Wind power plant site selection method based on multi-level offshore wind energy resource assessment

Abstract

The invention relates to a wind power plant site selection method based on multi-level offshore wind energy resource assessment, which comprises the following steps of: acquiring the position of an established wind power plant and wake flow data thereof, and calculating a position index and a wake flow index of the wind power plant; obtaining water depth data and offshore distance data, and calculating a water depth index and an offshore distance index; acquiring wind speed data, and calculating a wind resource index; calculating a comprehensive ocean wind energy index of each region based on the indexes; areas with the comprehensive ocean wind energy indexes larger than a preset threshold value are screened as candidate areas; acquiring a wind resource condition of a specified height by using the wind shear index of the candidate region, and performing statistical analysis on the wind resource condition of the candidate region by using Weibull probability distribution to obtain an average wind power density; and selecting a wind power plant installation address from the candidate area based on the average wind power density. Compared with the prior art, the method has the advantages of effective and accurate wind power plant site selection and the like.

Description

Technical Field

[0001] The present invention relates to the field of wind farm site selection, and in particular to a wind farm site selection method based on multi-level offshore wind energy resource assessment. Background Art

[0002] As an important part of renewable energy (RES), offshore wind energy is favored by many coastal countries because of its advantages such as no land occupation, no pollution, and high wind energy utilization rate. However, the site selection of wind farms depends on the wind resources. If the wind energy resources cannot be effectively evaluated, the site selection of wind farms will be ineffective and the power generation efficiency will be low.

[0003] Offshore wind resource assessments are generally divided into two types: low-resolution macro-assessments (scale of thousands of kilometers), which provide overall information on wind resources in large areas; and high-resolution micro-assessments (scale of tens to hundreds of kilometers), which provide detailed wind resource information in micro-areas.

[0004] In terms of macro wind resource assessment, since the site selection of offshore wind farm projects depends not only on the local wind speed conditions (including annual average wind speed, seasonal changes in wind speed, effective wind speed time, etc.), but also on the offshore distance and water depth (representing the difficulty of maintenance, installation and construction), the existing macro wind resource assessment often takes these factors into consideration. Although the current macro wind resource assessment has taken many factors into consideration, no research has yet incorporated existing wind farms and their wake effects into the macro wind resource assessment, which has caused errors in wind resource assessment.

[0005] In microscopic wind resource assessment, the main purpose is to determine the high-resolution wind resource situation in the microscopic area and provide information for the microscopic site selection of wind farms. Currently, it is mainly divided into Weather Research and Forecasting (WRF) and SAR satellite combined with Weibull distribution methods, both of which can provide up to 1km 2 Even higher small-area wind energy information, studies have shown that SAR can capture sea surface wind speed characteristics (such as spatial distribution) more accurately than WRF, making SAR a valuable tool for obtaining microscopic high-precision wind resource information. However, SAR satellites obtain wind speeds at a height of 10m above the sea surface, which must be converted to the height of the wind turbine hub to obtain more accurate wind resource information. Traditional methods generally rely on a fixed wind shear exponent or logarithmic law to convert it to the height of the wind turbine hub, but the laws are different in different regions, so the accuracy of the conversion results is affected.

[0006] Therefore, there is currently a lack of a method for effectively evaluating wind resources to achieve wind farm site selection. Summary of the invention

[0007] The purpose of the present invention is to provide a wind farm site selection method based on multi-level offshore wind energy resource assessment, taking into account the existing wind farms and wake effects, evaluating the wind energy resources from a macro perspective, and then using SAR satellites combined with Weibull distribution to evaluate the high-resolution wind resource density in the local area, thereby achieving accurate and effective wind farm site selection, so that the wind farm can efficiently utilize wind resources.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment comprises the following steps:

[0010] Obtain the location and wake data of the completed wind farms, and calculate the wind farm location index and wake index;

[0011] Obtain water depth data and offshore distance data, and calculate water depth index and offshore distance index;

[0012] Obtain wind speed data and calculate wind resource indicators;

[0013] Calculate the comprehensive marine wind energy index for each area based on the wind farm location index, wake index, water depth index, offshore distance index and wind resource index;

[0014] Screening areas where the comprehensive ocean wind energy index is greater than a preset threshold as candidate areas;

[0015] The wind resource conditions at a specified height are obtained by using the wind shear index of the candidate area, and the wind resource conditions of the candidate area are statistically analyzed by using the Weibull probability distribution to obtain the average wind power density;

[0016] A wind farm installation site is selected from the candidate areas based on the average wind power density.

