A method and device for early warning of cut-out of a wind turbine generator set

Abstract

The present application discloses a method and device for early warning of wind turbine generator set cut-out. The method comprises: for a target wind farm, numerically forecasting wind speed for the forecast date using global background field data for the forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain a wind speed ensemble forecast for the forecast date; numerically re-forecasting wind speed using reanalysis data of multiple historical dates, reanalysis data of a set of preset dates before and after multiple historical dates, a preset weather forecast model, and a physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date; the multiple historical dates and the forecast date are the same date in different years. Based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme wind index prediction algorithm based on preset quantiles, a predicted extreme wind index for the forecast date is calculated; and if it is determined that the predicted extreme wind index is greater than the preset extreme wind index, a wind turbine generator set cut-out warning is issued for the target wind farm.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for early warning of cut-out of a wind turbine generator set. Background Art

[0002] Numerical weather forecasting (NWF) involves using large computers to perform numerical calculations based on atmospheric conditions, using certain initial and boundary conditions, to solve the fluid dynamics and thermodynamics equations that describe weather evolution, thereby predicting atmospheric motion and weather phenomena for a specific time period. Currently, wind speed prediction within NWF is a core component of the safe and economical operation of wind turbines in wind farms, and is particularly crucial for early warning of wind turbine shutdowns.

[0003] The essence of numerical wind speed forecasting is to use differential methods to numerically solve a set of fluid mechanics and thermodynamics equations. Even if techniques such as observational data assimilation are used to improve the accuracy of numerical wind speed forecasting, the forecast results of numerical wind speed forecasting are relatively smooth. Extreme strong winds indicate a rapid increase in wind speed, so numerical wind speed forecasting cannot accurately predict extreme strong winds, resulting in an inability to meet the demand for accurate warning of wind turbine shutdown. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a method and apparatus for early warning of wind turbine generator set cut-out, which can accurately predict extreme gale and thus accurately early warning of wind turbine generator set cut-out.

[0005] In a first aspect, an embodiment of the present application provides a method for early warning of cut-out of a wind turbine generator set, the method comprising:

[0006] For a target wind farm, numerically forecasting wind speed based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model is performed to obtain an ensemble forecast of wind speed on the forecast date;

[0007] Performing a numerical wind speed re-forecast based on reanalysis data of multiple historical dates, reanalysis data of multiple sets of preset dates before and after the historical dates, the preset weather forecast model, and the physical parameter perturbation method to construct a historical wind speed simulation sample for the predicted date; the multiple historical dates and the predicted date are the same date in different years;

[0008] Determining a predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile;

[0009] If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine generator set cut-out warning is issued for the target wind farm.

[0010] Optionally, for a target wind farm, performing a numerical wind speed forecast based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain an ensemble wind speed forecast on the forecast date includes:

[0011] determining the preset weather forecast mode;

[0012] determining the physical parameter disturbance mode according to the geographic information of the target wind farm;

[0013] The global background field data is used as the initial field of the preset weather forecast model, and the wind speed numerical forecast is performed in combination with the physical parameter perturbation method to obtain the wind speed ensemble forecast.

[0014] Optionally, the numerical re-forecasting of wind speed based on reanalysis data of multiple historical dates, reanalysis data of multiple sets of preset dates before and after the historical dates, the preset weather forecast mode, and the physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date includes:

[0015] Using the reanalysis data of each of the historical dates as the initial field of the preset weather forecast model, and combining the physical parameter perturbation method to perform wind speed numerical re-forecasting, to obtain a wind speed simulation value for each of the historical dates;

[0016] Using the reanalysis data of each date in the set of preset dates before and after each of the historical dates as the initial field of the preset weather forecast model, and combining the physical parameter perturbation method to perform wind speed numerical re-forecasting, to obtain a wind speed simulation value for each date in the set of preset dates before and after each of the historical dates;

[0017] The historical wind speed simulation samples are constructed based on the wind speed simulation values of multiple historical dates and the wind speed simulation values of multiple dates in a set of preset dates before and after the historical dates.

[0018] Optionally, determining the predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile includes:

[0019] Determining a first cumulative distribution probability corresponding to the preset quantile based on the historical wind speed simulation sample;

[0020] Determining, based on the wind speed ensemble forecast, a second cumulative distribution probability corresponding to the target wind speed threshold, where the target wind speed threshold refers to a wind speed simulation value corresponding to the preset quantile in the historical wind speed simulation sample;

[0021] The predicted extreme gale index is determined according to the first cumulative distribution probability, the second cumulative distribution probability, and the extreme gale index prediction algorithm.

[0022] Optionally, determining the first cumulative distribution probability corresponding to the preset quantile based on the historical wind speed simulation sample includes:

[0023] Sorting the wind speed simulation values in the historical wind speed simulation samples from small to large to obtain a probability distribution of the wind speed simulation values in the historical wind speed simulation samples;

[0024] The first cumulative distribution probability is determined according to the probability distribution and the preset quantile.

