A wind field prediction system and method for complex terrain areas of power grid transmission channels

Abstract

The present invention discloses a wind field prediction system and method for complex terrain areas of power grid transmission channels, which collects historical observation data and pattern grid reanalysis data of selected areas of the transmission channel to obtain historical data for prediction; obtains site observation data variables of the prediction area; obtains wind field prediction coefficients based on historical observation data and site observation data variables; and predicts the wind field based on site observation data variables and wind field prediction coefficients. The wind field prediction method for complex terrain of power grid transmission channels of the present invention includes complex terrain, near-ground wind field and temperature field as the main influencing factors. The present invention establishes a data set to achieve more practical regional division and time division. The established prediction equation relies on the wind field, temperature and air pressure data of the power grid transmission channel over the years. On the basis of clarifying the main factors of complex terrain, near-ground wind field and temperature field, the grid reanalysis data is used to provide strong data support for the prediction equation, and the accuracy of the prediction equation is greatly improved.

Description

Technical Field

[0001] The invention relates to a wind field prediction method in the meteorological field, in particular to a wind field prediction method in a complex terrain area of ​​a power transmission channel. Background Art

[0002] With the development of supercomputers and the increasing maturity of numerical weather forecasting technology, numerical weather forecast models are used in meteorological services as an important forecasting method, which can provide relatively accurate forecasts of meteorological elements for applications such as weather, climate, water conservancy, and electricity. However, in the process of solving numerical weather forecasts, due to the imperfect initial conditions and model defects, the numerical weather forecast model has initial value uncertainty and model uncertainty, and this type of uncertainty will restrict the forecasting skills. In the transmission channel area, the terrain is more complex. Because the model does not fully describe the terrain and the observation data of complex terrain cannot be fully utilized, the forecast error in the transmission channel area is relatively large. Moreover, it is difficult to combine weather systems of different scales with such complex and steep terrain. Therefore, it is not easy for numerical forecast models to make good forecasts in these areas. On the one hand, this involves the problem of complex terrain processing, and on the other hand, it involves the problem of data processing on complex terrain.

[0003] The existing wind field prediction methods are mainly aimed at the meteorological field, for example: directly using the evolution of similar historical cases as the basis for current prediction, or directly establishing a numerical solution model by discretizing the dynamic equation. However, there are few applications in complex terrain areas of power grids, and the impact of complex terrain of power grid transmission channels on prediction is not fully considered. Summary of the invention

[0004] The purpose of the present invention is to provide a wind field prediction system and method for complex terrain areas in power grid transmission channels. By collecting site observation data and pattern reanalysis data of power grid transmission channels over the years for analysis, a transmission channel wind field prediction method is creatively proposed in which complex terrain, near-ground wind field and temperature field are the main factors affecting the actual wind field of the transmission channel.

[0005] To achieve this purpose, the present invention designs a wind field prediction system for complex terrain areas of power grid transmission channels, including a data collection and processing module, a prediction area data acquisition module, a wind field prediction coefficient acquisition module and a wind field prediction module; the data collection and processing module is used to collect historical observation data and pattern grid reanalysis data in a selected area of ​​the transmission channel to obtain historical data for prediction; the prediction area data acquisition module is used to obtain site observation data variables of each automatic weather station in the selected area of ​​the transmission channel; the wind field prediction coefficient acquisition module obtains wind field prediction coefficients based on the historical observation data of the collection and processing module and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel; the wind field prediction module predicts the wind field based on the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel and the wind field prediction coefficients acquired by the wind field prediction coefficient acquisition module.

[0006] The specific implementation method of the data collection and processing module is:

[0007] S110, dividing the selected area into longitude and latitude grids at a predetermined distance according to the power transmission channel, and dividing the natural year into time intervals at a predetermined time interval;

[0008] S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years, wherein p in this embodiment is preferably 10 years;

[0009] S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the grid observation data after interpolation is G ij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ij represents the interpolated grid observation data of the jth predetermined time interval at the i-th station location;

[0010] S140, reanalyze the interpolated grid data set G ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is:

[0011]

[0012] The historical observation data includes the ground automatic weather station 10-meter wind direction, the ground automatic weather station 10-meter wind speed, the ground 2-meter temperature of the automatic weather station and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel in each predetermined time interval;

[0013] The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, the model reanalysis ground 10-meter wind speed, the model reanalysis ground 2-meter temperature and the model reanalysis ground air pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grids is the horizontal resolution of the grid reanalysis data.