[0017] The location of the completed wind farm is obtained by extracting wind turbine coordinate information from SAR remote sensing images using the target detection algorithm YoloV8s. The wind farm location index of the area where the completed wind farm is located is 1, and the wind farm location index of other areas is 0.

[0018] The wake data is extracted from SAR wind farm remote sensing images using visual interpretation methods. The wake extension area of ​​each wind farm is outlined in each remote sensing image of the region within a preset time, and the outlined wake extension area is synthesized to extract the area with wake effect. The wake index of the area with wake effect is 1, and the wake index of the area without wake effect is 0.

[0019] The wind resource indicators include wind resource power density, effective wind speed, wind energy abundance, wind energy variation rate, monthly variation rate, and extreme wind speed, among which:

[0020] The wind resource power density represents the amount of wind energy per square meter and is calculated by the following formula:

[0021]

[0022] Among them, WPD is the wind resource power density, ρ is the regional air density, and V is the wind speed;

[0023] The effective wind speed refers to the ratio of the average number of hours of regional wind speed in the effective interval to the total number of hours, and the formula is as follows:

[0024]

[0025] Where EWSO is the effective wind speed, V min ≤V≤V max The wind speed is V in a year. min To V max hours, t is the number of hours in a year, V min 、V max Respectively represent the lower and upper limits of effective wind speed;

[0026] The wind energy abundance is the time when the wind resource power density is greater than a preset threshold, indicating the abundance of wind resources. The calculation formula is as follows:

[0027]

[0028] Among them, RLO is wind energy abundance, WPD s is the wind resource power density threshold, WPD ≥ WPD s The number of hours in a year when the wind resource power density is greater than or equal to the wind resource power density threshold;

[0029] The wind energy variation rate is determined based on the average value and standard deviation of wind resource power density, and the calculation formula is as follows:

[0030]

[0031] Among them, CV is the wind energy variation rate, σ p is the standard deviation of wind resource power density, is the average power density of wind resources;

[0032] The monthly change rate is determined based on the monthly average change degree of wind resource power density, and the calculation formula is as follows:

[0033]

[0034] Among them, MVI is the monthly change rate, P M1 With P M2are the maximum and minimum monthly average power density of wind resources, P year is the annual average value of wind resource power density;

[0035] The calculation formula of the extreme wind speed is:

[0036]

[0037] Among them, EWS is the extreme wind speed that occurs once in 50 years;

[0038]

[0039] is the average value of a set of wind speed data, and σ is the standard deviation of a set of wind speed data.

[0040] The water depth index is represented by the value of the regional water depth.

[0041] The offshore distance index is obtained by randomly generating multiple points in the area and calculating the average distance from each point to the coastline.

[0042] The wind resource power density, effective wind speed, and wind energy abundance are positive indicators, and the wind energy variability, monthly change rate, extreme wind speed, water depth index, and offshore distance index are negative indicators. All positive and negative indicators are normalized and unified within the range of 0-1. The larger the normalized indicator value, the more favorable the conditions for wind farm construction.

[0043] The calculation formula of the comprehensive ocean wind energy index is as follows:

[0044] COWEI=0.22×WPD+0.219×EWSO+0.102×RLO+0.096×CV

[0045] +0.047×MVI+EWS×0.145+WD×0.0065+DC×0.106-0.1×FW-WF

[0046] Among them, WPD is the wind resource power density, EWSO is the effective wind speed, RLO is the wind energy abundance, CV is the wind energy variability, MVI is the monthly change rate, EWS is the extreme wind speed, WD is the water depth index, DC is the offshore distance index, FW is the wake index, and WF is the wind farm location index.

[0047] The method of obtaining the wind resource situation at a specified height by using the wind shear index of the candidate area is as follows:

[0048] Using the wind shear exponent law, Z r The wind speed at altitude is converted to the wind speed at Z altitude:

[0049]

[0050] Where U represents the wind speed at Z height, U r Represents Z r Wind speed at height, wind speed unit is m / s, height unit is m, α is the wind shear coefficient.