[0025] Optionally, the method further includes:

[0026] Setting the preset quantile to different quantiles, wherein the different quantiles correspond to different extreme wind levels;

[0027] The step of determining the predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on a preset quantile comprises:

[0028] Determining, based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on the different quantiles, a plurality of predicted extreme gale indices for the predicted date and a plurality of extreme gale levels corresponding to the plurality of predicted extreme gale indices;

[0029] If the predicted extreme gale index is greater than the preset extreme gale index, issuing a wind turbine generator set cut-out warning for the target wind farm includes:

[0030] If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine generator set cut-out warning is issued to the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index.

[0031] Optionally, the preset weather forecast mode includes a mesoscale weather forecast mode.

[0032] In a second aspect, an embodiment of the present application provides a device for early warning of cut-out of a wind turbine generator set, the device comprising:

[0033] an obtaining unit, configured to perform a numerical wind speed forecast for a target wind farm based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model, to obtain an ensemble wind speed forecast for the forecast date;

[0034] a construction unit, configured to perform a numerical wind speed re-forecast based on reanalysis data of a plurality of historical dates, reanalysis data of a plurality of preset date sets before and after the historical dates, the preset weather forecast mode, and the physical parameter perturbation mode, to construct a historical wind speed simulation sample for the predicted date; the plurality of historical dates and the predicted date being the same date in different years;

[0035] a determination unit, configured to determine a predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile;

[0036] The early warning unit is used to issue a wind turbine generator set cut-out early warning to the target wind farm if the predicted extreme high wind index is greater than a preset extreme high wind index.

[0037] Optionally, the obtaining unit includes a first determining subunit, a second determining subunit and a first obtaining subunit;

[0038] The first determining subunit is configured to determine the preset weather forecast mode;

[0039] The second determining subunit is configured to determine the physical parameter disturbance mode according to the geographic information of the target wind farm;

[0040] The first obtaining subunit is configured to use the global background field data as the initial field of the preset weather forecast model, perform wind speed numerical forecast in combination with the physical parameter perturbation method, and obtain the wind speed ensemble forecast.

[0041] Optionally, the construction unit includes a second obtaining subunit, a third obtaining subunit and a construction subunit;

[0042] The second obtaining subunit is configured to use the reanalysis data of each of the historical dates as the initial field of the preset weather forecast model, perform wind speed numerical re-forecasting in combination with the physical parameter perturbation method, and obtain a wind speed simulation value for each of the historical dates;

[0043] The third obtaining subunit is configured to use the reanalysis data of each date in the set of preset dates before and after each of the historical dates as the initial field of the preset weather forecast model, and to perform wind speed numerical re-forecasting in combination with the physical parameter perturbation method to obtain a wind speed simulation value for each date in the set of preset dates before and after each of the historical dates;

[0044] The construction subunit is used to construct the historical wind speed simulation sample based on the wind speed simulation values of multiple historical dates and the wind speed simulation values of multiple dates in the preset date set before and after the multiple historical dates.

[0045] Optionally, the determining unit includes a third determining subunit, a fourth determining subunit and a fifth determining subunit;

[0046] The third determining subunit is configured to determine a first cumulative distribution probability corresponding to the preset quantile based on the historical wind speed simulation sample;

[0047] The fourth determining subunit is configured to determine, based on the wind speed ensemble forecast, a second cumulative distribution probability corresponding to the target wind speed threshold, where the target wind speed threshold refers to a wind speed simulation value corresponding to the preset quantile in the historical wind speed simulation sample;

[0048] The fifth determining subunit is configured to determine the predicted extreme gale index based on the first cumulative distribution probability, the second cumulative distribution probability, and the extreme gale index prediction algorithm.

[0049] Optionally, the third determining subunit includes an obtaining module and a determining module;

[0050] The obtaining module is used to sort the wind speed simulation values in the historical wind speed simulation samples from small to large, and obtain the probability distribution of the wind speed simulation values in the historical wind speed simulation samples;

[0051] The determination module is configured to determine the first cumulative distribution probability based on the probability distribution and the preset quantile.

[0052] Optionally, the device further includes:

[0053] A setting unit, configured to set the preset quantile to be divided into different quantiles, wherein the different quantiles correspond to different extreme wind levels;

[0054] The determining unit is configured to:

[0055] Determining, based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on the different quantiles, a plurality of predicted extreme gale indices for the predicted date and a plurality of extreme gale levels corresponding to the plurality of predicted extreme gale indices;

[0056] The early warning unit is used to:

[0057] If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine generator set cut-out warning is issued to the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index.

[0058] Optionally, the preset weather forecast mode includes a mesoscale weather forecast mode.

[0059] In a third aspect, an embodiment of the present application provides a computer device, the computer device including a processor and a memory:

[0060] The memory is used to store program code and transmit the program code to the processor;

[0061] The processor is configured to execute the wind turbine generator set cut-out warning method described in the first aspect according to instructions in the program code.

[0062] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the method for early warning of cut-out of a wind turbine generator set described in the first aspect above.

