the "high-frequency high-current carrying capacity interval", that is, it characterizes the high-probability current carrying scenario, and can accurately characterize the operating risk of the target line approaching the overload threshold. Therefore, the largest valid quantile is selected as the target quantile.

[0210] S42. Extracting the current carrying capacity prediction probability value corresponding to the target quantile from the plurality of current carrying capacity prediction probability sequences as the target current carrying capacity prediction probability value.

[0211] In the embodiment of the present invention, the current carrying capacity prediction probability value corresponding to the target quantile is extracted from multiple current carrying capacity prediction probability sequences as the target current carrying capacity prediction probability value, that is, the current carrying capacity prediction probability value associated with the quantile of 90% in each current carrying capacity prediction probability sequence is extracted. As the predicted probability value of target current carrying capacity.

[0212] S43. Calculate the prediction ratio of each target current carrying capacity prediction probability value.

[0213] In an embodiment of the present invention, a ratio operation is performed between each target current carrying capacity prediction probability value and the associated measured current carrying capacity to obtain multiple prediction ratios, wherein the calculation of the measured current carrying capacity is the same as the process of calculating the current carrying capacity observation value in step 202, which will not be repeated here.

[0214] In a specific implementation, to facilitate the implementation of the method, the above process can be converted into a formula encapsulation form, where the expression of the predicted ratio can be as follows:

[0215]

[0216] Where, Indicates the measured current carrying capacity at time t+h.

[0217] S44. Sort the predicted ratios and determine the median predicted ratio.

[0218] In this embodiment of the present invention, it is assumed that multiple predicted ratios are calculated to be 106.67, 112.00, 97.22, 105.26, 104.17, 93.75, 105.88, 96.67, 106.45, and 103.33, and they are sorted to obtain the sorted predicted ratios: 93.75, 96.67, 97.22, 103.33, 104.17, 105.26, 105.88, 106.45, 106.67, and 112.00. Since the total number is an even number, the predicted ratios of the 5th (104.17) and the 6th (105.26) are selected and averaged to obtain a median predicted ratio of approximately 104.72%.

[0219] S45. Compare the median prediction ratio with a preset standard prediction ratio threshold.

[0220] In an embodiment of the present invention, the median prediction ratio is compared with a preset standard prediction ratio threshold, and the preset standard prediction ratio threshold is preferably 80%.

[0221] S46. When the median prediction ratio is greater than or equal to the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model meets the line capacity utilization qualification requirement.

[0222] In the embodiment of the present invention, when the median prediction ratio is greater than or equal to the preset standard prediction ratio threshold, and the median prediction ratio is ≥80%, it indicates that the output prediction of the probability prediction model is relatively accurate, and the line capacity utilization is in a reasonable range. It can be understood that the prediction of the probability prediction model can basically adapt to the actual operation, and the line capacity utilization is reasonable (neither wasteful nor overly risky). Therefore, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model meets the qualified requirements for line capacity utilization.

[0223] S47. When the median prediction ratio is less than the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model does not meet the line capacity utilization qualification requirement.

[0224] In an embodiment of the present invention, when the median prediction ratio is less than the preset standard prediction ratio threshold, the median prediction ratio is less than 80%, which means that the probability prediction model may be overly conservative or the prediction deviation is large, and the line capacity utilization may have problems such as resource waste or insufficient risk assessment. It can be understood that the probability prediction model is too conservative, resulting in resource waste. This determines that the current carrying capacity prediction probability sequence output by the preset probability prediction model does not meet the qualified requirements for line capacity utilization.

[0225] It's worth noting that traditional static ratings can't adapt to dynamic weather conditions. This invention dynamically assesses line capacity utilization by quantifying the "median forecast ratio," exploring the potential current-carrying capacity of lines under different weather scenarios and avoiding the capacity waste caused by static ratings.

[0226] In the present invention, firstly, a capacity prediction model is dynamically constructed by integrating measured weather data and basic thermal characteristic parameters, breaking through the limitation of traditional BP neural network prediction model that only relies on historical data. Then, the predicted weather data is input into the capacity prediction model to predict the capacity. The capacity prediction result is subjected to quantile processing to determine the capacity prediction probability sequence of multiple probability intervals, which solves the problem that the static rating method only outputs a single capacity value and cannot quantify the risk level. Finally, multiple capacity prediction probability sequences are used for risk assessment, and the output of the capacity probability prediction model is ensured to be credible through statistical verification, which fills the gap of the traditional method in lacking systematic verification of the prediction results and solves the problem of lack of reliability verification in the traditional method, thereby ultimately improving the safe operation stability of the transmission line.