[0051] The Weibull probability distribution is used to statistically analyze the wind resource conditions in the candidate area to obtain the average wind power density, which is specifically:

[0052] The Weibull probability distribution follows:

[0053]

[0054] Among them, V is the wind speed, k is the scale parameter, and A is the shape parameter, which are calculated by the following formulas:

[0055]

[0056]

[0057] Among them, δ is the variance of a set of wind speeds, is the average value of a set of wind speed data, г() is the Gamma function;

[0058] The average wind power density AWPD is expressed as:

[0059]

[0060] Where ρ is the regional air density.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention combines a macro-scale assessment using ERA5 reanalysis data, a GEBCO water depth data set and SAR satellite data with a micro-scale assessment based on high-resolution SAR satellite data. In the macro-scale analysis, in addition to classical factors such as wind speed, wind resource power density, water depth and offshore distance, the location of existing wind farms and the influence of wake effects are also combined to improve the effectiveness of the macro-assessment results of wind resources. After determining areas with high wind energy potential in the macro-scale analysis, SAR satellite data combined with Weibull distribution are used to further perform high-resolution assessments, thereby obtaining more detailed areas with better wind energy resources, thereby selecting wind farm sites, which can greatly improve the effectiveness of the site selection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is a flow chart of the method of the present invention;

[0064] Figure 21 is an S-1GRD image and an enlarged image thereof in an embodiment, wherein (a) is an S-1GRD image, (b) is an image of a wind turbine, and (c) is an image of a ship;

[0065] Figure 3 is an S-1OCN image in one embodiment;

[0066] Figure 4 Schematic diagram for comparing wind farm site selection results in an embodiment, wherein (a) is a schematic diagram of a wind shear index, (b) is a schematic diagram of a wind farm site selection result of a wind energy resource assessment based only on a comprehensive ocean wind energy index, and (c) is a wind farm site selection result of a wind energy resource assessment using the present invention. DETAILED DESCRIPTION

[0067] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0068] This embodiment provides a wind farm site selection method based on multi-level offshore wind energy resource assessment, such as Figure 1 As shown, the following steps are included:

[0069] S1, obtain the location of the built wind farm and its wake data, and calculate the wind farm location index and wake index.

[0070] In this embodiment, the location of the built wind farm is obtained by extracting wind turbine coordinate information from SAR remote sensing images using the target detection algorithm YoloV8s. The wind farm location index WF in the area where the built wind farm is located is 1, and the wind farm location index WF in other areas is 0.

[0071] In S-1GRD image data, offshore wind turbines usually appear as high reflectivity features, similar to man-made metal objects against the ocean background, and are therefore easy to distinguish. Figure 2 (a) shows an example of an S-1GRD image that includes offshore wind turbines and ships. Figure 2 (b) provides an enlarged view of the turbine features, Figure 2 (c) shows the surrounding ships. While both turbines and ships exhibit high reflectivity typical of metallic objects on the ocean, their unique shapes allow for the effective application of object detection algorithms to identify turbines.

[0072] In the field of target detection, the YOLO (You Only Look Once) algorithm is widely recognized for its high accuracy, low computational cost, and compatibility with various hardware platforms including personal computers. The present invention adopts the latest YOLOv8 detection algorithm, including five models: YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x. Considering that OWT appears as small objects in wide-width SAR images, the YOLOv8s model was selected because of its excellent efficiency in small object detection. The hardware environment used in the present invention includes an Intel i5-12400F processor and an NVIDIA RTX 3070 graphics card. The training of the YOLOv8s model is implemented in the PyTorch environment. Initially, each of the 27 pre-processed SAR images covering all offshore areas of a region was cropped into an image of 500×500 pixels. Among them, 142 images containing turbines and other offshore man-made objects were randomly selected for manual annotation based on the unique features of OWT in SAR images. These annotated images were then used for model training, which consisted of 159 iterations. After training, the processed SAR images were fed into the YOLOv8 detection algorithm for object detection. The Intersection over Union (IoU) threshold was set to 0.08 to avoid the problem of duplicate detection of small objects due to high IoU. A confidence level of 0.4 was chosen because a higher confidence level could result in missing many turbine detections, while a lower confidence level could result in mistaking other offshore objects for turbines.