[0063] Compared with the prior art, this application has at least the following advantages:

[0064] According to the technical solution of the embodiment of the present application, for the target wind farm, a numerical wind speed forecast is performed using the global background field data of the forecast date, a preset weather forecast mode, and a physical parameter perturbation method of the preset weather forecast mode to obtain a wind speed ensemble forecast for the forecast date; a numerical wind speed re-forecast is performed using reanalysis data of multiple historical dates, reanalysis data of a set of preset dates before and after multiple historical dates, a preset weather forecast mode, and a physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date; wherein, the multiple historical dates and the forecast date are the same date in different years. Based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on preset quantiles, the predicted extreme gale index for the forecast date is calculated; if it is determined that the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine cut-out warning needs to be issued for the target wind farm. It can be seen that this method uses the quantile method to predict the extreme wind index for the target wind farm through the wind speed ensemble forecast of the forecast date and the historical wind speed simulation sample constructed by reanalysis data, so as to achieve accurate forecast of extreme winds and thus realize accurate warning of wind turbine switching out. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0066] Figure 1 A schematic diagram of a system framework involved in an application scenario in an embodiment of the present application;

[0067] Figure 2A flow chart of a method for early warning of a wind turbine generator set cut-out provided in an embodiment of the present application;

[0068] Figure 3 A schematic structural diagram of a wind turbine generator set cut-out warning device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0069] In order to help those skilled in the art better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0070] Currently, numerical wind speed forecasting essentially relies on using differential methods to numerically solve a system of fluid dynamics and thermodynamics equations. Even with the use of techniques like observational data assimilation to improve wind speed forecast accuracy, the inventors have discovered that numerical wind speed forecasting produces relatively smooth results. However, extreme gales, which represent a rapid increase in wind speed, cannot accurately predict these extreme gales, making it impossible to accurately predict wind turbine shutdowns.

[0071] To solve this problem, in an embodiment of the present application, for a target wind farm, a wind speed numerical forecast is performed using global background field data for the forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain a wind speed ensemble forecast for the forecast date; a wind speed numerical re-forecast is performed using reanalysis data of multiple historical dates, reanalysis data of a preset set of dates before and after multiple historical dates, a preset weather forecast model, and a physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date; wherein, the multiple historical dates and the forecast date are the same date in different years. Based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on preset quantiles, a predicted extreme gale index for the forecast date is calculated; if it is determined that the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine cut-out warning needs to be issued for the target wind farm. It can be seen that this method uses the quantile method to predict the extreme wind index for the target wind farm through the wind speed ensemble forecast of the forecast date and the historical wind speed simulation sample constructed by reanalysis data, so as to achieve accurate forecast of extreme winds and thus realize accurate warning of wind turbine switching out.

[0072] For example, one of the scenarios of the embodiment of the present application may be applied to Figure 1In the scenario shown, the scenario includes a processor 101 and a database 102, wherein the database 102 stores global background field data and reanalysis data. The processor 101 calls the data in the database 102 and uses the embodiment provided in this application to perform a wind turbine generator set cut-out warning.

[0073] First, in the above application scenario, although the action description of the implementation method provided by the embodiment of the present application is executed by the server 102; however, the embodiment of the present application is not restricted in terms of the execution subject, as long as the actions disclosed in the implementation method provided by the embodiment of the present application are executed.

[0074] Secondly, the above scenario is only an example scenario provided by the embodiment of the present application, and the embodiment of the present application is not limited to this scenario.

[0075] The specific implementation of the method and device for wind turbine generator set cut-out warning in the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0076] Exemplary Methods

[0077] See also Figure 2 , shows a flow chart of a method for early warning of wind turbine generator set cut-out in an embodiment of the present application. In this embodiment, the method may include the following steps:

[0078] Step 201: For a target wind farm, numerically forecast wind speed based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain an ensemble forecast of wind speed on the forecast date.

[0079] In the embodiments of the present application, since numerical wind speed forecasting is the core basis for the safe and economical operation of wind turbines in a wind farm, it is particularly important for warning whether to shut down wind turbines; therefore, it is first necessary to perform numerical wind speed forecasting on the wind farm to obtain forecast results. Taking any wind farm as the target wind farm, global background field data for the forecast date can be obtained. A numerical wind speed forecast is then performed using a preset weather forecast model and the physical parameter perturbation method of the preset weather forecast model. A set of different wind speed forecast results for the forecast date are obtained, which serve as the wind speed ensemble forecast for the forecast date.

[0080] When step 201 is specifically implemented, first, it is necessary to determine the weather forecast mode used for numerical wind speed forecasting as the preset weather forecast mode; then, based on the preset weather forecast mode, it is necessary to determine the physical parameter perturbation method of the preset weather forecast mode, and the physical parameter perturbation method is related to the geographical information of the target wind farm; finally, the global background field data of the forecast date is used as the initial field of the determined preset weather forecast mode, and the wind speed numerical forecast is performed in combination with the determined physical parameter perturbation method to obtain a set of different wind speed forecast results for the forecast date, that is, a set of different wind speed forecast values for the forecast date, which constitute the wind speed ensemble forecast for the forecast date. Therefore, in an optional implementation of the embodiment of the present application, step 201 may, for example, include the following steps 2011-2013:

[0081] Step 2011: Determine the preset weather forecast mode.

[0082] Among them, when performing numerical wind speed forecasting for the target wind farm, for example, a new generation of weather forecasting mode that is increasingly widely used in the meteorological field can be adopted, that is, the mesoscale weather forecasting (Weather Research and Forecasting, WRF) mode, which has high accuracy, new schemes, and includes a variety of earth system processes. Therefore, in an optional implementation of an embodiment of the present application, the preset weather forecast mode includes a WRF mode. Of course, the preset weather forecast mode may also be other weather forecast modes, and in the embodiment of the present application, it is not limited that the preset weather forecast mode must be a WRF mode.

