A method for deducing microclimate of power transmission line dynamic capacity enhancement

Abstract

The application provides a kind of transmission line dynamic capacity increasing microclimate deduction method, belong to power system technical field, including: based on intrinsic orthogonal decomposition is carried out feature extraction to wind field simulation data under different working conditions, obtain corresponding mode, construct wind field reduced order model;With the wind field data of sparse monitoring point as constraint condition, the least square method is used to solve each mode weight coefficient, and the whole line reference wind field is obtained by reconstruction;According to the deviation between the measured wind field data of monitoring point and the whole line reference wind field obtained by reconstruction, the local wind field error of non-monitoring point is calculated and residual error is corrected, and the whole line wind field is obtained;Determine the dynamic thermal bottleneck point, and determine the maximum allowable flow capacity according to the wind speed of dynamic thermal bottleneck point.Have the beneficial effect: through offline physical reduction, online sparse reconstruction and residual error correction, the complex wind field of the whole line can be accurately inversed by limited monitoring points, the low-cost, high-precision transmission line microclimate space-time deduction is realized, and the safety margin of dynamic capacity increasing is improved.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method for dynamic capacity expansion simulation of transmission lines using micro-meteorological data. Background Technology

[0002] With the continuous growth of power grid load demand, Dynamic Line Rating (DLR) technology has become a key means to tap the transmission potential of existing transmission lines. The core of DLR technology lies in accurately acquiring the real-time temperature and heat dissipation environment of the conductors. Among these parameters, wind speed and direction are the most sensitive to determining the conductor's heat dissipation capacity. Therefore, accurately monitoring wind speed and direction information is crucial for precisely calculating the conductor's current carrying capacity and realizing dynamic capacity expansion of transmission lines.

[0003] However, existing dynamic capacity expansion systems typically install a micro-weather station every few tens of kilometers to monitor meteorological data such as wind speed and direction. But transmission lines often traverse complex terrains, such as valleys, mountain passes, and forests. These terrains obstruct and accelerate airflow, resulting in highly non-uniform local wind fields. Data obtained from a limited number of monitoring points cannot represent the actual wind field conditions along the entire transmission line, easily leading to missed detection of thermal bottlenecks located in localized low-wind-speed areas (i.e., dead-wind zones). This results in inflated current-carrying capacity calculations, potentially causing overheating, sagging, or even flashover accidents in the transmission lines.

[0004] Furthermore, while Computational Fluid Dynamics (CFD) can simulate wind fields in complex terrains with high accuracy, its computational time is extremely long, typically on the order of hours, which cannot meet the minute-level real-time scheduling requirements of dynamic capacity expansion systems. Although some methods utilize CFD to establish offline databases and obtain wind field data online through table lookups, this approach is limited by the resolution of discrete databases, making it difficult to handle continuous changes in monitoring data between two preset operating conditions. It also cannot correct for systematic errors in the CFD model caused by vegetation growth or surface changes, thus affecting the accuracy of dynamic capacity expansion calculations. Summary of the Invention

[0005] To address the above technical issues, this invention provides a method for dynamic capacity expansion micro-meteorological simulation of transmission lines, solving the problem that sparse monitoring points in complex terrain cannot detect the drift of thermal bottlenecks along the entire line; overcoming the slow online calculation speed of traditional computational fluid dynamics (CFD), achieving millisecond-level reconstruction of the wind field across the entire line; and improving the safety margin of dynamic capacity expansion by integrating physical mechanism intrinsic orthogonal decomposition (POD) modes with residual correction of measured data.

[0006] The technical problem solved by this invention can be achieved by the following technical solutions:

[0007] A method for dynamic capacity expansion micro-meteorological simulation of transmission lines includes: Step S1, extracting features from wind field simulation data under different operating conditions based on the intrinsic orthogonal decomposition method to obtain the corresponding modes of the wind field, and constructing a wind field order reduction model based on the modes; Step S2, using wind field data collected by sparsely distributed monitoring points on the target transmission line as constraints, solving the weight coefficients of each mode using the least squares method to reconstruct the baseline wind field of the entire line; Step S3, calculating and correcting the local wind field error of non-monitoring points based on the deviation between the measured wind field data of the monitoring points and the reconstructed baseline wind field of the entire line, to obtain the entire wind field of the target transmission line; Step S4, determining the dynamic thermal bottleneck point based on the entire wind field, and determining the maximum allowable current carrying capacity of the target transmission line based on the wind speed of the dynamic thermal bottleneck point.

[0008] The method for dynamic capacity expansion of transmission lines according to the present invention includes the following steps in step S1: obtaining the wind field simulation data: Step S101, obtaining digital elevation data and surface roughness data of the target transmission line and establishing a three-dimensional geometric model; Step S102, setting the background wind speed and background wind direction corresponding to different operating conditions; Step S103, for each operating condition, numerically solving the Reynolds-averaged Navier-Stokes equations according to computational fluid dynamics to obtain the wind speed vector field data of each discrete point on the target transmission line under each operating condition, and using it as the wind field simulation data.