[0227] See also Figure 3 , Figure 3 This is a structural block diagram of a line current carrying capacity prediction and risk assessment system provided in Example 3 of the present invention.

[0228] The present invention provides a line current carrying capacity prediction and risk assessment system, comprising:

[0229] An acquisition module 301 is used to acquire measured weather data, predicted weather data, and basic thermal characteristic parameters of a target line;

[0230] A construction module 302 is used to construct a current carrying capacity prediction model based on measured weather data and basic thermal characteristic parameters;

[0231] The prediction module 303 is used to use the predicted weather data to input the capacity prediction model to perform capacity prediction, and perform quantile processing based on the capacity prediction results to determine multiple capacity prediction probability sequences;

[0232] The evaluation module 304 is configured to perform risk evaluation using multiple current carrying capacity prediction probability sequences to obtain risk evaluation results.

[0233] Furthermore, the measured weather data is a plurality of measured weather data at the current and historical moments, and the construction module 302 includes:

[0234] The current carrying capacity observation value submodule is used to perform observation current carrying capacity calculations using each measured weather data and the associated thermal characteristic basic parameters to determine each current carrying capacity observation value;

[0235] The regression coefficient submodule is used to use the current carrying capacity observation values ​​and the associated measured weather data to input the preset linear regression model for solution to obtain multiple regression coefficients;

[0236] The prediction error submodule is used to calculate the prediction error of each regression coefficient and select the regression coefficient with the smallest prediction error to build the current carrying capacity prediction model.

[0237] Furthermore, the measured weather data includes sunshine intensity, ambient temperature and wind speed. The basic thermal characteristic parameters include the conductor surface heat absorption coefficient, conductor diameter, conductor surface emissivity, Stephen Boltzmann constant, conductor temperature, air viscosity, conductor surface air density, air thermal conductivity and conductor resistance. The specific calculation of the observed current carrying capacity is as follows:

[0238] The solar radiation heat absorption data is determined using the sunlight intensity, conductor surface heat absorption coefficient and conductor diameter;

[0239] The radiation heat dissipation data is determined using the conductor diameter, conductor surface emissivity, Stephen-Boltzmann constant, ambient temperature and conductor temperature;

[0240] The Reynolds number is determined using the conductor diameter, wind speed, air viscosity, and air density on the conductor surface;

[0241] Compare wind speed with preset wind speed test threshold;

[0242] Based on the comparison results, the Reynolds number, ambient temperature, conductor temperature and air thermal conductivity are used to determine the forced convection heat dissipation data;

[0243] The observed current carrying capacity is determined using solar radiation heat absorption data, radiation heat dissipation data, forced convection heat dissipation data and conductor resistance values.

[0244] Furthermore, based on the comparison results, the Reynolds number, ambient temperature, conductor temperature, and air thermal conductivity are used to determine the forced convection heat dissipation data, including:

[0245] When the wind speed is less than the preset wind speed test threshold, the Reynolds number, ambient temperature, conductor temperature and air thermal conductivity are used to input the first preset forced convection heat dissipation function to obtain forced convection heat dissipation data;

[0246] When the wind speed is greater than or equal to the preset wind speed test threshold, the Reynolds number, ambient temperature, conductor temperature and air thermal conductivity are used to input a second preset forced convection heat dissipation function to obtain forced convection heat dissipation data.

[0247] Furthermore, the predicted weather data is the predicted weather data for multiple moments in a preset future time period obtained by prediction using a preset mesoscale weather forecast model. The prediction module 303 includes:

[0248] The current carrying capacity prediction value submodule is used to use each forecast weather data input corresponding current carrying capacity prediction model to obtain multiple current carrying capacity prediction values;

[0249] The quantile current carrying capacity prediction value sequence submodule is used to perform interval quantile operation on multiple current carrying capacity prediction values ​​to obtain multiple quantile current carrying capacity prediction value sequences;

[0250] The solution submodule is used to input the preset probability prediction model into each quantile current carrying capacity prediction value sequence for solution to obtain multiple current carrying capacity prediction probability sequences.