[0014] The specific implementation of the interpolation algorithm is as follows:

[0015] If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is:

[0016]

[0017] Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2);

[0018] The interpolated grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij (wdir), interpolated 10-meter wind speed G ij (spd), interpolated 2-meter temperature G ij (T) and interpolated surface pressure G ij (P).

[0019] The specific implementation method of the prediction area data acquisition module is:

[0020] Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τk represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time.

[0021] The specific implementation method of the wind field prediction coefficient acquisition module is:

[0022] S310, using the station observation data variables in the selected area of ​​the transmission channel to estimate the initial guess value CFu of the zonal wind forecast i (t) and the initial guess of the meridional wind forecast CFv i (t), where i represents the i-th site location and t represents the prediction time; the estimation formula is:

[0023]

[0024]

[0025] in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment longitudinal wind of the time, k represents the ordinal number of the time before the prediction time, Indicates rounding up;

[0026] S320, calculate the comprehensive distance between the initial guess of wind field forecast and all historical reanalysis grid data

[0027]

[0028]

[0029] In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, represents rounding up, n represents the ordinal number of the four variables, σ n is the weight of the nth variable;

[0030] S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all historical reanalysis grid data ij :

[0031]

[0032] S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates:

[0033]

[0034] in Indicates rounding up;

[0035] Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im , where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij :

[0036]

[0037] The n=1 represents the variable ground 10 meters wind direction, n=2 represents the variable 10 meters wind speed, n=3 represents the variable 2 meters temperature, and n=4 represents the variable ground air pressure.

[0038] The specific implementation method of the wind field prediction module 4 is:

[0039] Using the historical analysis observation dataset Z ij 10m wind speed Z ij (wsp) and 10-meter wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is:

[0040] Zu ij =-Z ij (wsp)×sin(Z ij (wdir))

[0041] Z ij =-Z ij (wsp)×cos(Z ij (wdir))

[0042] Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is:

[0043]

[0044]

[0045] Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

[0046] A method for predicting wind fields in complex terrain areas of power grid transmission channels comprises the following steps:

[0047] Step 1: Collect historical observation data and model grid reanalysis data in the selected area of ​​the transmission channel to obtain historical data for prediction;

[0048] Step 2, obtaining the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel;

[0049] Step 3, obtaining a wind field prediction coefficient based on the historical observation data and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel;

[0050] Step 4: predicting the wind field based on the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel and the wind field prediction coefficient.

[0051] The specific implementation method of step 1 is:

[0052] S110, dividing the selected area into longitude and latitude grids at a predetermined distance according to the power transmission channel, and dividing the natural year into time intervals at a predetermined time interval;

[0053] S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years, wherein p in this embodiment is preferably 10 years;

[0054] S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the grid observation data after interpolation is G ij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ijrepresents the interpolated grid observation data of the jth predetermined time interval at the i-th station location;

[0055] S140, reanalyze the interpolated grid data set G ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is:

[0056]

[0057] The historical observation data includes the ground automatic weather station 10-meter wind direction, the ground automatic weather station 10-meter wind speed, the ground 2-meter temperature of the automatic weather station and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel in each predetermined time interval;

[0058] The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, model reanalysis ground 10-meter wind speed, model reanalysis ground 2-meter temperature and model reanalysis ground pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grid is the horizontal resolution of the grid reanalysis data;

[0059] The specific implementation of the interpolation algorithm in step S130 is:

[0060] If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is:

[0061]

[0062] Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2);

[0063] The interpolated grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij (wdir), interpolated 10-meter wind speed G ij (spd), interpolated 2-meter temperature G ij (T) and interpolated surface pressure G ij (P);

[0064] The specific implementation method of step 2 is:

[0065] Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τ k represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time.