[0009] The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to the present invention includes the following steps in step S1: Step S111, averaging the wind field simulation data to calculate the average wind field; Step S112, constructing a pulsation matrix based on the difference between the wind speed vector field data under each operating condition in the wind field simulation data and the average wind field; Step S113, performing singular value decomposition on the pulsation matrix to extract the modes corresponding to the wind field; Step S114, linearly combining the average wind field and the modes to construct the wind field order reduction model.

[0010] The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to the present invention includes the following steps in step S2: Step S21, extracting sub-modes corresponding to monitoring point locations from the modes extracted in step S1; Step S22, constructing an overdetermined or well-posed equation system by linearly combining the wind field data collected at the monitoring point and the sub-modes; Step S23, solving the weight coefficients of each mode in the overdetermined or well-posed equation system using the weighted least squares method to obtain the optimal coefficient vector; Step S24, substituting the obtained optimal coefficient vector into the wind field order reduction model to reconstruct the baseline wind field of the entire line.

[0011] The method for calculating the optimal coefficient vector in the dynamic capacity expansion micro-meteorological simulation method for transmission lines described in this invention is as follows:

[0012]

[0013] in, Represents the optimal coefficient vector; Represents a mode; Let the weight matrix be denoted as . A diagonal matrix, where P represents the number of monitoring points; Indicates the first t Time of the first m Measured wind field data from each monitoring point m =1,2,...,P; Indicates the first m The average wind field at each monitoring point.

[0014] The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to the present invention includes the following method for constructing the weight matrix:

[0015]

[0016] in, Indicates the first t Diagonal elements at any time; The adjustment coefficient representing the static instrument error; Indicates the first Static instrument error under each working condition; An adjustment coefficient representing the uncertainty of the dynamic environment; Indicates the first The dynamic environmental uncertainty of each working condition:

[0017]

[0018] in, Indicates the number of samples within the preset time window; Indicates the sample index; Indicates the first Time of the first i Measured wind field data for each operating condition; Indicates the sampling interval; This indicates the average wind speed within a preset time window.

[0019] The micro-meteorological simulation method for dynamic capacity expansion of transmission lines according to the present invention includes the following steps in step S3: Step S31, for each monitoring point, calculate the deviation between the measured wind field data of the monitoring point and the reconstructed wind field data of the corresponding monitoring point in the benchmark wind field of the entire line, and obtain the residuals of all monitoring points; Step S32, using the terrain complexity factor as a covariate, map the residuals of all monitoring points to the entire line, and obtain the residual distribution of the entire line; Step S33, perform residual correction on the benchmark wind field of the entire line according to the residual distribution of the entire line, and obtain the wind field of the entire line of the target transmission line.

[0020] The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to the present invention includes the following steps in step S32: Step S3201: For any target location point on the target transmission line, calculate the elevation gradient modulus of the target location point using the finite difference method based on the digital elevation model grid where the target location point is located and its surrounding neighborhood, to obtain a local slope component; Step S3202: Calculate the standard deviation of the elevation within the neighborhood to obtain the local topographic relief; Step S3203: Normalize the local slope component and the local topographic relief; Step S3204: Weight the normalized local slope component and the normalized local topographic relief to obtain the topographic complexity factor of the target location point.

[0021] The micro-meteorological simulation method for dynamic capacity expansion of transmission lines according to the present invention includes the following steps in step S32: mapping the residuals of all monitoring points to the entire line, which comprises: step S3211, establishing a trend model and solving for regression coefficients based on the residuals of all monitoring points and the terrain complexity factor; step S3212, calculating the random residuals of each monitoring point based on the difference between the residuals of each monitoring point and the trend term residuals of the corresponding positions in the trend model; step S3213, performing Kriging interpolation based on the random residuals of each monitoring point to obtain the estimated random residuals at any target location on the target transmission line; and step S3214, superimposing the trend term residuals with the estimated random residuals to obtain the residual distribution of the entire target transmission line.

[0022] The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to the present invention includes step S4: step S41, identifying the location with the lowest wind speed and located in the heavy-load section based on the wind field of the entire line, and determining it as the dynamic thermal bottleneck point; step S42, determining the maximum allowable current carrying capacity of the target transmission line based on the wind speed of the dynamic thermal bottleneck point.

