Method for estimating a standardized wind energy resource drought index based on reanalysis data

Abstract

The application discloses an estimation method of a standardized wind energy resource drought index based on reanalysis data, and comprises the following steps: S1, defining the standardized wind energy resource drought index SWRI by wind resources, wherein the wind resources are represented by wind energy density; S2, dividing drought grades according to the defined standardized wind energy resource drought index; S3, calculating the standardized wind energy resource drought index of a target region; and S4, determining the drought grade of the target region. The application proposes a brand-new standardized wind energy resource drought index, which is based on high-resolution reanalysis wind speed data, is converted into wind energy density variation to represent drought conditions, has the advantages of simple algorithm, easy-to-obtain required observation data, and the ability to accurately describe the severity of drought in wind energy resources and wind power generation systems, and can be applied to drought monitoring in aspects of wind energy resource assessment, wind power plant site selection, wind power plant operation, energy planning and risk management.

Description

Technical Field

[0001] This invention relates to a standardized wind energy resource drought index, and more particularly to a method for estimating a standardized wind energy resource drought index based on reanalysis data. Background Technology

[0002] Drought is a frequent and highly destructive extreme climate event that severely impacts the ecological environment, agricultural production, economic activities, and social life. Existing drought indices used to monitor drought intensity include the Standardized Precipitation Index (SPI), the Standardized Precipitation Evapotranspiration Index (SPEI), and the Palmer Drought Index (PDSI). The algorithms for these drought indices are as follows:

[0003] (1) Standardized Precipitation Index (SPI)

[0004]

[0005] Where P is the cumulative precipitation in month n. σ represents the average monthly precipitation over many years for the same period in month n. P denoted as , where is the standard deviation of monthly precipitation over many years in month n. The Standardized Precipitation Index (SPI) only reflects drought conditions caused by insufficient atmospheric precipitation supply. Since precipitation data typically follows a gamma distribution, in practical applications, the precipitation data is first fitted to a gamma distribution and then standardized to calculate the SPI value.

[0006] (2) Standardized Precipitation Evapotranspiration Index (SPEI)

[0007]

[0008] or

[0009]

[0010] Where D = P - PET, is the difference between precipitation P and potential evapotranspiration PET. The SPEI index is the average of the differences over many years; since the original data sequence D may contain negative values, a three-parameter log-logistic probability distribution is used:

[0011]

[0012] Where a, β, and γ are parameters. This index takes into account both precipitation and evapotranspiration, but there are various methods for calculating PET (such as the Penman-Monteith method, the Thornthwaite method, etc.), and the results of different methods vary greatly, resulting in a large degree of uncertainty in the SPEI results.

[0013] (3) Palmer Drought Index (PDSI)

[0014]

[0015] Among them, c i Z is the weighting coefficient. i The Palmer Drought Index (PDSI) is a moisture anomaly index that reflects the difference between water supply and demand. Calculation of the PDSI is complex, involving various meteorological and soil data. Due to significant differences in climate and soil characteristics across different regions, the applicability of the PDSI is limited.

[0016] In summary, although there are various existing drought indices, they mainly focus on the degree of drought in water resources, with very little attention paid to the monitoring of wind energy resources. Therefore, they cannot be effectively applied to drought monitoring in areas such as wind energy resource assessment, wind farm site selection, wind farm operation, energy planning, and risk management. Summary of the Invention

[0017] To address the shortcomings of the aforementioned technologies, this invention provides a method for estimating a standardized wind energy resource drought index based on reanalysis data.

[0018] To address the above technical problems, the technical solution adopted in this invention is: a method for estimating the standardized wind energy resource drought index based on reanalysis data, comprising the following calculation steps:

[0019] S1. The standardized wind energy resource drought index SWRI is defined by wind energy resources, and wind energy resources are characterized by wind energy density.

[0020] S2. Based on the defined standardized wind energy resource drought index, drought levels are classified into mild drought, moderate drought, severe drought, and extreme drought.

[0021] S3. Calculate the standardized wind energy resource drought index for the target area: Using hourly wind speed time data from the reanalysis dataset, calculate the hourly wind energy density for each grid cell; then obtain the moving average time series of wind energy resource changes at this time scale, and calculate the standardized wind energy resource drought index at different time scales.

[0022] S4. Determine the drought level of the target area based on the drought level defined in S2 and the standardized wind energy resource drought index of the target area calculated in S3.