[0066] The specific implementation method of step 3 is:

[0067] S310, using the station observation data variables in the selected area of ​​the transmission channel to estimate the initial guess value CFu of the zonal wind forecast i (t) and the initial guess of the meridional wind forecast CFv i (t), where i represents the i-th site location and t represents the prediction time; the estimation formula is:

[0068]

[0069]

[0070] in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment longitudinal wind of the time, k represents the ordinal number of the time before the prediction time, Indicates rounding up;

[0071] S320, calculate the comprehensive distance between the initial guess of wind field forecast and all historical reanalysis grid data

[0072]

[0073]

[0074] In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, represents rounding up, n represents the ordinal number of the four variables, σ n is the weight of the nth variable;

[0075] S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all historical reanalysis grid data ij :

[0076]

[0077] S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates:

[0078]

[0079] in Indicates rounding up;

[0080] Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im , where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij :

[0081]

[0082] The n=1 represents the variable ground 10 meters wind direction, n=2 represents the variable 10 meters wind speed, n=3 represents the variable 2 meters temperature, and n=4 represents the variable ground pressure;

[0083] The specific implementation method of step 4 is:

[0084] Using the historical analysis observation dataset Z ij 10m wind speed Z ij (wsp) and 10-meter wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is:

[0085] Zu ij =-Z ij (wsp)×sin(Z ij (wdir))

[0086] Z ij =-Z ij (wsp)×cos(Z ij (wdir))

[0087] Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is:

[0088]

[0089]

[0090] Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

[0091] The beneficial effects of the present invention are:

[0092] The complex terrain wind field prediction method for the power grid transmission channel includes complex terrain, near-ground wind field and temperature field as the main influencing factors. By collecting the wind field, temperature and air pressure data of the power grid transmission channel over the years, a creative formula for the initial guess of the wind field based on the terrain altitude is proposed.

[0093] By establishing a data set, the complex terrain areas of the power grid transmission channel are gridded, achieving a more practical regional division. At the same time, time division is performed from 0:00 to 24:00 every minute, every hour or every natural day, achieving a more practical time division. The established prediction equation relies on the wind field, temperature and air pressure data of the power grid transmission channel over the years. On the basis of clarifying the main factors of complex terrain, near-ground wind field and temperature field, the grid reanalysis data is used to provide strong data support for the prediction equation, and the accuracy of the prediction equation is greatly improved.

[0094] By calculating the impact of wind field, temperature and air pressure on wind field prediction through historical grid data, the zonal wind and meridional wind can be accurately calculated, and the actual wind field, temperature field and air pressure field in the monitored area to be predicted can be included in the data set, thereby optimizing the prediction equation and improving the accuracy of the prediction equation. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 The system structure diagram of the present invention is

[0096] Among them, 1-data collection and processing module, 2-prediction area data acquisition module, 3-wind field prediction coefficient acquisition module and 4-wind field prediction module. DETAILED DESCRIPTION

[0097] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0098] A wind field prediction system for complex terrain areas in power grid transmission channels, such as Figure 1 As shown, it includes a data collection and processing module 1, a prediction area data acquisition module 2, a wind field prediction coefficient acquisition module 3 and a wind field prediction module 4;

[0099] The data collection and processing module 1 is used to collect historical observation data and pattern grid reanalysis data of the selected area of ​​the transmission channel to obtain historical data for prediction;

[0100] The prediction area data acquisition module 2 is used to obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel;

[0101] The wind field prediction coefficient acquisition module 3 acquires the wind field prediction coefficient based on the historical observation data of the collection and processing module 1 and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel of the prediction area data acquisition module 2;

[0102] The wind field prediction module 4 predicts the wind field based on the site observation data variables of each automatic weather station in the transmission channel selected area of ​​the prediction area data acquisition module 2 and the wind field prediction coefficient obtained by the wind field prediction coefficient acquisition module 3.

[0103] In the above technical solution, the specific implementation method of the data collection and processing module 1 is:

[0104] S110, dividing the selected area into longitude and latitude grids at predetermined distances according to the power transmission channel, and dividing the natural year into time intervals at predetermined time intervals;

[0105] S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years, wherein p in this embodiment is preferably 10 years;

[0106] S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the grid observation data after interpolation is Gij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ij represents the interpolated grid observation data of the jth predetermined time interval at the i-th station location;

[0107] S140, normalization of grid data, reanalysis of the interpolated grid data set G ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is:

[0108]

[0109] In the above technical solution, the historical observation data includes the ground automatic weather station 10-meter wind direction, the ground 10-meter wind speed, the ground 2-meter temperature and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel within each predetermined time interval;

[0110] The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, the model reanalysis ground 10-meter wind speed, the model reanalysis ground 2-meter temperature and the model reanalysis ground air pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grids is the horizontal resolution of the grid reanalysis data.