[0023] The advantages or beneficial effects of the technical solution of this invention are as follows: This invention utilizes intrinsic orthogonal decomposition to extract the dominant spatial modes of the simulated wind field to construct a reduced-order model, and combines sparse monitoring point data with the least squares method to reconstruct the benchmark wind field. This enables rapid data processing, improves the efficiency of wind field calculation, overcomes the drawbacks of the long computation time of traditional computational fluid dynamics, and meets the needs of real-time scheduling at the minute level for dynamic capacity expansion systems. It introduces topology residual correction, utilizing the deviation between the measured values ​​and the reconstructed values ​​of the monitoring points, and combines the spatial topology correlation of the line to calculate the local errors of non-monitoring points, obtaining a high-resolution wind field across the entire line. This achieves accurate inversion of wind fields under complex terrain, solving the problems of insufficient representativeness of sparse monitoring point data and inability to detect the drift of thermal bottlenecks across the entire line in existing technologies, thus improving the monitoring and early warning capabilities for thermal bottlenecks in transmission lines. Furthermore, this invention integrates physical mechanisms and residual correction of measured data, accurately inverting the complex wind field across the entire line with a limited number of monitoring points. This avoids the cost of laying a large number of monitoring devices, achieving low-cost, high-precision micro-meteorological spatiotemporal extrapolation of transmission lines, and improving the safety margin of dynamic capacity expansion. Attached Figure Description

[0024] Figure 1 A flowchart illustrating the micro-meteorological simulation method for dynamic capacity expansion of transmission lines, as shown in a preferred embodiment of the present invention.

[0025] Figure 2 This is a flowchart illustrating the steps for acquiring wind field simulation data in a preferred embodiment of the present invention.

[0026] Figure 3 This is a flowchart illustrating step S1 in a preferred embodiment of the present invention.

[0027] Figure 4 This is a flowchart illustrating step S2 in a preferred embodiment of the present invention.

[0028] Figure 5 This is a flowchart illustrating step S3 in a preferred embodiment of the present invention.

[0029] Figure 6 This is a flowchart illustrating the method for calculating the terrain complexity factor in step S32, as a preferred embodiment of the present invention.

[0030] Figure 7 In a preferred embodiment of the present invention, step S32 is a flowchart illustrating the process of mapping the residuals of all monitoring points to the entire line.

[0031] Figure 8 This is a flowchart illustrating step S4 in a preferred embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0034] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0035] In a preferred embodiment of the present invention, based on the above-mentioned problems existing in the prior art, a micro-meteorological inference method for dynamic capacity expansion of transmission lines is provided, which is based on reduced-order flow field reconstruction and topological residual correction using Proper Orthogonal Decomposition (POD). This method can retain the physical accuracy of computational fluid dynamics, meet the real-time requirements of dynamic capacity expansion systems, and accurately invert the wind field of the entire line under complex terrain using limited monitoring points, thus providing reliable support for dynamic capacity expansion of transmission lines.

[0036] like Figure 1 As shown, the method for dynamic capacity expansion micro-meteorological simulation of transmission lines includes the following steps:

[0037] Step S1: Based on the intrinsic orthogonal decomposition (POD) method, feature extraction is performed on the wind field simulation data under different operating conditions to obtain the corresponding modes of the wind field, and a reduced order model (ROM) of the wind field is constructed based on the modes.

[0038] Specifically, in the offline phase, a computational fluid dynamics (CFD) flow field snapshot library under complex terrain is first constructed; then, wind field modes are extracted and reduced in order based on intrinsic orthogonal decomposition (POD).

[0039] In the construction phase of the computational fluid dynamics (CFD) flow field snapshot library under complex terrain, comprehensive and accurate wind field simulation data can be obtained through geometric modeling, working condition sampling, and computational fluid dynamics simulation.

[0040] like Figure 2As shown, in step S1, the steps for acquiring wind field simulation data include: step S101, acquiring digital elevation data and surface roughness data of the target transmission line, and establishing a three-dimensional geometric model; step S102, setting the background wind speed and background wind direction corresponding to different operating conditions; step S103, for each operating condition, numerically solving the Reynolds-Averaged Navier-Stokes (RANS) equation according to computational fluid dynamics to obtain the wind speed vector field data of each discrete point on the target transmission line under each operating condition, and using it as wind field simulation data.

[0041] Specifically, in the geometric modeling stage, digital elevation model (DEM) data and surface roughness data are first acquired within a predetermined range on both sides of the transmission line corridor, typically within a 2km range. The DEM data reflects the topographic relief within the area, including natural terrain features such as mountains, valleys, and rivers that influence wind direction and speed. Surface roughness data includes surface objects such as vegetation and buildings, which impede airflow and create friction; different surface roughness levels result in varying degrees of turbulence and velocity attenuation in the near-surface layer. Then, using Geographic Information System (GIS) software and 3D modeling tools, the acquired DEM and surface roughness data are integrated and processed to construct a 3D geometric model of the transmission line and its surrounding environment. This model is designed to closely approximate the actual terrain, providing an accurate physical scene for subsequent computational fluid dynamics simulations.