[0023] Preferably, in S1, the characterization formula for the standardized wind energy resource drought index (SWRI) is as follows:

[0024]

[0025] Wherein, WR is the time series of wind energy density variation at different time scales in a certain region, with the unit of W m -2 , and the different time scales usually refer to daily or monthly; is the multi-year average value of the wind energy density variation on this time scale; σ WR is the standard deviation of the multi-year wind energy density variation on this time scale;

[0026] or

[0027]

[0028] Wherein, W is a fitting function that the time series of WR at the daily or multi-day time scale follows a Weibull distribution, a log-normal distribution or a Gamma distribution; is the multi-year average value of the W time series on this time scale; σ W is the standard deviation of the multi-year W time series on this time scale.

[0029] As a preference, in S2, the division criteria for drought grades are as follows:

[0030] When -1.0<SWRI≤-0.5, it is mild drought;

[0031] When -1.5<SWRI≤-1.0, it is moderate drought;

[0032] When -2.0<SWRI≤-1.5, it is severe drought;

[0033] When SWRI≤-2.0, it is extreme drought.

[0034] As a preference, in S3, the calculation of the standardized wind resource index SWRI of the target region at different time scales includes the following steps:

[0035] S31. Acquire the hourly wind speed time data v of the reanalysis dataset t , and calculate the hourly power density time series WR of the wind energy resource for each grid cell t , with the calculation formula as follows:

[0036]

[0037] Wherein, ρ is the air density, which is usually taken as 1.225kg m 3 ; v t is the wind speed time data at a height of 100 meters, with the unit of m s -1 ;

[0038] S32. According to the application requirement of drought monitoring at a certain time scale, calculate the moving average time series of wind energy resource variation on this time scale according to Formula 3;

[0039] S33. The standardized wind energy resource drought index at different time scales is calculated from the moving average time series of wind energy resource changes at different time scales.

[0040] Preferably, in S33, if the changes in wind resources follow a normal distribution on a certain time scale, the standardized wind energy resource drought index SWRI is calculated according to Formula 1.

[0041] If the changes in wind resources follow other distributions at a certain time scale, including the Weiber distribution, the log-normal distribution, or the gamma distribution, the standardized wind resource drought index (SWRI) that follows the corresponding distribution can be calculated using Formula 2.

[0042] This invention discloses a method for estimating a standardized wind energy resource drought index based on reanalysis data. It proposes a novel standardized wind energy resource drought index, which is based on high-resolution reanalysis wind speed data and transforms it into wind energy density changes to characterize drought conditions. It has the advantages of simple algorithm, easy acquisition of required observation data, and accurate description of the severity of drought in wind energy resources and wind power generation systems. It can be applied to drought monitoring in wind energy resource assessment, wind farm site selection, wind farm operation, energy planning and risk management. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the standardized wind energy resource drought index calculation process for the target area of ​​this invention.

[0044] Figure 2 This is a time series diagram of SWRI in Northeast China from 1004 to 2023 at different time scales, as shown in this embodiment of the invention.

[0045] Figure 3 This is a spatial distribution map of SWRI in Northeast China from January to March 2014, as shown in this embodiment of the invention. Detailed Implementation

[0046] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0047] The specific technical solution of the present invention is as follows:

[0048] S1. Define the standardized wind energy resource drought index:

[0049] This invention defines a new drought index called the Standardized Wind Resource Index (SWRI), whose formula is as follows:

[0050]

[0051] Wherein, WR is the time series of wind energy density variation at different time scales in a certain region, and the unit is W m -2 , and the different time scales usually refer to day or month; is the multi-year average value of the wind energy density variation on this time scale; σ WR is the multi-year standard deviation of the wind energy density variation on this time scale.

[0052] or

[0053]

[0054] Wherein, W is the fitting function of the time series of WR at the daily or multi-day time scale following Weibull distribution, log-normal distribution (Norm) or Gamma distribution; is the multi-year average value of the W time series on this time scale; σ W is the multi-year standard deviation of the W time series on this time scale.

[0055] S2: Drought grade division is performed according to the defined standardized wind resource drought index:

[0056] When the standardized wind resource drought index SWRI is greater than -0.5, it indicates that the wind energy resource is in the normal range, that is, the wind energy resource is in an abundant state; on the contrary, when the standardized wind resource drought index SWRI is less than -0.5, it indicates that the wind energy resource is lower than the normal value, that is, the wind energy resource is scarce. The magnitude of SWRI represents the degree of deviation from the normal value. According to the drought system classification proposed by the World Meteorological Organization (WMO), the drought grades are divided into no drought, light drought, moderate drought, severe drought and extreme drought, and the separation points between different drought grades are 0.5σ, 1.0σ, 1.5σ and 2σ respectively, where σ is the standard deviation.