[0111] In the above technical solution, the specific implementation of the interpolation algorithm is:

[0112] If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is:

[0113]

[0114] Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2);

[0115] The interpolated grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij(wdir), in degrees, interpolated 10-meter wind speed G ij (spd), in m / s, interpolated 2-meter temperature G ij (T), in K and interpolated surface pressure G ij (P), unit is hPa.

[0116] In the above technical solution, the specific implementation method of the prediction area data acquisition module 2 is:

[0117] Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τ k represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time.

[0118] In the above technical solution, the specific implementation method of the wind field prediction coefficient acquisition module 3 is:

[0119] S310, calculate the initial guess value of wind field forecast, and estimate the initial guess value CFu of zonal wind forecast using the station observation data variables in the selected area of ​​the transmission channel i (t) (in m / s) and the initial guess of the meridional wind forecast CFv i (t) (unit: m / s), where i represents the location of the i-th site and t represents the predicted time; the estimation formula is:

[0120]

[0121]

[0122] in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment longitudinal wind of the time, k represents the ordinal number of the time before the prediction time, The unit is m, Indicates rounding up;

[0123] S320, calculate the comprehensive distance between the initial guess of wind field forecast and all historical reanalysis grid data

[0124] Calculate the initial guess of the zonal wind and The distance Su at the i-th station location ij , calculate the initial guess of the meridional wind and the The distance Sv of the jth predetermined time interval ij , the specific calculation formula is:

[0125]

[0126]

[0127] In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, The unit is m, represents rounding up, n represents the ordinal number of the four variables, σ n is the weight of the nth variable. In this embodiment

[0128] S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all historical reanalysis grid data ij :

[0129]

[0130] S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates:

[0131]

[0132] in Indicates rounding up;

[0133] Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im, where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij :

[0134]

[0135] In the above technical solution, n=1 represents the variable ground 10 meters wind direction, n=2 represents the variable 10 meters wind speed, n=3 represents the variable 2 meters temperature, and n=4 represents the variable ground air pressure.

[0136] In the above technical solution, the specific implementation method of the wind field prediction module 4 is:

[0137] Using the historical analysis observation dataset Z ij 10m wind speed Z ij (wsp) and 10-meter wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is:

[0138] Zu ij =-Z ij (wsp)×sin(Z ij (wdir))

[0139] Z ij =-Z ij (wsp)×cos(Z ij (wdir))

[0140] Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is:

[0141]

[0142]

[0143] Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

[0144] A method for predicting wind fields in complex terrain areas of power grid transmission channels comprises the following steps:

[0145] Step 1: Collect historical observation data and model grid reanalysis data in the selected area of ​​the transmission channel to obtain historical data for prediction;

[0146] Step 2, obtaining the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel;

[0147] Step 3, obtaining a wind field prediction coefficient based on the historical observation data and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel;

[0148] Step 4: predicting the wind field based on the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel and the wind field prediction coefficient.

[0149] The specific implementation method of step 1 is:

[0150] S110, dividing the selected area into longitude and latitude grids at a predetermined distance according to the power transmission channel, and dividing the natural year into time intervals at a predetermined time interval;

[0151] S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years, wherein p in this embodiment is preferably 10 years;

[0152] S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the grid observation data after interpolation is G ij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ij represents the interpolated grid observation data of the jth predetermined time interval at the i-th station location;

[0153] S140, reanalyze the interpolated grid data set G ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is:

[0154]

[0155] The historical observation data includes the ground automatic weather station 10-meter wind direction, the ground automatic weather station 10-meter wind speed, the ground 2-meter temperature of the automatic weather station and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel in each predetermined time interval;

[0156] The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, model reanalysis ground 10-meter wind speed, model reanalysis ground 2-meter temperature and model reanalysis ground pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grid is the horizontal resolution of the grid reanalysis data;

[0157] The specific implementation of the interpolation algorithm in step S130 is:

[0158] If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is:

[0159]

[0160] Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2);

[0161] The interpolated grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij (wdir), interpolated 10-meter wind speed G ij (spd), interpolated 2-meter temperature G ij (T) and interpolated surface pressure G ij (P);

[0162] The specific implementation method of step 2 is:

[0163] Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τ k represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time.