[0083] Step 2012: Determine the physical parameter disturbance mode according to the geographic information of the target wind farm.

[0084] Among them, the geographical information such as the topographic information and grid surface information of the location of the target wind farm is related to the parameterization schemes of physical processes such as the boundary layer, near-surface layer, long-wave radiation and short-wave radiation of the WRF model. The sensitivity test method can be used to determine the physical parameter perturbation mode of the WRF model through the combination configuration of different physical perturbation modes.

[0085] Step 2013: Using the global background field data as the initial field of the preset weather forecast model, numerical wind speed forecasting is performed in combination with the physical parameter perturbation method to obtain a wind speed ensemble forecast.

[0086] The implementation method of step 201 obtains the wind speed ensemble forecast for the forecast date, which not only provides the possible wind speed forecast for the forecast date, but also quantitatively estimates the uncertainty of the wind speed forecast. The wind speed ensemble forecast for the forecast date can better predict some local details of the wind speed on the forecast date, so that it can be more accurately judged whether there is a risk of extreme high winds on the forecast date.

[0087] As an example, the global background field data of four representative forecast dates, namely the Global Forecast System (GFS), the Global Environmental Model (GEM), the Global Spectral Model (GSM), and the European Centre for Medium-Range Weather Forecasts (ECMWF), are used as the initial fields of the WRF model respectively. Combined with the physical parameter perturbation method of the WRF model, the wind speed numerical forecast is performed to obtain 60 different wind speed forecast results for the forecast date, that is, 60 different wind speed forecast values for the forecast date, which constitute the wind speed ensemble forecast for the forecast date.

[0088] Step 202: numerically re-forecast the wind speed based on the re-analysis data of multiple historical dates, the re-analysis data of a set of preset dates before and after the multiple historical dates, the preset weather forecast model and the physical parameter perturbation method, and construct a historical wind speed simulation sample for the predicted date; the multiple historical dates and the predicted date are the same date in different years.

[0089] Since the essence of numerical wind speed forecasting is to use the differential method to numerically solve the fluid mechanics and thermodynamics equations, even if techniques such as observational data assimilation are used to improve the accuracy of numerical wind speed forecasting; however, the forecast results of numerical wind speed forecasting are relatively smooth, and extreme strong winds indicate a rapid increase in wind speed. Therefore, numerical wind speed forecasting cannot accurately predict extreme strong winds, resulting in an inability to meet the demand for accurate warning of wind turbine shutdown.

[0090] Therefore, in the embodiment of the present application, on the basis of obtaining the wind speed ensemble forecast for the forecast date in step 201, in order to be able to accurately forecast extreme strong winds in the future and meet the needs of quasi-warning wind turbine switching out; extreme strong winds are relative to the same period in history, it is necessary to comprehensively consider the wind speed ensemble forecast for the forecast date and the relevant wind speed ensemble forecasts for multiple historical dates corresponding to the forecast date, so as to determine whether there are extreme strong winds on the forecast date, it is necessary to obtain the relevant wind speed ensemble forecasts for multiple historical dates corresponding to the forecast date.

[0091] Obtaining the relevant wind speed ensemble forecasts for multiple historical dates corresponding to the predicted date is actually to obtain the global background field data corresponding to the predicted date, obtain the reanalysis data for multiple historical dates, and the reanalysis data for the preset date sets before and after the multiple historical dates, combine the preset weather forecast mode and the physical parameter perturbation method of the preset weather forecast mode to perform wind speed numerical re-forecasting, and obtain the relevant wind speed re-forecast results for each historical date, that is, the relevant wind speed simulation values for each historical date, to form the relevant wind speed ensemble forecast for each historical date; and construct the relevant wind speed ensemble forecasts for multiple historical dates into historical wind speed simulation samples corresponding to the predicted date.

[0092] During the specific implementation of step 202, not only is it necessary to use the reanalysis data of each historical date as the initial field of the preset weather forecast model, and to perform wind speed numerical re-forecasting in combination with the physical parameter perturbation method to obtain the wind speed re-forecast result of each historical date, that is, to obtain the wind speed simulation value of each historical date; but it is also necessary to use the reanalysis data of each date in the preset date set before and after each historical date as the initial field of the preset weather forecast model, and to perform wind speed numerical re-forecasting in combination with the physical parameter perturbation method to obtain the wind speed re-forecast result of each date in the preset date set before and after each historical date, that is, to obtain the wind speed simulation value of each date in the preset date set before and after each historical date. Based on this, the wind speed simulation value of each historical date and the wind speed simulation value of each date in the preset date set before and after the historical date constitute the relevant wind speed simulation value of each historical date, that is, the relevant wind speed set forecast for each historical date; then the relevant wind speed set forecast for multiple historical dates constructs a historical wind speed simulation sample corresponding to the forecast date, which means that the wind speed simulation values of multiple historical dates and the wind speed simulation values of multiple dates in the preset date set before and after the multiple historical dates construct a historical wind speed simulation sample corresponding to the forecast date. Therefore, in an optional implementation of the embodiment of the present application, step 202 can, for example, include the following steps 2021-2023:

[0093] Step 2021: Use the reanalysis data of each historical date as the initial field of the preset weather forecast model, combine the physical parameter perturbation method to re-forecast the wind speed value, and obtain the wind speed simulation value of each historical date.