[0042] In the operating condition sampling phase, by setting operating conditions, the wind field conditions of transmission lines under various meteorological combinations can be comprehensively and accurately simulated, providing a rich variety of input conditions for subsequent computational fluid dynamics simulations. Operating condition settings include, but are not limited to, key meteorological parameters such as background wind speed and background wind direction. For example, background wind speed... The setting range is 0-30m / s; background wind direction The setting range is 0-360°, divided into 22.5° intervals.

[0043] After completing geometric modeling and operating condition sampling, specialized computational fluid dynamics (CFD) software was used to simulate the wind field under different operating conditions. Commonly used CFD software, such as OpenFOAM, possesses powerful numerical calculation capabilities and flexible simulation functions, enabling accurate solutions to complex fluid flow problems. During the CFD simulation, the Reynolds-averaged Navier-Stokes equations were numerically solved for each preset operating condition. These equations are fundamental to describing fluid motion; solving them yields the values ​​of discrete points along the transmission line (including towers and midpoints of spans). Snapshot data of wind speed vector fields at discrete points along the line under different operating conditions. These wind field simulation data are stored in the form of a snapshot library for easy subsequent querying and analysis.

[0044] These snapshot sets are denoted as a matrix. :

[0045]

[0046] in, Indicates the first Under various operating conditions, the entire line Wind speed data vectors at each location point.

[0047] In the wind field mode extraction and order reduction stage based on intrinsic orthogonal decomposition (POD), the snapshot matrix is... Dimensionality reduction is performed to reduce online computation. Intrinsic orthogonal decomposition (POD) is a data-driven dimensionality reduction technique that extracts the dominant orthogonal modes with the highest energy proportion from high-dimensional snapshot data through singular value decomposition (SVD) to construct a low-dimensional subspace to efficiently characterize the core dynamic features of the system.

[0048] like Figure 3 As shown, step S1 includes:

[0049] Step S111: Average the wind field simulation data and calculate the average wind field. :

[0050]

[0051] Among them, subscript Indicates the working condition index. , Indicates the number of operating conditions;

[0052] Step S112: Based on the wind speed vector field data and average wind field under various operating conditions in the wind field simulation data... The difference is used to construct the pulsation matrix. :

[0053]

[0054] Step S113, for the pulsation matrix Perform Singular Value Decomposition (SVD) before extraction Key feature modes ,this Each mode can contain more than 99% of the flow field energy;

[0055] Step S114: Linearly combine the mean wind field and modes to construct a wind field order-reduced model. At this point, the wind field at any given time is... It can be approximated as a linear combination of the mean wind field and the modes:

[0056]

[0057] Among them, subscript Indicates modal index, , This represents the total number of key feature modes extracted; Indicates the first The mode in the th ... The time coefficient at any given moment; Indicates location point The average wind field; Indicates location point In the The wind speed vector field at any given moment.

[0058] Step S2: Using the wind field data collected by the sparsely distributed monitoring points on the target transmission line as constraints, the least squares method is used to solve the weight coefficients of each mode, and the baseline wind field of the entire line is reconstructed.

[0059] Specifically, in actual operation, modal coefficients are solved online based on sparse monitoring data. During actual operation, only [equipment / data] are deployed in the target transmission line. One monitoring point is used to provide real-time measured wind field data. ,in .

[0060] like Figure 4 As shown, step S2 includes:

[0061] Step S21, the modalities extracted from step S1 In the process, sub-modes are extracted from the corresponding monitoring point locations. ;

[0062] Step S22, based on the wind field data collected from the monitoring points Sub-mode By performing linear combinations, we can construct overdetermined or well-posed systems of equations:

[0063]

[0064] in, Indicates the first m Wind field data collected from each monitoring point, ; Indicates the first m Average wind field at each monitoring point;

[0065] Step S23: Use the weighted least squares method to solve for the weight coefficients of each mode in the overdetermined or well-determined equation system to obtain the optimal coefficient vector. Where the superscript T denotes the transpose matrix;

[0066] The optimal coefficient vector is calculated as follows:

[0067]

[0068] in, Represents the optimal coefficient vector; The modal matrix representing the location of the monitoring point; This represents the weight matrix set according to the sensor accuracy. The weight matrix is: A diagonal matrix, where P represents the number of monitoring points; Indicates the first t Time of the first m Measured wind field data from each monitoring point m =1,2,...,P; Indicates the first m The average wind field at each monitoring point.

[0069] In step S23, the weight matrix The value of directly determines the accuracy of the overall reconstruction when the quality of monitoring data is inconsistent, such as when a sensor is disturbed by a temporary strong gust of wind. This invention does not simply use an identity matrix, but proposes an adaptive weighted algorithm based on the sensor's static accuracy and dynamic turbulence intensity.