[0164] The specific implementation method of step 3 is:

[0165] S310, using the station observation data variables in the selected area of ​​the transmission channel to estimate the initial guess value CFu of the zonal wind forecast i (t) and the initial guess of the meridional wind forecast CFv i (t), where i represents the i-th site location and t represents the prediction time; the estimation formula is:

[0166]

[0167]

[0168] in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, is the time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, is the time before the prediction of the location of the i-th station The historical moment longitude wind of the time, k represents the ordinal number of the time before the prediction time, Indicates rounding up;

[0169] S320, calculate the comprehensive distance between the initial guess of wind field forecast and all historical reanalysis grid data

[0170]

[0171] In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, represents rounding up, n represents the ordinal number of the four variables, σ n is the weight of the nth variable;

[0172] S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all historical reanalysis grid data ij :

[0173]

[0174] S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates:

[0175]

[0176] in Indicates rounding up;

[0177] Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im , where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij :

[0178]

[0179] The n=1 represents the variable ground 10 meters wind direction, n=2 represents the variable 10 meters wind speed, n=3 represents the variable 2 meters temperature, and n=4 represents the variable ground pressure;

[0180] The specific implementation method of step 4 is:

[0181] Using the historical analysis observation dataset Z ij 10m wind speed Z ij (wsp) and 10-meter wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is:

[0182] Zu ij =-Z ij (wsp)×sin(Z ij (wdir))

[0183] Z ij =-Z ij (wsp)×cos(Z ij (wdir))

[0184] Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is:

[0185]

[0186]

[0187] Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

[0188] The contents not described in detail in this specification belong to the prior art known to those skilled in the art. It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0189] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0190] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0191] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0192] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit its protection scope. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims of the invention.

Claims

1. A wind field prediction system for complex terrain areas of power grid transmission channels, characterized by: It includes a data collection and processing module (1), a prediction area data acquisition module (2), a wind field prediction coefficient acquisition module (3) and a wind field prediction module (4); The data collection and processing module (1) is used to collect historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel to obtain historical data for prediction; The prediction area data acquisition module (2) is used to acquire the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel; The wind field prediction coefficient acquisition module (3) acquires the wind field prediction coefficient based on the historical data for prediction acquired by the collection and processing module (1) and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel of the prediction area data acquisition module (2); The wind field prediction module (4) predicts the wind field based on the site observation data variables of each automatic weather station in the transmission channel selected area of ​​the prediction area data acquisition module (2) and the wind field prediction coefficient acquired by the wind field prediction coefficient acquisition module (3); The specific implementation method of the wind field prediction coefficient acquisition module (3) is: S310, estimating a preliminary guess value of the zonal wind forecast and a preliminary guess value of the meridional wind forecast using station observation data variables in the selected area of ​​the transmission channel; S320, calculating the comprehensive distance between the initial guess values ​​of the zonal wind and meridional wind forecasts of the wind field and all the historical model grid reanalysis data; S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all the historical model grid reanalysis data ij : S340, calculating the wind field prediction coefficient, using the ground altitude, calculating the total number of prediction estimates M, and using the comprehensive distance S ij , select the smallest M distances and construct the nearest distance vector to calculate the wind field prediction coefficient; The specific implementation method of the wind field prediction module (4) is: The wind speed observations and wind direction observations in the historical analysis observation data set are used to calculate the historically observed zonal winds and meridional winds. The predicted zonal winds and meridional winds are calculated using the historically observed zonal winds and meridional winds and wind field prediction coefficients.

2. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 1, characterized in that: S310, using the station observation data variables in the selected area of ​​the transmission channel to estimate the initial guess value CFu of the zonal wind forecast i (t) and the initial guess of the meridional wind forecast CFv i (t), where i represents the location of the i-th station and t represents the prediction time; the estimation formula is: in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, is the time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, The time before the prediction of the location of the i-th station The historical moment longitude wind of the time, k represents the ordinal number of the time before the prediction time, Indicates rounding up. Indicates the ground altitude; S320, calculating the comprehensive distance between the initial guess values ​​of the zonal wind and meridional wind forecasts of the wind field and all the historical model grid reanalysis data; In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, represents rounding up, n represents the ordinal number of the surface wind direction variable, surface wind speed variable, surface temperature variable, and surface air pressure variable, σ n is the weight of the nth variable, Fu i and Fv i denote the predicted zonal and meridional winds, respectively; S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all the historical model grid reanalysis data ij : S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates: in Indicates rounding up; Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im , where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij : The specific implementation method of the wind field prediction module (4) is: Using historical analysis observation dataset Z ij Wind speed Z ij (wsp) and wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is: Zu ij =-Z ij (co)×sin(Z ij (wdir)) Zv ij =-Z ij (co)×cos(Z ij (wdir)) Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is: Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

3. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 1, characterized in that: The specific implementation method of the data collection and processing module (1) is: S110, dividing the selected area into longitude and latitude grids at a predetermined distance according to the power transmission channel, and dividing the natural year into time intervals at a predetermined time interval; S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years; S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the interpolated grid data set is G ij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ij represents the interpolated grid observation data at the jth predetermined time interval at the i-th station location. The interpolated model grid reanalysis dataset is G ij '; S140, reanalyze the data set G for the interpolated model grid ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is:

4. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 3, characterized in that: The historical observation data includes the wind direction at 10 meters above the ground of the automatic weather station, the wind speed at 10 meters above the ground of the automatic weather station, the temperature at 2 meters above the ground of the automatic weather station, and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel within each predetermined time interval; The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, the model reanalysis ground 10-meter wind speed, the model reanalysis ground 2-meter temperature and the model reanalysis ground air pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grids is the horizontal resolution of the grid reanalysis data.

5. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 3, characterized in that: The specific implementation of the interpolation algorithm is as follows: If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is: Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2); The interpolated model grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij (wdir), interpolated 10-meter wind speed G ij (spd), interpolated 2-meter temperature G ij (T) and interpolated surface pressure G ij (P).

6. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 3, characterized in that: The specific implementation method of the prediction area data acquisition module (2) is: Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τ k represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time.

7. A wind field prediction system for complex terrain areas of power grid transmission channels according to claim 1, characterized in that: n=1 represents the variable ground 10m wind direction, n=2 represents the variable 10m wind speed, n=3 represents the variable 2m temperature, and n=4 represents the variable ground air pressure.

8. A method for predicting wind fields in complex terrain areas of power grid transmission channels based on the system of claim 1, characterized in that: It includes the following steps: Step 1: Collect historical observation data and model grid reanalysis data in the selected area of ​​the transmission channel to obtain historical data for prediction; Step 2, obtaining the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel; Step 3, obtaining a wind field prediction coefficient based on the historical data for prediction and the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel; Step 4: predicting the wind field based on the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel and the wind field prediction coefficient.