[0251] Furthermore, the quantile current carrying capacity prediction value sequence submodule includes:

[0252] A target width interval unit is used to select a maximum value and a minimum value from a plurality of current carrying capacity prediction values, and determine a boundary value of a preset width interval based on the maximum value and the minimum value to obtain a target width interval;

[0253] The fixed width interval unit is used to divide the target width interval into intervals to obtain multiple fixed width intervals;

[0254] an ascending sorting unit, configured to divide the plurality of predicted current carrying capacity values ​​into corresponding fixed width intervals, and to sort the predicted current carrying capacity values ​​within each fixed width interval in ascending order;

[0255] The quantile current carrying capacity prediction value unit is used to calculate the quantile current carrying capacity prediction values ​​corresponding to multiple quantiles in the fixed width interval after ascending sorting;

[0256] The statistical unit is used to count multiple quantile current carrying capacity prediction values ​​to obtain a quantile current carrying capacity prediction value sequence corresponding to each sorted fixed width interval.

[0257] Furthermore, the current carrying capacity prediction probability sequence includes current carrying capacity prediction probability values ​​corresponding to each quantile, and the evaluation module 304 includes:

[0258] The reliability assessment submodule is used to perform reliability assessment using the predicted probability values ​​of each current carrying capacity to obtain a reliability assessment result;

[0259] The line capacity utilization analysis submodule is used to perform line capacity utilization analysis using the predicted probability values ​​of each current carrying capacity to obtain the line capacity utilization analysis results;

[0260] The risk assessment results include reliability assessment results and line capacity utilization analysis results.

[0261] Furthermore, the reliability assessment submodule includes:

[0262] A target prediction probability sequence unit is used to extract the current carrying capacity prediction probability value corresponding to the same quantile from multiple current carrying capacity prediction probability sequences to obtain target prediction probability sequences corresponding to multiple different quantile types;

[0263] The target prediction probability sequence includes multiple current carrying capacity prediction probability values ​​corresponding to the same quantile;

[0264] Reliability index unit, used to calculate the reliability index corresponding to each target prediction probability sequence;

[0265] An over-limit frequency unit is used to perform difference calculations between each reliability index and a preset over-limit frequency coefficient to obtain multiple over-limit frequencies;

[0266] A target deviation unit is used to perform difference calculations on each over-limit frequency and the associated quantile to obtain multiple target deviations;

[0267] A first comparison unit is used to compare each target deviation with a preset allowable deviation threshold;

[0268] The first data processing unit is configured to determine that the target prediction probability sequence output by the preset probability prediction model is reliable when the target deviation is less than or equal to a preset allowable deviation threshold;

[0269] The second data processing unit is used to determine that the target prediction probability sequence output by the preset probability prediction model is unreliable when the target deviation is greater than a preset allowable deviation threshold.

[0270] Furthermore, the line capacity utilization analysis submodule includes:

[0271] A selection unit is used to select the maximum quantile as the target quantile;

[0272] A target capacity prediction probability value unit is used to extract the capacity prediction probability value corresponding to the target quantile from multiple capacity prediction probability sequences as the target capacity prediction probability value;

[0273] A prediction ratio unit, used to calculate the prediction ratio of each target current carrying capacity prediction probability value;

[0274] a median prediction ratio unit, used to sort the prediction ratios and determine the median prediction ratio;

[0275] a second comparison unit, configured to compare the median prediction ratio with a preset standard prediction ratio threshold;

[0276] The third data processing unit is configured to determine that the current carrying capacity prediction probability sequence output by the preset probability prediction model meets the line capacity utilization qualification requirement when the median prediction ratio is greater than or equal to a preset standard prediction ratio threshold;

[0277] The fourth data processing unit is used to determine that the current carrying capacity prediction probability sequence output by the preset probability prediction model does not meet the line capacity utilization qualification requirement when the median prediction ratio is less than the preset standard prediction ratio threshold.