[0057] Since SWRI follows the standard normal distribution (μ=0; σ=1), the drought grade division results are shown in Table 1. When -1.0<SWRI≤-0.5, it is light drought; when -1.5<SWRI≤-1.0, it is moderate drought; when -2.0<SWRI≤-1.5, it is severe drought; when SWRI≤-2.0, it is extreme drought. Moreover, according to the characteristics of the normal distribution, the probabilities of light drought, moderate drought, severe drought and extreme drought are 15.0%, 9.2%, 4.4% and 2.3% respectively. The return periods of light, moderate, severe and extreme drought in drought events are about 10, 15, 20 and 50 years respectively.

[0058] Table 1. Drought grade division of standardized wind resource index (SWRI)

[0059] Mild drought -1.0<SWRI≤-0.5 15.0 10 Moderate drought -1.5<SWRI≤-1.0 9.2 15 Severe drought -2.0<SWRI≤-1.5 4.4 20 Extreme drought SWRI≤-2.0 2.3 50

[0060] S3. Calculate the standardized wind energy resource drought index for the target area;

[0061] The calculation process of the standardized wind resource index (SWRI) for the target area of ​​this invention at different time scales (such as daily or monthly scales) is as follows: Figure 1 As shown, the calculation steps are as follows:

[0062] S31. Obtain hourly wind speed time data v from the reanalysis dataset (such as ERA5 wind speed time series data WS). t Calculate the hourly power density time series WR of wind energy resources for each grid cell. t The calculation formula is as follows:

[0063]

[0064] Where ρ is the air density, typically taken as 1.225 kg m³. 3 ,v t This is time-varying wind speed data at a height of 100 meters, in milliseconds (ms). -1 .

[0065] S32. Based on the drought monitoring application requirements at a certain time scale (such as daily or monthly time scale), calculate the moving average time series of wind energy resource changes at that time scale according to Formula 3, that is, calculate the standardized time series of WPD at different time scales for each grid unit.

[0066] S33. Calculate the standardized wind energy resource drought index at different time scales from the moving average time series of wind energy resource changes at different time scales.

[0067] Specifically, if the wind energy resource change (WR) follows a normal distribution on a certain time scale, the standardized wind energy resource drought index SWRI is calculated according to Formula 1.

[0068] If the wind energy resource change (WR) follows other distributions at a certain time scale, such as the Weibull distribution, the log-normal distribution, or the gamma distribution, then the distribution function is first used to fit the wind energy resource data to generate a new time series. Then, the time series is standardized, and the standardized wind energy resource drought index SWRI that follows the corresponding distribution is calculated by Equation 2.

[0069] S4. Determine the drought level of the target area based on the drought level defined in S2 and the standardized wind energy resource drought index of the target area calculated in S3.

[0070] In summary, the present invention defines a new drought index, namely the Standardized Wind Resource Drought Index, which is characterized by a standardized time series of wind energy density. The Standardized Wind Resource Drought Index (SWRI-n) at a certain time scale (n) is obtained through standardized calculation of the average time series of wind energy density variation at that time scale. Different from existing hydrological drought indices, such as the Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI), and Palmer Drought Severity Index (PDSI), the SWRI drought index of the present invention is mainly used to characterize the drought condition of wind energy resources, and is currently the only index describing wind energy resource drought. It is more suitable for drought monitoring in wind energy resource assessment and wind power production, and has the advantages of simple calculation, easy accessibility, and abundant long-term series observation data in historical periods.

[0071] It should also be noted that the present invention uses wind energy density to characterize wind resources (WR), but wind speed or capacity factor can also be used to characterize wind resources according to research needs. According to the distribution that the wind speed or capacity factor conforms to, the data is fitted to generate a new time series, and then the time series is standardized to obtain the Standardized Wind Resource Index (SWRI) that obeys the corresponding distribution.

[0072] Example:

[0073] The method for estimating the Standardized Wind Resource Drought Index based on reanalysis data proposed by the present invention will be further described below in combination with specific application examples.