[0070] weight matrix It is A diagonal matrix, where P represents the number of monitoring points, and its diagonal elements... Corresponding to the The credibility weight of each monitoring point is calculated using the following formula:

[0071]

[0072] in, Represents the weight matrix The Middle t Diagonal elements at any time; Indicates the first The monitoring point at the 1st The total error variance at any given time consists of two parts: static instrument error and dynamic environmental uncertainty.

[0073]

[0074] in, Indicates the first Static instrument error under each working condition; Indicates the first The dynamic environmental uncertainty of each working condition; The adjustment coefficient representing the static instrument error; An adjustment coefficient representing the uncertainty of the dynamic environment; and These are all preset empirical coefficients used to balance the impact of equipment errors and environmental fluctuations. For example... .

[0075] Static instrument error ( This is determined by the hardware specifications of the micro-weather station and is a constant. For example, if the accuracy of the wind speed sensor is... Then take .

[0076] Dynamic environmental uncertainty ( This reflects the turbulence intensity of the airflow at the current moment. If the wind speed at a point changes drastically within a short period, it indicates that the point is in a highly unstable turbulent region, and its representativeness is poor; therefore, its weight should be reduced. The calculation method is as follows:

[0077] Retrieve past preset time window Standard deviation of wind speed within:

[0078]

[0079] Among them, among them, Indicates the number of samples within the preset time window; Indicates the sample index; Indicates the first Time of the first i Measured wind field data for each operating condition; Indicates the sampling interval; This indicates the average wind speed within a preset time window; preset time window It can be set as needed, for example, to 10 minutes.

[0080] Step S24, the obtained optimal coefficient vector Substituting into the wind field reduction model, the overall wind field formula In the process, the baseline wind field for the entire line is obtained through reconstruction. .

[0081] Step S3: Based on the deviation between the measured wind field data at the monitoring points and the reconstructed baseline wind field, the local wind field error at non-monitoring points is calculated and the residual is corrected to obtain the wind field of the entire target transmission line.

[0082] Specifically, since the simplified boundary conditions of the computational fluid dynamics model deviate from the actual microclimate, there are still errors when relying solely on intrinsic orthogonal decomposition and reconstruction. This invention introduces spatial interpolation residual correction to eliminate the influence of local terrain.

[0083] like Figure 5 As shown, step S3 includes:

[0084] Step S31: For each monitoring point, calculate the deviation between the measured wind field data at the monitoring point and the reconstructed wind field data at the corresponding monitoring point location in the baseline wind field along the entire line, and obtain the residuals for all monitoring points:

[0085]

[0086] in, Indicates the first m The monitoring point at the 1st Time residuals; This represents the actual wind field data measured at the monitoring point; This represents the baseline wind field along the entire line; in step S32, using Gaussian process regression (Kriging) or radial basis function (RBF), with the Terrain Complexity Index (TCI) as a covariate, the residuals of all monitoring points are calculated. Mapping to the entire line yields the residual distribution across the entire line. ;like Figure 6 As shown, in step S32, the method for calculating the terrain complexity factor includes: Step S3201, for any target location point on the target transmission line, the elevation gradient modulus of the target location point is calculated using the finite difference method based on the digital elevation model grid where the target location point is located and its surrounding neighborhood, to obtain the local slope component; Step S3202, the standard deviation of the elevation in the neighborhood is calculated to obtain the local terrain undulation; Step S3203, the local slope component and the local terrain undulation are normalized; Step S3204, the normalized local slope component and the normalized local terrain undulation are weighted to obtain the terrain complexity factor of the target location point.

[0087] Specifically, in the experimental design, covariates are independent variables (explanatory variables), which are not manipulated by the experimenter but still affect the response. In this embodiment, the terrain complexity factor (TCI) is used as a covariate. The terrain complexity factor is defined as a weighted combination of local slope and local topographic relief, used to quantify any target location point on the target transmission line. The complexity of the local micro-topography.

[0088] For any target location point on the target transmission line Based on the target location point The digital elevation model (DEM) grid in which it is located and its surroundings (For example Calculate the terrain complexity factor within the grid neighborhood:

[0089]

[0090] in, Indicates the target location point The terrain complexity factor; Indicates the local slope component; Indicates the degree of local topographic relief; The weighting coefficients representing the local slope components; The weighting coefficients representing the local topographic relief, for example, taking .

[0091] Among them, the local slope component The elevation gradient modulus of this point can be calculated using the finite difference method.

[0092]

[0093] in, Indicates elevation; Represents planar coordinates.

[0094] Local topographic relief Reflecting the degree of surface fragmentation, by calculating the surrounding area. The standard deviation of elevation within the neighborhood is obtained as follows:

[0095]

[0096] in, Indicates the elevation of each grid point within the neighborhood. This represents the average elevation of the neighborhood.

[0097] To eliminate the influence of dimensions, the local slope components of the entire line were analyzed. and local topographic relief Normalization process Interval.