[0073] The wake data is extracted from SAR wind farm remote sensing images using visual interpretation methods. The wake extension area of ​​each wind farm is outlined in each remote sensing image of the region within a preset time, and the outlined wake extension area is synthesized to extract the area with wake effect. The wake index FW of the area with wake effect is 1, and the wake index FW of the area without wake effect is 0.

[0074] In the S1-OCN image data, different pixels represent wind speed information at different corresponding coordinates. For example, Figure 3This is an S-1OCN image of a coastal area of ​​a certain place taken on a certain day, covering six offshore wind farms. The wake effect area of ​​the largest wind farm extends to an astonishing 90 kilometers, covering an area of ​​about 1,000 square kilometers. In addition, the wake lengths of the other two wind farms reach at least 75 kilometers and 80 kilometers respectively, with a total coverage area of ​​about 1,000 square kilometers, excluding the impact on onshore wind turbines. The wake collection method for offshore wind farms is mainly to use visual interpretation after importing S-1OCN data into ArcGIS Pro software, and outline the wake extension area of ​​each wind farm in each remote sensing image of the region over a period of time. Then use the merge function of ArcGIS Pro to synthesize the outlined wake area, and finally obtain the wake coverage information of all wind farms at different times.

[0075] S2, obtaining water depth data and offshore distance data, and calculating water depth index and offshore distance index.

[0076] Water depth (unit: m) and distance from shore are two important factors that affect the site selection and construction and maintenance costs of offshore wind turbines. Generally speaking, the deeper the sea depth and the farther the distance from shore, the higher the cost of laying submarine cables and the higher the maintenance and operation costs after construction. Therefore, these two types of data need to be considered in the early site selection of wind farms.

[0077] In this embodiment, the water depth data directly uses the GEBCO data set, and the water depth index is expressed as the value of the regional water depth. The offshore distance index is obtained by randomly generating multiple points in the region using ArcGIS Pro software and calculating the average distance from each point to the coastline. Obviously, the increase in water depth and offshore distance will lead to an increase in the cost of offshore wind farms, so the water depth index and offshore distance index are both negative indicators.

[0078] S3, obtain wind speed data and calculate wind resource indicators.

[0079] Considering that the typical life cycle of an offshore wind farm is 20 to 30 years, long-term wind speed assessment is crucial. Therefore, this example uses ERA5 reanalysis wind speed data from 1983 to 2023 to assess the wind resource index at a height of 100 meters.

[0080] The reanalysis data used in this embodiment is the 100m height wind speed data of ERA5 from 1983 to 2023, which is used to evaluate the wind resource conditions at the hub height of a typical wind turbine. The 10m height wind speed data and 100m height data of ERA5 at the same time are used to estimate the regional wind shear index, with a spatial resolution of 0.25 degrees (approximately 31KM).

[0081] Wind resource indicators include wind resource power density, effective wind speed, wind energy abundance, wind energy variability, monthly change rate, and extreme wind speed.

[0082] Wind resource power density is the most critical parameter used to evaluate wind energy resources in a region. It indicates the amount of wind energy per square meter, and the unit is W / m 2 , calculated by the following formula:

[0083]

[0084] Where WPD is the wind resource power density, ρ is the regional air density, which is 1.225 kg / m 3 , V is the wind speed at a height of 100m, in m / s. The higher the WPD, the better the wind resource, so WPD is a positive indicator.

[0085] The effective wind speed indicates the ratio of the average number of hours in which the regional wind speed is in the effective interval to the total number of hours. The formula is as follows:

[0086]

[0087] Where EWSO is the effective wind speed, V min ≤V≤V max The wind speed is V in a year. min To V max hours, t is the number of hours in a year, V min 、V max Respectively represent the lower limit and upper limit of the effective wind speed; in this embodiment, V min Take 3m / s, V max Take 25m / s. The higher the EWSO, the longer the effective power generation time, so EWSO is a positive indicator.