9. A method for predicting wind fields in complex terrain areas of power grid transmission channels according to claim 7, characterized in that: The specific implementation method of step 1 is: S110, dividing the selected area into longitude and latitude grids at a predetermined distance according to the power transmission channel, and dividing the natural year into time intervals at a predetermined time interval; S120, collecting historical observation data and model grid reanalysis data of the selected area of ​​the transmission channel in the past p years; S130, using an interpolation algorithm to interpolate the model grid point reanalysis data within each predetermined time interval to each automatic weather station site location in the historical observation data; and using the interpolated model grid point reanalysis data and the historical observation data to establish each automatic weather station site location, an interpolated historical analysis observation data set and an interpolated grid point data set for each predetermined time interval Δt; the historical analysis observation data set is Z ij , the interpolated grid data set is G ij , where Z ij represents the observation value of the i-th station at the j-th predetermined time interval, G ij represents the interpolated grid observation data at the jth predetermined time interval at the i-th station location. The interpolated model grid reanalysis dataset is G ij '; S140, reanalyze the data set G for the interpolated model grid ij 'Perform normalization to obtain a normalized data set The formula for the normalization process is: The historical observation data includes the wind direction at 10 meters above the ground of the automatic weather station, the wind speed at 10 meters above the ground of the automatic weather station, the temperature at 2 meters above the ground of the automatic weather station, and the ground air pressure of the automatic weather station observed by the automatic weather station on the transmission channel within each predetermined time interval; The model grid reanalysis data includes the model reanalysis ground 10-meter wind direction, model reanalysis ground 10-meter wind speed, model reanalysis ground 2-meter temperature and model reanalysis ground pressure of the model grid reanalysis within each latitude and longitude grid and each predetermined time interval, wherein the minimum distance of the latitude and longitude grid is the horizontal resolution of the grid reanalysis data; The specific implementation of the interpolation algorithm in step S130 is: If the station location spacing is greater than or equal to the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the nearest neighbor method; if the station location spacing is less than the horizontal resolution of the grid reanalysis data, the interpolation algorithm is selected as the bilinear interpolation method, and the bilinear interpolation formula is: Where Z ij The location of the i-th station is (x, y), and the reanalysis data of the four adjacent grid points are Said The position is (x1, y1), The position is (x2, y1), The position is (x1, y2), The position is (x2,y2); The interpolated model grid reanalysis dataset G ij 'Including the interpolated 10-meter wind direction G ij (wdir), interpolated 10-meter wind speed G ij (spd), interpolated 2-meter temperature G ij (T) and interpolated surface pressure G ij (P); The specific implementation method of step 2 is: Obtain the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel, wherein the site observation data variables of each automatic weather station in the selected area of ​​the transmission channel include zonal wind Meridional wind and ground altitude Where i represents the i-th site location, τ k represents the k-th historical moment before the predicted time obtained according to the predetermined time interval, Δt represents the predetermined time interval, and the k-th historical moment is Δt×k hours before the predicted time; The specific implementation method of step 3 is: S310, using the station observation data variables in the selected area of ​​the transmission channel to estimate the initial guess value CFu of the zonal wind forecast i (t) and the initial guess of the meridional wind forecast CFv i (t), where i represents the location of the i-th station and t represents the prediction time; the estimation formula is: in, is the historical zonal wind at the first hour before the prediction time of the i-th station location, is the historical zonal wind at the k+1th time before the prediction time of the i-th station location, is the historical zonal wind at the kth time before the prediction time of the i-th station location, is the time before the prediction of the location of the i-th station The historical moment of the zonal wind, is the historical meridional wind at the first hour before the prediction time of the i-th station location, is the historical meridional wind at the k+1th time before the prediction time of the i-th station location, is the historical meridional wind at the kth time before the prediction time of the i-th station location, is the time before the prediction of the location of the i-th station The historical moment longitude wind of the time, k represents the ordinal number of the time before the prediction time, Indicates rounding up. Indicates the ground altitude; S320, calculate the comprehensive distance between the initial guess of the zonal wind and meridional wind forecast values ​​and all historical model grid reanalysis data In the formula, represents the value of the nth variable normalized at the i-th station location and the jk-th scheduled time interval, where k represents the ordinal number of the time before the prediction time, represents rounding up, n represents the ordinal number of the surface wind direction variable, surface wind speed variable, surface temperature variable, and surface air pressure variable, σ n is the weight of the nth variable, Fu i and Fv i denote the predicted zonal and meridional winds, respectively; S330, calculate the comprehensive distance S between the initial guess vector of the zonal wind, the initial guess vector of the longitudinal wind and all the historical model grid reanalysis data ij : S340, calculate the wind field prediction coefficient, using the ground altitude Calculate the total number of forecast estimates: in Indicates rounding up; Using the comprehensive distance S ij , select the smallest M distances and construct the closest distance vector Smin im , where i represents the location of the i-th station, m represents the ordinal number of the closest distance, m = 1…M; calculate the wind field prediction coefficient ω ij : The specific implementation method of step 4 is: Using historical analysis observation dataset Z ij Wind speed Z ij (wsp) and wind direction observation value Z ij (wdir), calculate the historically observed zonal wind Zu ij and meridional wind Zv ij , the calculation formula is: Zu ij =-Z ij (co)×sin(Z ij (wdir)) Zv ij =-Z ij (co)×cos(Z ij (wdir)) Calculate the predicted zonal wind Fu i and the meridional wind Fv i , the calculation formula is: Where i represents the location of the i-th site, ω ij is the wind field prediction coefficient.

Images