[0098] Furthermore, this invention not only utilizes the residuals at monitoring points, but also uses the terrain complexity factor (TCI) data for the entire route to calculate the residuals at non-monitoring points using the Regression Kriging (RK) algorithm, thereby obtaining the residual distribution for the entire route. .

[0099] like Figure 7 As shown, in step S32, mapping the residuals of all monitoring points to the entire line includes:

[0100] Step S3211, Establish a trend model: Assuming that the residuals and terrain complexity have a linear or non-linear correlation, based on... Known residuals at each monitoring point and the terrain complexity factor at its location Fit a trend function Establish a trend model:

[0101]

[0102] in, and They represent the regression coefficients, respectively. Indicates the target location point The terrain complexity factor; This represents a trend model function.

[0103] Solving the regression coefficients using the least squares method and .

[0104] Step S3212, Calculate the regression residuals: Based on the difference between the residuals of each monitoring point and the residuals of the trend term at the corresponding position in the trend model, calculate the stochastic residuals of each monitoring point. :

[0105]

[0106] in, Indicates the monitoring point index. ; express; Represents random residuals;

[0107] Step S3213, perform ordinary kriging interpolation on the random residuals: according to Random residuals at each monitoring point Spatial correlation is analyzed using variograms, and kriging interpolation is performed to obtain the location of any target point on the target transmission line. Random residual estimates :

[0108]

[0109] Among them, weight It is obtained by solving the Kriging equations;

[0110] Step S3214, synthesize the final overall residuals: superimpose the trend term residuals with the estimated random residuals to obtain the overall residual distribution of the target transmission line. :

[0111]

[0112] in, This represents the overall residual distribution, which is a spatial random term. Indicates the target location point The terrain complexity factor is a terrain-related trend term.

[0113] Step S33: Perform residual correction on the baseline wind field along the entire line based on the residual distribution to obtain the wind field along the entire target transmission line.

[0114]

[0115] in, Indicates the wind field along the entire line; This represents the residual distribution across the entire line.

[0116] Step S4: Determine the dynamic thermal bottleneck point based on the wind field along the entire line, and determine the maximum allowable current carrying capacity of the target transmission line based on the wind speed at the dynamic thermal bottleneck point.

[0117] like Figure 8 As shown, step S4 includes: step S41, based on the wind field along the entire line. Step S42 identifies the location with the lowest wind speed and located in the heavy-load section, defining it as the Dynamic Hotspot. Based on the wind speed at the Dynamic Hotspot, step S43 determines the maximum allowable current carrying capacity of the target transmission line. According to IEEE 738 or CIGRE standards, the wind speed, ambient temperature, and conductor parameters at that point are substituted to calculate the current maximum allowable current carrying capacity.

[0118] Specifically, addressing the issues in existing technologies where finite and sparse monitoring points easily miss local low-wind-speed bottlenecks leading to inflated load-carrying capacity calculations, and the time-consuming online computational fluid dynamics (CFD) calculations failing to meet real-time requirements, while offline data is prone to low accuracy due to model system errors, this invention employs an architecture of offline physical reduction, online sparse reconstruction, and residual correction. Specifically, firstly, multi-condition CFD simulation is performed based on a high-precision digital elevation model (DEM). Intrinsic orthogonal decomposition (POD) is used to extract features from the CFD simulation data to obtain the dominant spatial modes of the wind field, and a reduced-order model (ROM) is constructed based on these modes. During online execution, sparse monitoring point data is used as constraints, and the weight coefficients of each mode are quickly solved using the least squares method to reconstruct the baseline wind field. Finally, a topology residual correction module is introduced. Using the deviation between the measured and reconstructed values ​​at monitoring points, combined with the spatial topological correlation of the line, the local errors of non-monitoring points are calculated, ultimately obtaining a high-resolution wind speed field for the entire line.

[0119] To explain this technical solution more clearly and comprehensively, a detailed description will be provided below through a specific embodiment.

[0120] This embodiment describes a dynamic capacity expansion system for a 220kV transmission line crossing a mountainous region. The system executes the following steps:

[0121] Dynamic capacity expansion system for a 220kV transmission line crossing a mountainous area

[0122] Step 1: Data Preparation and Modeling (Offline):

[0123] An area of ​​approximately 30 km in length along the route was selected as the research object. This area contains three micrometeorological monitoring stations, located at towers 10, 50, and 85.

[0124] Acquire data from the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTERG DEM) for this region with an accuracy of 30m.

[0125] A mesh model was built in the professional computational fluid dynamics software OpenFOAM. The wind direction from 0° to 360° was divided into 22.5° intervals as one working condition. At the same time, three different background wind speeds of 5 m / s, 10 m / s and 20 m / s were considered, and steady-state Reynolds-averaged Navigator-Stokes simulation was performed.