[0094] Step 2022: Use the reanalysis data of each date in the preset date set before and after each historical date as the initial field of the preset weather forecast model, and re-forecast the wind speed value in combination with the physical parameter perturbation method to obtain the wind speed simulation value for each date in the preset date set before and after each historical date.

[0095] Step 2023: Construct a historical wind speed simulation sample based on the wind speed simulation values of multiple historical dates and the wind speed simulation values of multiple dates in a preset date set before and after the multiple historical dates.

[0096] A historical wind speed simulation sample is constructed through the implementation method of step 202. The historical wind speed simulation sample is a wind speed numerical re-forecast based on the reanalysis data, that is, it is obtained by simulating and back-calculating historical data. The wind speed ensemble forecast relative to the forecast date includes a variety of wind speed simulation values and wind speed change situations, and is a wind speed simulation sample with very high credibility.

[0097] As an example, based on the above example, the US Final Operational Global Analysis Data (FNL), Climate Forecast System Reanalysis Data (CFSR), European Center for Meteorological Research (ECMWF) Reanalysis Data Version 5 (ERA-5), Japan Meteorological Agency 55-year Reanalysis Data (JRA-55), and NASA 2nd Generation Reanalysis Data (MERRA-2) for multiple historical dates in the past 30 years, as well as the above reanalysis data for 20 days before and after multiple historical dates, are used as the initial fields of the WRF model. Combined with the WRF model's physical parameter perturbation method, wind speed numerical reforecasts are performed. 30 × 41 × 60 = 73,800 different wind speed simulation values corresponding to the prediction date are obtained to construct a historical wind speed simulation sample for the prediction date.

[0098] Step 203: Determine the predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation samples, and the extreme gale index prediction algorithm based on the preset quantile.

[0099] In an embodiment of the present application, after obtaining the wind speed ensemble forecast for the forecast date in step 201 and constructing the historical wind speed simulation sample for the forecast date in step 202, the wind speed ensemble forecast for the forecast date and the wind speed ensemble forecasts for multiple historical dates corresponding to the forecast date are comprehensively considered to determine whether there will be extreme strong winds on the forecast date. In fact, it means that based on the wind speed ensemble forecast for the forecast date and the historical wind speed simulation samples for the forecast date, combined with the extreme strong wind index prediction algorithm based on the preset quantile, the predicted extreme strong wind index for the forecast date under the preset quantile is calculated to determine whether there will be extreme strong winds on the forecast date.

[0100] When step 203 is specifically implemented, first, the first cumulative distribution probability corresponding to the preset quantile can be determined in the historical wind speed simulation sample of the preset date; then, by taking the wind speed simulation value corresponding to the preset quantile in the historical wind speed simulation sample as the target wind speed threshold, the second cumulative distribution probability corresponding to the target wind speed threshold is determined in the wind speed ensemble forecast of the preset date; finally, the first cumulative distribution probability and the second cumulative distribution probability are input into the extreme gale index prediction algorithm, and the difference between the first cumulative distribution probability and the second cumulative distribution probability is calculated to determine the predicted extreme gale index. The greater the difference, the greater the predicted extreme gale index, and the more likely there is an extreme gale risk. Therefore, in an optional implementation of the embodiment of the present application, step 203 may, for example, include the following steps 2031-2033:

[0101] Step 2031: Determine a first cumulative distribution probability corresponding to a preset quantile based on historical wind speed simulation samples.

[0102] Among them, when step 2031 is specifically implemented, first, the wind speed simulation values in the historical wind speed simulation sample of the preset date need to be sorted in order from small to large, and the number of each wind speed simulation value and the total number of each wind speed simulation value are counted. Based on this, the probability distribution of the wind speed simulation values in the historical wind speed simulation sample is obtained; then, based on the probability distribution of the wind speed simulation values in the historical wind speed simulation sample, the cumulative distribution probability of the wind speed simulation values in the historical wind speed simulation sample can be obtained, and the cumulative distribution probability corresponding to the preset quantile is determined as the first cumulative distribution probability. Therefore, in an optional implementation of the embodiment of the present application, step 2031, for example, may include the following steps A-B:

[0103] Step A: sorting the wind speed simulation values in the historical wind speed simulation samples from small to large to obtain the probability distribution of the wind speed simulation values in the historical wind speed simulation samples;

[0104] Step B: Determine a first cumulative distribution probability based on the probability distribution and a preset quantile.

[0105] Step 2032: Based on the wind speed ensemble forecast, determine a second cumulative distribution probability corresponding to a target wind speed threshold, where the target wind speed threshold refers to a wind speed simulation value corresponding to a preset quantile in a historical wind speed simulation sample.