[0088] Wind energy abundance refers to the time when the wind resource power density is greater than the preset threshold, indicating the abundance of wind resources. The calculation formula is as follows:

[0089]

[0090] Among them, RLO is wind energy abundance, WPD s is the wind resource power density threshold, which is 200W / m 2 , WPD ≥ WPD s It is the number of hours in a year when the wind resource power density is greater than or equal to the wind resource power density threshold; the larger the RLO, the better the regional wind energy conditions, so it is a positive indicator.

[0091] The wind energy variability is determined based on the average and standard deviation of wind resource power density, which characterizes the dispersion of WPD in historical data. Generally speaking, CV reflects the stability of wind resources. The larger the value, the greater the difference in wind power density at different time points in this region. This will lead to a negative impact of wind farms in this region on the stability of the regional power grid and also affect the life of wind turbines. Therefore, it is a negative indicator. Its calculation formula is as follows:

[0092]

[0093] Among them, CV is the wind energy variation rate, σ p is the standard deviation of wind resource power density, is the average power density of wind resources.

[0094] The monthly change rate is determined based on the monthly average change in wind resource power density, and is used to evaluate the degree of variation in regional wind energy within a year. The smaller the value, the smaller the change in regional wind resources within a year, so it is a negative indicator. The calculation formula is as follows:

[0095]

[0096] Among them, MVI is the monthly change rate, P m1 With P m2 are the maximum and minimum monthly average power density of wind resources, P year is the annual average value of wind resource power density.

[0097] Due to the particularity of wind turbine structure, extreme wind speed can cause wind turbine tower collapse and blades to be damaged due to excessive force, which seriously endangers the integrity of wind turbine structure. Therefore, extreme wind speed needs to be included as one of the factors for wind farm site selection.

[0098] Gumbel distribution has been used as extreme value distribution in many neighborhoods (including wind energy). The greater the extreme wind speed, the greater the probability of wind turbine damage, so the extreme wind speed EWS is a negative indicator. Therefore, this embodiment uses Gumbel distribution combined with 41 years of wind speed data to calculate the extreme wind speed of the region once in 50 years. The cumulative probability distribution function F(V) of Gumbel distribution is as follows:

[0099] F(v)=exp[-exp{-a(vb)}]

[0100] Among them, v is the extreme wind speed, a is the scale parameter, b is the location parameter,

[0101]

[0102] is the average value of a set of wind speed data, and σ is the standard deviation of a set of wind speed data.

[0103] The calculation formula for the extreme wind speed EWS once in 50 years is:

[0104]

[0105] S4, calculates the comprehensive marine wind energy index for each area based on the wind farm location index, wake index, water depth index, offshore distance index and wind resource index.

[0106] This embodiment divides the wind resources macroscopically by combining the 10 factors of WPD, EWSO, RLO, CV, MVI, EWS, WD, DC, WF and FW calculated above. Table 1 briefly summarizes these 10 indicators. Since the units of these parameters are different, each parameter must be normalized before the final comprehensive calculation. A normalization formula is applied to both positive and negative indicators, as shown below.

[0107] 1). Positive indicators:

[0108]

[0109] 2). Negative indicators

[0110]

[0111] Among them, x i is the value of the ith indicator, y i is the normalized index value; after normalization, except for FW and WF which are set to 1 (indicating the existence of wake effect and wind farm), the higher the positive and negative index values ​​of the other 8 categories are, the more favorable the conditions for wind farm construction are.

[0112] The above indicators are combined by weighted method to obtain the comprehensive ocean wind energy index:

[0113] COWEI=0.22×WPD+0.219×EWSO+0.102×RLO+0.096×CV

[0114] +0.047×MVI+EWS×0.145+WD×0.0065+DC×0.106-0.1×FW-WF

[0115] Among them, WPD is the wind resource power density, EWSO is the effective wind speed, RLO is the wind energy abundance, CV is the wind energy variability, MVI is the monthly change rate, EWS is the extreme wind speed, WD is the water depth index, DC is the offshore distance index, FW is the wake index, and WF is the wind farm location index.

[0116] After comprehensively calculating the above indicators, a higher COWEI value means that the area is more suitable for installing wind turbines, while an area with a value less than 0 indicates that wind turbines have already been installed and is no longer suitable for building a new wind farm.

[0117] Table 1 Evaluation criteria affecting offshore wind farm construction

[0118]

[0119] S5, screening areas where the comprehensive ocean wind energy index is greater than a preset threshold as candidate areas.