[0278] In the present invention, firstly, a capacity prediction model is dynamically constructed by integrating measured weather data and basic thermal characteristic parameters, breaking through the limitation of traditional BP neural network prediction model that only relies on historical data. Then, the predicted weather data is input into the capacity prediction model to predict the capacity. The capacity prediction result is subjected to quantile processing to determine the capacity prediction probability sequence of multiple probability intervals, which solves the problem that the static rating method only outputs a single capacity value and cannot quantify the risk level. Finally, multiple capacity prediction probability sequences are used for risk assessment, and the output of the capacity probability prediction model is ensured to be credible through statistical verification, which fills the gap of the traditional method in lacking systematic verification of the prediction results and solves the problem of lack of reliability verification in the traditional method, thereby ultimately improving the safe operation stability of the transmission line.

[0279] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0280] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0281] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0282] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0283] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0284] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for predicting and assessing line current carrying capacity, characterized in that: include: Obtain measured weather data, predicted weather data and basic thermal characteristic parameters of the target line; Constructing a current carrying capacity prediction model based on the measured weather data and the basic thermal characteristic parameters; The predicted weather data is input into the capacity prediction model to perform capacity prediction, and quantile processing is performed according to the capacity prediction result to determine multiple capacity prediction probability sequences; A plurality of the current carrying capacity prediction probability sequences are used to perform risk assessment to obtain a risk assessment result.

2. The line current carrying capacity prediction and risk assessment method according to claim 1, characterized in that: The measured weather data is a plurality of measured weather data at current and historical moments. The current carrying capacity prediction model is constructed based on the measured weather data and the thermal characteristic basic parameters, including: Performing observation current carrying capacity calculations using each of the measured weather data and the associated thermal characteristic basic parameters to determine each current carrying capacity observation value; Using each of the current carrying capacity observation values ​​and the associated measured weather data to input into a preset linear regression model for solution to obtain a plurality of regression coefficients; The prediction error of each regression coefficient is calculated, and the regression coefficient with the smallest prediction error is selected to construct a current carrying capacity prediction model.

3. The line current carrying capacity prediction and risk assessment method according to claim 2, characterized in that: The measured weather data includes sunshine intensity, ambient temperature and wind speed. The basic thermal characteristic parameters include conductor surface heat absorption coefficient, conductor diameter, conductor surface emissivity, Stephen Boltzmann constant, conductor temperature, air viscosity, conductor surface air density, air thermal conductivity and conductor resistance. The observed current carrying capacity calculation is specifically as follows: Determining solar radiation heat absorption data using the sunlight intensity, the conductor surface heat absorption coefficient, and the conductor diameter; Determining radiation heat dissipation data using the wire diameter, the wire surface emissivity, the Stephen-Boltzmann constant, the ambient temperature, and the wire temperature; Determining a Reynolds number using the conductor diameter, the wind speed, the air viscosity coefficient, and the air density on the conductor surface; comparing the wind speed with a preset wind speed test threshold; Based on the comparison result, the Reynolds number, the ambient temperature, the wire temperature and the air thermal conductivity are used to determine forced convection heat dissipation data; The solar radiation heat absorption data, the radiation heat dissipation data, the forced convection heat dissipation data and the conductor resistance value are used to determine an observed current carrying capacity value.

4. The line current carrying capacity prediction and risk assessment method according to claim 3, characterized in that: Determining forced convection heat dissipation data based on the comparison result using the Reynolds number, the ambient temperature, the wire temperature, and the air thermal conductivity includes: When the wind speed is less than a preset wind speed test threshold, the Reynolds number, the ambient temperature, the conductor temperature, and the air thermal conductivity are used to input a first preset forced convection heat dissipation function to obtain forced convection heat dissipation data; When the wind speed is greater than or equal to the preset wind speed test threshold, the Reynolds number, the ambient temperature, the wire temperature and the air thermal conductivity are used to input a second preset forced convection heat dissipation function to obtain forced convection heat dissipation data.

5. The line current carrying capacity prediction and risk assessment method according to claim 1, characterized in that: The predicted weather data is the predicted weather data for multiple moments in a preset future time period obtained by predicting a preset mesoscale weather forecast model. The predicted weather data is input into the capacity prediction model to perform capacity prediction, and quantile processing is performed based on the capacity prediction results to determine multiple capacity prediction probability sequences, including: Using each of the forecast weather data to input the corresponding capacity prediction model to obtain a plurality of capacity prediction values; Performing interval quantile operation on the plurality of the current carrying capacity prediction values ​​to obtain a plurality of quantile current carrying capacity prediction value sequences; Each of the quantile current carrying capacity prediction value sequences is input into a preset probability prediction model for solution to obtain a plurality of current carrying capacity prediction probability sequences.