[0106] Among them, on the basis of the cumulative distribution probability of the wind speed simulation values in the above-mentioned historical wind speed simulation samples, the wind speed simulation values corresponding to the preset quantiles in the historical wind speed simulation samples can be first obtained as the target wind speed threshold; then, the wind speed forecast values in the wind speed ensemble forecast for the preset date are sorted in ascending order, and the number of each wind speed forecast value and the total number of each wind speed forecast value are counted, based on which the probability distribution of the wind speed forecast values in the wind speed ensemble forecast is obtained; then, with the probability distribution of the wind speed forecast values in the wind speed ensemble forecast, the cumulative distribution probability of the wind speed forecast values in the wind speed ensemble forecast can be obtained, and the cumulative distribution probability corresponding to the target wind speed threshold is determined as the second cumulative distribution probability.

[0107] Step 2033: Determine the predicted extreme gale index based on the first cumulative distribution probability, the second cumulative distribution probability, and the extreme gale index prediction algorithm.

[0108] As an example, the extreme wind index prediction algorithm uses the following formula:

[0109]

[0110] Where p represents the first cumulative distribution probability corresponding to the preset quantile p in the historical wind speed simulation sample, W(p) represents the second cumulative distribution probability corresponding to the target wind speed threshold in the wind speed ensemble forecast, and the target wind speed threshold refers to the wind speed simulation value corresponding to the preset quantile p in the historical wind speed simulation sample; Wind cut Indicates the predicted extreme wind index.

[0111] Step 204: If the predicted extreme high wind index is greater than the preset extreme high wind index, a wind turbine generator set cut-out warning is issued for the target wind farm.

[0112] In the embodiment of the present application, after the predicted extreme gale index for the predicted date is determined in step 203, on the basis that the larger the predicted extreme gale index is, the more likely the extreme gale risk is, a lower limit of the extreme gale index indicating the existence of an extreme gale risk is pre-set for the value range of the extreme gale index as the preset extreme gale index; and a determination is made as to whether the predicted extreme gale index is greater than the preset extreme gale index. If so, it indicates that the target wind farm has an extreme gale risk on the predicted date, and a wind turbine generator set cut-out warning needs to be issued to the target wind farm.

[0113] In addition, in the embodiment of the present application, when different preset quantiles are selected, the levels of extreme gale risk are different, that is, different quantiles correspond to different extreme gale levels; based on this, step 203 adopts an extreme gale index prediction algorithm based on different quantiles to determine multiple predicted extreme gale indices and their corresponding multiple extreme gale levels on the predicted date; thus, step 204 determines that the predicted extreme gale index is greater than the preset extreme gale index, and it is necessary to perform a wind turbine cut-out warning for the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index, so as to realize wind turbine cut-out warning for different extreme gale levels.

[0114] That is, in an optional implementation of the embodiment of the present application, the method may further include the following step C: setting the preset quantile to be divided into different quantiles, and different quantiles correspond to different extreme gale levels; correspondingly, step 203 may include, for example: determining multiple predicted extreme gale indices for the predicted date and multiple extreme gale levels corresponding to the multiple predicted extreme gale indices based on the wind speed ensemble forecast, historical wind speed simulation samples and extreme gale index prediction algorithms based on different quantiles; step 204 may include, for example: if the predicted extreme gale index is greater than the preset extreme gale index, according to the extreme gale level corresponding to the predicted extreme gale index, a wind turbine cut-out warning is issued for the target wind farm.

[0115] Through the various implementation methods provided in this embodiment, for a target wind farm, a numerical wind speed forecast is performed using global background field data for the forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain a wind speed ensemble forecast for the forecast date. A numerical wind speed re-forecast is performed using reanalysis data of multiple historical dates, reanalysis data of a set of preset dates before and after multiple historical dates, a preset weather forecast model, and a physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date. The multiple historical dates and the forecast date are the same date in different years. Based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on preset quantiles, a predicted extreme gale index for the forecast date is calculated. If the predicted extreme gale index is determined to be greater than the preset extreme gale index, a wind turbine cut-out warning is required for the target wind farm. It can be seen that this method uses the quantile method to predict the extreme wind index for the target wind farm through the wind speed ensemble forecast of the forecast date and the historical wind speed simulation sample constructed by reanalysis data, so as to achieve accurate forecast of extreme winds and thus realize accurate warning of wind turbine switching out.

[0116] Exemplary devices

[0117] See also Figure 3, shows a schematic structural diagram of a device for early warning of a wind turbine generator set cut-out in an embodiment of the present application. In this embodiment, the device may specifically include:

[0118] An obtaining unit 301 is configured to perform a numerical wind speed forecast for a target wind farm based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model, to obtain an ensemble wind speed forecast for the forecast date.

[0119] A construction unit 302 is configured to perform a numerical wind speed re-forecast based on reanalysis data of multiple historical dates, reanalysis data of a set of preset dates before and after the multiple historical dates, a preset weather forecast model, and a physical parameter perturbation method, to construct a historical wind speed simulation sample for the predicted date; the multiple historical dates and the predicted date are the same date in different years;

[0120] A determination unit 303 is configured to determine a predicted extreme gale index for a predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile;

[0121] The early warning unit 304 is configured to issue a wind turbine generator set cut-out early warning to the target wind farm if the predicted extreme high wind index is greater than a preset extreme high wind index.