[0074] In this example, the change of wind energy resources in Northeast China (accounting for nearly 20% of the total installed capacity), which has the highest proportion of wind power installed capacity in China from 1994 to 2023, is taken as an application example of drought monitoring. Figure 2 The time series comparison of the Standardized Wind Resource Drought Index (SWRI) at time scales of 1 month, 3 months and 6 months in Northeast China is shown. In the figure, the periods covered by the negative red areas are drought periods. From Figure 2 it can be seen that SWRI characterizes at least 6 drought events since 1994, occurring in 2015, 2000, 2002-2003, 2008, 2011-2022 and 2024. Among them, severe drought events occurred in 2012 and 2014 (-2.0<SWRI-1≤-1.5). Further analysis shows that Figure 3 the spatial distribution map of the Standardized Wind Resource Index in Northeast China from January to March 2014 is shown, and the time scale of the drought index is 1 month. The red covered area in the figure indicates the drought period. It can be seen from the figure that the drought level developed from light drought in the northern and northeastern parts of Heilongjiang Province in January 2014 to severe drought level covering the entire Northeast China in March, and a large area reached the level of SWRI<-1.5.

[0075] Therefore, the standardized wind energy resource index (SWRI) of this invention can fully reflect the severe impact of extreme drought events on the negative anomaly of wind energy resources, thereby enabling long-term drought monitoring of wind energy resources and contributing to disaster loss assessment for wind power projects, the State Grid, and power security.

[0076] The above embodiments are not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the technical solution of the present invention are also within the protection scope of the present invention.

Claims

1. A method for estimating a standardized wind energy resource drought index based on reanalysis data, characterized in that: The calculation steps include the following: S1. Standardized wind energy resource drought index defined by wind resources SWRI Wind resources are characterized by wind energy density; In S1, the standardized wind energy resource drought index is defined. SWRI The characterization formula is: （1） in, WR This is a time series of wind energy density variations at different time scales over a specific region, with units of W / m³. -2 Different time scales refer to days or months; This represents the average value of wind energy density changes over many years on this timescale; It is the standard deviation of wind energy density variation over many years on that time scale; or （2） in, W for WR A fitting function for a time series on a one- or multi-day time scale that follows a Weiber distribution, a log-normal distribution, or a gamma distribution; For many years on this time scale W The average value of a time series; For many years on this time scale W Standard deviation of time series; S2. Based on the defined standardized wind energy resource drought index, drought levels are classified into mild drought, moderate drought, severe drought, and extreme drought. S3. Calculate the standardized wind energy resource drought index of the target area: Using hourly wind speed time data from the reanalysis dataset, calculate the hourly wind energy density for each grid; then obtain the moving average time series of wind energy resource changes at this time scale, and calculate the standardized wind energy resource drought index at different time scales. S3 includes the following steps: S31. Obtain hourly wind speed time data from the reanalysis dataset. Calculate the hourly power density time series of wind energy resources for each grid cell. The calculation formula is as follows: （3） in, Air density, unit: kg / m³ -3 ; Wind speed time data at a height of 100 meters, in milliseconds (ms). -1 ; S32. Based on the drought monitoring application requirements of a certain time scale, calculate the moving average time series of wind energy resource changes at that time scale according to formula (3); a certain time scale includes daily or monthly time scales. S33. Calculate the standardized wind energy resource drought index at different time scales from the moving average time series of wind energy resource changes at different time scales. S4. Determine the drought level of the target area based on the drought level defined in S2 and the standardized wind energy resource drought index of the target area calculated in S3.

2. The method for estimating the standardized wind energy resource drought index based on reanalysis data according to claim 1, characterized in that: In S2, the criteria for classifying drought levels are as follows: When -1.0 < SWRI A value ≤-0.5 indicates mild drought; When -1.5 < SWRI A value ≤-1.0 indicates moderate drought; When -2.0 < SWRI A value ≤-1.5 indicates severe drought; when SWRI When the temperature is ≤-2.0, it is considered extremely dry.

3. The method for estimating the standardized wind energy resource drought index based on reanalysis data according to claim 1, characterized in that: In S33, if the changes in wind resources follow a normal distribution on a certain time scale, the standardized wind resource drought index can be calculated according to formula (1). SWRI ; If wind resource changes follow other distributions on a certain time scale, including the Weiber distribution, log-normal distribution, or gamma distribution, the standardized wind resource drought index following the corresponding distribution can be calculated using formula (2). SWRI .

Images