[0120] S6, using the wind shear index of the candidate area to obtain the wind resource conditions at a specified height, and using Weibull probability distribution to perform statistical analysis on the wind resource conditions of the candidate area to obtain an average wind power density.

[0121] Since the wind speed information acquired by the SAR satellite is the wind speed at a height of 10m above the sea surface, it needs to be converted to the traditional wind turbine hub height of 100m. Therefore, this embodiment mainly uses the ERA5 100m height and 10m height wind speed to derive the regional wind shear index.

[0122] Both the wind shear exponent and the logarithmic law can be used to extrapolate wind speeds at different heights. However, the logarithmic law requires surface roughness data, which is difficult to obtain. Therefore, the present invention uses the wind shear exponent law to extrapolate the wind speed at a height of 10m to the wind speed at 100m. The wind shear exponent law used is:

[0123]

[0124] Where U represents the wind speed at height Z, U r Represents Z r The wind speed at height, wind speed unit is m / s, height unit is m, α is the wind shear coefficient. In traditional research, α is set to a fixed constant of 0.11, but since α in each region is not fixed, directly using a fixed value may cause errors in wind resource assessment. Therefore, the present invention uses the wind speed at 100m height and the wind speed at 10m height provided by ERA5 data to infer, and α can be expressed as:

[0125]

[0126] Among them U 100 and U 10 The distribution represents the wind speed at 100m and 10m of ERA5, Z 100 and Z 10 Representing 100m and 10m respectively. In this way, the regional wind shear index can be calculated for the extrapolation of the 10m height wind speed data of S-1OCN to the 100m height.

[0127] Since S-1 collects wind speed data from the same ocean location at specific time intervals, it is necessary to use Weibull distribution to perform statistical analysis on S-1OCN wind speed data within a given time period to obtain high-precision WPD values ​​for the area. Weibull distribution is often used for statistical analysis in wind energy research to assess regional wind energy resources. As remote sensing satellites continuously collect wind speed data from the same area, offshore wind speed data collected by S-1OCN can also be fitted with the regional wind speed distribution using Weibull distribution. Weibull probability distribution obeys:

[0128]

[0129] Among them, V is the wind speed, k is the scale parameter, and A is the shape parameter, which are calculated by the following formulas:

[0130]

[0131]

[0132] Among them, δ is the variance of a set of wind speeds, is the average value of a set of wind speed data, г() is the Gamma function;

[0133] The average wind power density AWPD is expressed as:

[0134]

[0135] Where ρ is the regional air density.

[0136] S7: Selecting a wind farm installation site from the candidate areas based on the average wind power density.

[0137] This embodiment uses a specific case to illustrate the effectiveness of the wind farm site selection method of the present invention.

[0138] When selecting a suitable offshore wind turbine installation area in a huge area (thousands of kilometers) and requiring refined evaluation requirements, if only macroscopic evaluation methods are used, although the wind energy situation in this vast area is obtained, it is limited by the resolution of macroscopic evaluation data (31km 2 ), but often cannot directly use this macro data in actual construction, because currently a wind farm generally occupies an area of ​​tens of kilometers, and the macro data under this area often only has a few data points, and the amount of data cannot meet the actual construction needs. Therefore, the introduction of micro wind resource assessment method can solve this problem well. The assessment method based on S-1 satellite data used in the present invention provides 1km 2 The data resolution can provide thousands of data points within a wind farm area of ​​dozens of kilometers, meeting the actual construction needs of the wind farm.

[0139] First, the macro-comprehensive evaluation method introduced in steps S1-S4 is used to evaluate an entire region. After considering the comprehensive data of wind resource power density, wind energy abundance, effective wind speed, wind energy variability, wind energy monthly change rate, extreme wind speed, offshore distance, water depth, built wind farms, and wind farm wake, the overall distribution of COWEI is obtained.

[0140] In small areas with excellent COWEI values, a microscopic wind resource assessment method based on S-1 satellite data given in steps S5-S6 is used for microscopic assessment to obtain the wind resource conditions at an altitude of 100m in the small area, so as to select a suitable wind turbine installation site based on the high-resolution wind resource conditions in the small area.