[0122] In an optional implementation of the embodiment of the present application, the obtaining unit 301 includes a first determining subunit, a second determining subunit, and a first obtaining subunit;

[0123] A first determining subunit, configured to determine a preset weather forecast mode;

[0124] A second determining subunit is configured to determine a physical parameter disturbance mode based on geographic information of the target wind farm;

[0125] The first acquisition subunit is used to use the global background field data as the initial field of the preset weather forecast model, combine the physical parameter perturbation method to perform wind speed numerical forecast, and obtain the wind speed ensemble forecast.

[0126] In an optional implementation of the embodiment of the present application, the construction unit 302 includes a second obtaining subunit, a third obtaining subunit and a construction subunit;

[0127] The second acquisition subunit is used to use the reanalysis data of each historical date as the initial field of the preset weather forecast model, combine the physical parameter perturbation method to perform wind speed numerical re-forecast, and obtain the wind speed simulation value for each historical date;

[0128] The third obtaining subunit is used to use the reanalysis data of each date in the set of preset dates before and after each historical date as the initial field of the preset weather forecast model, and to re-forecast the wind speed value in combination with the physical parameter perturbation method to obtain the wind speed simulation value for each date in the set of preset dates before and after each historical date;

[0129] The construction subunit is used to construct historical wind speed simulation samples based on wind speed simulation values of multiple historical dates and wind speed simulation values of multiple dates in a preset date set before and after the multiple historical dates.

[0130] In an optional implementation of the embodiment of the present application, the determining unit 303 includes a third determining subunit, a fourth determining subunit, and a fifth determining subunit;

[0131] A third determining subunit is configured to determine a first cumulative distribution probability corresponding to a preset quantile based on historical wind speed simulation samples;

[0132] a fourth determining subunit, configured to determine, based on the wind speed ensemble forecast, a second cumulative distribution probability corresponding to a target wind speed threshold, where the target wind speed threshold refers to a wind speed simulation value corresponding to a preset quantile in a historical wind speed simulation sample;

[0133] The fifth determining subunit is configured to determine a predicted extreme gale index based on the first cumulative distribution probability, the second cumulative distribution probability, and an extreme gale index prediction algorithm.

[0134] In an optional implementation of the embodiment of the present application, the third determining subunit includes an obtaining module and a determining module;

[0135] An acquisition module is used to sort the wind speed simulation values in the historical wind speed simulation samples from small to large, and obtain the probability distribution of the wind speed simulation values in the historical wind speed simulation samples;

[0136] The determining module is used to determine a first cumulative distribution probability according to the probability distribution and a preset quantile.

[0137] In an optional implementation of the embodiment of the present application, the device further includes:

[0138] A setting unit, used to set the preset quantile to be divided into different quantiles, and different quantiles correspond to different extreme wind levels;

[0139] The determining unit 303 is configured to:

[0140] Based on the wind speed ensemble forecast, historical wind speed simulation samples and extreme gale index prediction algorithms based on different quantiles, multiple predicted extreme gale indices for the predicted date and multiple extreme gale levels corresponding to the multiple predicted extreme gale indices are determined;

[0141] The early warning unit 304 is configured to:

[0142] If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine cut-out warning will be issued to the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index.

[0143] In an optional implementation of the embodiment of the present application, the preset weather forecast mode includes a mesoscale weather forecast mode.

[0144] Through the various implementation methods provided in this embodiment, for a target wind farm, a numerical wind speed forecast is performed using global background field data for the forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain a wind speed ensemble forecast for the forecast date. A numerical wind speed re-forecast is performed using reanalysis data of multiple historical dates, reanalysis data of a set of preset dates before and after multiple historical dates, a preset weather forecast model, and a physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date. The multiple historical dates and the forecast date are the same date in different years. Based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on preset quantiles, a predicted extreme gale index for the forecast date is calculated. If the predicted extreme gale index is determined to be greater than the preset extreme gale index, a wind turbine cut-out warning is required for the target wind farm. It can be seen that this method uses the quantile method to predict the extreme wind index for the target wind farm through the wind speed ensemble forecast of the forecast date and the historical wind speed simulation sample constructed by reanalysis data, so as to achieve accurate forecast of extreme winds and thus realize accurate warning of wind turbine switching out.

[0145] The present application also provides a computer device, comprising a processor and a memory.

[0146] The memory is used to store program code and transmit the program code to the processor;

[0147] The processor is configured to execute the wind turbine generator set cut-out warning method described in the above method embodiment according to the instructions in the program code.

[0148] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the method for early warning of cut-out of a wind turbine generator set described in the above method embodiment.

[0149] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0150] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0151] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0152] The above description is only a preferred embodiment of the present application and does not constitute any formal limitation to the present application. Although the present application has been disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the art can use the above-disclosed methods and technical contents to make many possible changes and modifications to the technical solution of the present application without departing from the scope of the technical solution of the present application, or modify it into an equivalent embodiment with equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application still falls within the scope of protection of the technical solution of the present application.