6. The line current carrying capacity prediction and risk assessment method according to claim 5, characterized in that: The performing of interval quantile operation on the plurality of the predicted current carrying capacity values ​​to obtain a plurality of quantile current carrying capacity prediction value sequences includes: Selecting a maximum value and a minimum value from the plurality of predicted current carrying capacity values, and determining boundary values ​​of a preset width interval based on the maximum value and the minimum value to obtain a target width interval; Dividing the target width interval into intervals to obtain a plurality of fixed width intervals; Dividing the plurality of predicted current carrying capacity values ​​into corresponding fixed width intervals, and sorting the predicted current carrying capacity values ​​within each fixed width interval in ascending order; Calculate the predicted quantile current carrying capacity values ​​corresponding to multiple quantiles in each fixed width interval after ascending sorting; A plurality of the quantile current carrying capacity prediction values ​​are counted to obtain a quantile current carrying capacity prediction value sequence corresponding to each of the sorted fixed-width intervals.

7. The line current carrying capacity prediction and risk assessment method according to claim 6, characterized in that: The capacity prediction probability sequence includes a capacity prediction probability value corresponding to each quantile, and the risk assessment is performed using multiple capacity prediction probability sequences to obtain a risk assessment result, including: Performing a reliability assessment using each of the current carrying capacity prediction probability values ​​to obtain a reliability assessment result; Performing line capacity utilization analysis using each of the current carrying capacity prediction probability values ​​to obtain a line capacity utilization analysis result; The risk assessment result includes the reliability assessment result and the line capacity utilization analysis result.

8. The line current carrying capacity prediction and risk assessment method according to claim 7, characterized in that: The reliability assessment is performed using each of the current carrying capacity prediction probability values ​​to obtain a reliability assessment result, including: Extracting the current carrying capacity prediction probability value corresponding to the same quantile from the plurality of current carrying capacity prediction probability sequences, and obtaining target prediction probability sequences corresponding to a plurality of different quantile types; The target prediction probability sequence includes a plurality of current carrying capacity prediction probability values ​​corresponding to the same quantile; Calculating the reliability index corresponding to each target prediction probability sequence; Performing difference calculations between each of the reliability indicators and a preset over-limit frequency coefficient to obtain a plurality of over-limit frequencies; performing difference operations on each of the over-limit frequencies and the associated quantiles to obtain a plurality of target deviations; Comparing each of the target deviations with a preset tolerance threshold; When the target deviation is less than or equal to the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is reliable; When the target deviation is greater than the preset allowable deviation threshold, it is determined that the target prediction probability sequence output by the preset probability prediction model is unreliable.

9. The line current carrying capacity prediction and risk assessment method according to claim 7, characterized in that: The line capacity utilization analysis is performed using each of the current carrying capacity prediction probability values ​​to obtain a line capacity utilization analysis result, including: Select the largest quantile as the target quantile; Extracting the current carrying capacity prediction probability value corresponding to the target quantile from the plurality of current carrying capacity prediction probability sequences as the target current carrying capacity prediction probability value; Calculating a prediction ratio of each target current carrying capacity prediction probability value; ranking each of the predicted ratios and determining a median predicted ratio; comparing the median predicted ratio with a preset standard predicted ratio threshold; When the median prediction ratio is greater than or equal to the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model meets the line capacity utilization qualification requirement; When the median prediction ratio is less than the preset standard prediction ratio threshold, it is determined that the current carrying capacity prediction probability sequence output by the preset probability prediction model does not meet the line capacity utilization qualification requirement.

10. A line current carrying capacity prediction and risk assessment system, characterized in that: include: An acquisition module is used to obtain measured weather data, predicted weather data, and basic thermal characteristic parameters of the target line; A construction module, configured to construct a current carrying capacity prediction model based on the measured weather data and the basic thermal characteristic parameters; A prediction module, configured to use the predicted weather data to input the capacity prediction model to perform capacity prediction, and perform quantile processing based on the capacity prediction results to determine multiple capacity prediction probability sequences; An evaluation module is used to perform risk evaluation using a plurality of the current carrying capacity prediction probability sequences to obtain a risk evaluation result.

Images