[0126] Wind speed data at the midpoints of 100 spans along the entire line were extracted from the simulation results to form a snapshot matrix. This snapshot matrix contains wind speed information at the midpoints of each span along the transmission line under different operating conditions.

[0127] Step 2, Downgrading (Offline):

[0128] Singular value decomposition (SVD) was performed on the resulting snapshot matrix. The decomposition results show that the first 6 modes ( The cumulative contribution rate of these six modes reached 99.2%, indicating that they can highly summarize the wind field information contained in the snapshot matrix. Subsequently, the vectors of these six modes... to and mean wind field vector Save to the embedded industrial computer.

[0129] Step 3: Online simulation process:

[0130] Data transmission from monitoring point at time T

[0131] At time T, the three micro-meteorological monitoring stations will transmit real-time monitoring data back to the system. Specific data is as follows:

[0132] Tower 10 is located in a valley, where the wind speed it monitored was 2.1 m / s, and the wind direction was southeast.

[0133] Tower 50 is located on a ridge, where a wind speed of 8.5 m / s was recorded, and the wind direction was also southeast.

[0134] Tower No. 85 is located on flat ground, and the wind speed monitored was 4.2 m / s, with the wind direction also being southeast.

[0135] Coefficient Determination: The system extracts the modal values ​​corresponding to towers 10, 50, and 85 from the reduced-order model (ROM) and constructs... The observation matrix is ​​used. Since this matrix may be underdetermined, the system employs regularization methods to process it, or utilizes a neural network trained on historical data to predict coefficients, ultimately calculating the current coefficients for the six modes. .

[0136] Full-line reconstruction: Based on the calculated modal coefficients, the wind speed along the entire line is reconstructed and calculated. The calculation results showed that although the wind speed at tower 50 was relatively high, the reconstructed wind speed at tower 32 (located on a leeward slope and without a sensor installed) was only 1.5 m / s.

[0137] Residual correction: After calculation, the residual at tower No. 32 is 0.13 m / s. By adding the residual to the reconstructed wind speed, the corrected wind speed is 1.63 m / s.

[0138] Thermal bottleneck location: By comparing and analyzing the wind speeds of the entire reconstruction line, tower No. 32 was identified as the current thermal bottleneck point.

[0139] Capacity Calculation: Based on the corrected wind speed of 1.63 m / s for tower No. 32, the current carrying capacity of this line is calculated to be 650 A. However, if only the average data of 4.2 m / s from tower No. 85 is used for calculation, the current carrying capacity may be mistakenly judged as 900 A. This avoids the potential overheating risk that may be caused by misjudgment of the current carrying capacity and effectively ensures the safe and stable operation of the transmission line.

[0140] The advantages or beneficial effects of adopting the above technical solution are as follows:

[0141] (1) Panoramic perception capability: It breaks through the limitation of traditional single point representing the whole line, and can reproduce low wind speed areas under micro-topography such as valleys and leeward slopes, effectively capture hot bottleneck points that drift with the wind direction, and eliminate the safety blind spots of DLR.

[0142] (2) Extremely high computational efficiency: By using the POD reduced-order model, the complex CFD fluid dynamics operation is transformed into a simple linear algebra matrix operation, reducing the online calculation time from hours to milliseconds, which fully meets the real-time dispatching requirements of the power grid.

[0143] (3) Strong robustness: It combines the overall flow field structure constraints of the physical model with data-driven real-time residual correction. When monitoring data is missing, the physical model can ensure the basic trend is correct; when the environment changes, the residual correction can make up for the systematic error of the physical model.

[0144] (4) Low-cost implementation: There is no need to increase the number of sensors along the route on a large scale. High-precision simulation can be achieved by using only a small number of existing micro-weather stations combined with publicly available topographic data, which greatly reduces the cost of engineering transformation.

[0145] The above are merely preferred embodiments of the present invention and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should recognize that any equivalent substitutions and obvious changes made using the content of this specification and illustrations should be included within the protection scope of the present invention.