Claims

1. A method for early warning of wind turbine generator set cut-out, characterized in that: include: For a target wind farm, numerically forecasting wind speed based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model is performed to obtain an ensemble forecast of wind speed on the forecast date; Performing a numerical wind speed re-forecast based on reanalysis data of multiple historical dates, reanalysis data of multiple sets of preset dates before and after the historical dates, the preset weather forecast model, and the physical parameter perturbation method to construct a historical wind speed simulation sample for the predicted date; the multiple historical dates and the predicted date are the same date in different years; Determining a predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile; If the predicted extreme gale index is greater than the preset extreme gale index, issuing a wind turbine generator set cut-out warning for the target wind farm; The method further comprises: Setting the preset quantile to different quantiles, wherein the different quantiles correspond to different extreme wind levels; The step of determining the predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on a preset quantile comprises: Determining, based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on the different quantiles, a plurality of predicted extreme gale indices for the predicted date and a plurality of extreme gale levels corresponding to the plurality of predicted extreme gale indices; If the predicted extreme gale index is greater than the preset extreme gale index, issuing a wind turbine generator set cut-out warning for the target wind farm includes: If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine generator set cut-out warning is issued to the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index.

2. The method according to claim 1, characterized in that The method of performing a numerical wind speed forecast for a target wind farm based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model to obtain an ensemble wind speed forecast on the forecast date includes: determining the preset weather forecast mode; determining the physical parameter disturbance mode according to the geographic information of the target wind farm; The global background field data is used as the initial field of the preset weather forecast model, and the wind speed numerical forecast is performed in combination with the physical parameter perturbation method to obtain the wind speed ensemble forecast.

3. The method according to claim 1, characterized in that The method of performing numerical wind speed re-forecasting based on re-analysis data of multiple historical dates, re-analysis data of multiple sets of preset dates before and after the historical dates, the preset weather forecast mode, and the physical parameter perturbation method to construct a historical wind speed simulation sample for the forecast date includes: Using the reanalysis data of each of the historical dates as the initial field of the preset weather forecast model, and combining the physical parameter perturbation method to perform wind speed numerical re-forecasting, to obtain a wind speed simulation value for each of the historical dates; Using the reanalysis data of each date in the set of preset dates before and after each of the historical dates as the initial field of the preset weather forecast model, and combining the physical parameter perturbation method to perform wind speed numerical re-forecasting, to obtain a wind speed simulation value for each date in the set of preset dates before and after each of the historical dates; The historical wind speed simulation samples are constructed based on the wind speed simulation values of multiple historical dates and the wind speed simulation values of multiple dates in a set of preset dates before and after the historical dates.

4. The method according to claim 3, characterized in that The step of determining the predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on a preset quantile comprises: Determining a first cumulative distribution probability corresponding to the preset quantile based on the historical wind speed simulation sample; Determining, based on the wind speed ensemble forecast, a second cumulative distribution probability corresponding to a target wind speed threshold, wherein the target wind speed threshold refers to a wind speed simulation value corresponding to the preset quantile in the historical wind speed simulation sample; The predicted extreme gale index is determined according to the first cumulative distribution probability, the second cumulative distribution probability, and the extreme gale index prediction algorithm.

5. The method according to claim 4, characterized in that The determining, based on the historical wind speed simulation sample, a first cumulative distribution probability corresponding to the preset quantile includes: Sorting the wind speed simulation values in the historical wind speed simulation samples from small to large to obtain a probability distribution of the wind speed simulation values in the historical wind speed simulation samples; The first cumulative distribution probability is determined according to the probability distribution and the preset quantile.

6. The method according to any one of claims 1 to 5, characterized in that The preset weather forecast mode includes a mesoscale weather forecast mode.

7. A device for early warning of cut-out of a wind turbine generator set, characterized in that: include: an obtaining unit, configured to perform a numerical wind speed forecast for a target wind farm based on global background field data on a forecast date, a preset weather forecast model, and a physical parameter perturbation method of the preset weather forecast model, to obtain an ensemble wind speed forecast for the forecast date; a construction unit, configured to perform a numerical wind speed re-forecast based on reanalysis data of a plurality of historical dates, reanalysis data of a plurality of preset date sets before and after the historical dates, the preset weather forecast mode, and the physical parameter perturbation mode, to construct a historical wind speed simulation sample for the predicted date; the plurality of historical dates and the predicted date being the same date in different years; a determination unit, configured to determine a predicted extreme gale index for the predicted date based on the wind speed ensemble forecast, the historical wind speed simulation sample, and an extreme gale index prediction algorithm based on a preset quantile; an early warning unit, configured to issue a wind turbine generator set cut-out early warning to the target wind farm if the predicted extreme high wind index is greater than a preset extreme high wind index; The device further comprises: A setting unit, configured to set the preset quantile to be divided into different quantiles, wherein the different quantiles correspond to different extreme wind levels; The determining unit is configured to: Determining, based on the wind speed ensemble forecast, the historical wind speed simulation sample, and the extreme gale index prediction algorithm based on the different quantiles, a plurality of predicted extreme gale indices for the predicted date and a plurality of extreme gale levels corresponding to the plurality of predicted extreme gale indices; The early warning unit is used to: If the predicted extreme gale index is greater than the preset extreme gale index, a wind turbine generator set cut-out warning is issued to the target wind farm according to the extreme gale level corresponding to the predicted extreme gale index.

8. A computer device, characterized in that: The computer device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the method for early warning of cut-out of a wind turbine generator set according to any one of claims 1 to 6 according to instructions in the program code.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program codes, and the program codes are used to execute the method for early warning of cut-out of a wind turbine generator set according to any one of claims 1 to 6.

Images