Claims

1. A method for dynamic capacity expansion micro-meteorological simulation of transmission lines, characterized in that, include: Step S1: Based on the intrinsic orthogonal decomposition method, feature extraction is performed on the wind field simulation data under different operating conditions to obtain the corresponding modes of the wind field, and a wind field order reduction model is constructed based on the modes. Step S2: Using the wind field data collected by the sparsely distributed monitoring points on the target transmission line as constraints, the least squares method is used to solve the weight coefficients of each mode, and the baseline wind field of the entire line is reconstructed. Step S3: Based on the deviation between the measured wind field data at the monitoring points and the reconstructed baseline wind field, the local wind field error at non-monitoring points is calculated and the residual is corrected to obtain the wind field of the entire target transmission line. Step S4: Determine the dynamic thermal bottleneck point based on the wind field of the entire line, and determine the maximum allowable current carrying capacity of the target transmission line based on the wind speed of the dynamic thermal bottleneck point. Step S2 includes: Step S21: Extract the sub-mode corresponding to the monitoring point position from the modes extracted in step S1; Step S22: Based on the wind field data collected by the monitoring point and the sub-mode, a linear combination is performed to construct an overdetermined or well-determined set of equations; Step S23: The weighted least squares method is used to solve for the weight coefficients of each mode in the overdetermined or well-determined equation set to obtain the optimal coefficient vector. The method for calculating the optimal coefficient vector is as follows: ； in, Represents the optimal coefficient vector; Represents a mode; Let the weight matrix be denoted as . diagonal matrix, Indicates the number of monitoring points; This represents the measured wind field data at the m-th monitoring point at time t, where m = 1, 2, ..., P; This represents the average wind field at the m-th monitoring point; The diagonal elements of the weight matrix Corresponding to the The confidence weight of the monitoring point is the confidence weight of the th monitoring point. The monitoring point at the 1st The reciprocal of the total error variance at time t, which consists of two parts: static instrument error and dynamic environmental uncertainty; Step S24: Substitute the obtained optimal coefficient vector into the wind field order reduction model to reconstruct the baseline wind field for the entire line; Step S3 includes: Step S31: For each monitoring point, calculate the deviation between the measured wind field data of the monitoring point and the reconstructed wind field data of the corresponding monitoring point in the benchmark wind field of the whole line, and obtain the residuals of all monitoring points. Step S32: Using the terrain complexity factor as a covariate, map the residuals of all monitoring points to the entire line to obtain the residual distribution of the entire line; Step S33: Perform residual correction on the baseline wind field of the entire line according to the residual distribution of the entire line to obtain the wind field of the entire target transmission line.

2. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, In step S1, the steps for acquiring the wind field simulation data include: Step S101: Obtain digital elevation data and surface roughness data of the target transmission line, and establish a three-dimensional geometric model; Step S102: Set the background wind speed and background wind direction for different operating conditions; Step S103: For each operating condition, the Reynolds-averaged Navier-Stokes equations are numerically solved according to computational fluid dynamics to obtain the wind speed vector field data of each discrete point on the target transmission line under each operating condition, and this data is used as the wind field simulation data.

3. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, Step S1 includes: Step S111: Average the wind field simulation data and calculate the average wind field. Step S112: Construct a pulsation matrix based on the difference between the wind speed vector field data under each working condition and the average wind field in the wind field simulation data; Step S113: Perform singular value decomposition on the pulsation matrix to extract the modes corresponding to the wind field; Step S114: Linearly combine the average wind field and the mode to construct the wind field reduced-order model.

4. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, The method for constructing the weight matrix includes: ； in, Represents the diagonal elements at time t; The adjustment coefficient representing the static instrument error; Indicates the first Static instrument error under each working condition; An adjustment coefficient representing the uncertainty of the dynamic environment; Indicates the first The dynamic environmental uncertainty of each working condition: ； in, Indicates the number of samples within the preset time window; Indicates the sample index; Indicates the first Measured wind field data for the i-th operating condition at time i; Indicates the sampling interval; This indicates the average wind speed within a preset time window.

5. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, In step S32, the method for calculating the terrain complexity factor includes: Step S3201: For any target location point on the target transmission line, the elevation gradient modulus of the target location point is calculated using the finite difference method based on the digital elevation model grid where the target location point is located and its surrounding neighborhood, to obtain the local slope component. Step S3202: Calculate the standard deviation of elevation within the neighborhood to obtain the local topographic relief. Step S3203: Normalize the local slope component and the local topographic relief. Step S3204: Weight the normalized local slope component and the normalized local terrain undulation to obtain the terrain complexity factor of the target location point.

6. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, In step S32, mapping the residuals of all monitoring points to the entire line includes: Step S3211: Based on the residuals of all the monitoring points and the terrain complexity factor, establish a trend model and solve for the regression coefficients; Step S3212: Calculate the random residuals of each monitoring point based on the difference between the residuals of each monitoring point and the residuals of the corresponding trend terms in the trend model. Step S3213: Perform Kriging interpolation based on the random residuals of each monitoring point to obtain the estimated random residual value of any target location point on the target transmission line; Step S3214: The trend term residual is superimposed with the estimated random residual to obtain the residual distribution of the entire target transmission line.

7. The method for dynamic capacity expansion micro-meteorological simulation of transmission lines according to claim 1, characterized in that, Step S4 includes: Step S41: Based on the wind field along the entire line, identify the location with the lowest wind speed and located in the heavy-load section, and determine it as the dynamic thermal bottleneck point. Step S42: Determine the maximum allowable current carrying capacity of the target transmission line based on the wind speed at the dynamic heat bottleneck point.

Images