[0141] Specifically, Figure 4 The results of using ERA5 data in a small area (such as Figure 4 (b)) and S-1 satellite data (such as Figure 4 (c)) Examples of comparative wind resource assessments for small areas. Figure 4 (a) is the derived wind shear index. It can be seen that, due to the limitation of low data resolution, the ERA5 data (with the same resolution as COWEI) is very roughly distributed in this sea area. The wind resources from offshore to offshore are simply an upward relationship, and there is no obvious distribution in details. On the other hand, the wind resource distribution provided by the S-1 satellite data shows that it provides a good wind resource situation in detail. Figure 4 (b) If a wind farm is constructed based on the wind resource distribution, it is very likely that the wind farm will be built in an area with low wind resources, resulting in lower power generation than expected. Figure 4 The wind resource distribution in (b) sets the wind farm near the black box in the figure. Figure 4 (b) It seems that this area has good wind resource density and can provide effective wind energy for wind farms, but in fact Figure 4 (c) The high-resolution wind resources provided are not that high, which results in the actual power generation of the wind farm being smaller than estimated. The wind energy resource assessment method combining macro and micro in the present invention can more accurately assess the wind resources, thereby more accurately selecting the site for the wind farm.

[0142] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment, characterized in that: The following steps are involved: Obtain the location and wake data of the completed wind farms, and calculate the wind farm location index and wake index; Obtain water depth data and offshore distance data, and calculate water depth index and offshore distance index; Obtain wind speed data and calculate wind resource indicators; Calculate the comprehensive marine wind energy index for each area based on the wind farm location index, wake index, water depth index, offshore distance index and wind resource index; Screening areas where the comprehensive ocean wind energy index is greater than a preset threshold as candidate areas; The wind resource conditions at a specified height are obtained by using the wind shear index of the candidate area, and the wind resource conditions of the candidate area are statistically analyzed by using the Weibull probability distribution to obtain the average wind power density; A wind farm installation site is selected from the candidate areas based on the average wind power density.

2. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The location of the completed wind farm is obtained by extracting wind turbine coordinate information from SAR remote sensing images using the target detection algorithm YoloV8s. The wind farm location index of the area where the completed wind farm is located is 1, and the wind farm location index of other areas is 0.

3. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The wake data is extracted from SAR wind farm remote sensing images using visual interpretation methods. The wake extension area of ​​each wind farm is outlined in each remote sensing image of the region within a preset time, and the outlined wake extension area is synthesized to extract the area with wake effect. The wake index of the area with wake effect is 1, and the wake index of the area without wake effect is 0.

4. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The wind resource indicators include wind resource power density, effective wind speed, wind energy abundance, wind energy variation rate, monthly variation rate, and extreme wind speed, among which: The wind resource power density represents the amount of wind energy per square meter and is calculated by the following formula: Among them, WPD is the wind resource power density, ρ is the regional air density, and V is the wind speed; The effective wind speed refers to the ratio of the average number of hours of regional wind speed in the effective interval to the total number of hours, and the formula is as follows: Where EWSO is the effective wind speed, V min ≤V≤V max The wind speed is V in a year. min To V max hours, t is the number of hours in a year, V min 、V max Respectively represent the lower and upper limits of effective wind speed; The wind energy abundance is the time when the wind resource power density is greater than a preset threshold, indicating the abundance of wind resources. The calculation formula is as follows: Among them, RLO is wind energy abundance, WPD s is the wind resource power density threshold, WPD ≥ WPD s The number of hours in a year when the wind resource power density is greater than or equal to the wind resource power density threshold; The wind energy variation rate is determined based on the average value and standard deviation of wind resource power density, and the calculation formula is as follows: Among them, CV is the wind energy variation rate, σ p is the standard deviation of wind resource power density, is the average power density of wind resources; The monthly change rate is determined based on the monthly average change degree of wind resource power density, and the calculation formula is as follows: Among them, MVI is the monthly change rate, P M1 With P M2 are the maximum and minimum monthly average power density of wind resources, P year is the annual average value of wind resource power density; The calculation formula of the extreme wind speed is: Among them, EWS is the extreme wind speed that occurs once in 50 years; is the average value of a set of wind speed data, and σ is the standard deviation of a set of wind speed data.

5. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The water depth index is represented by the value of the regional water depth.

6. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The offshore distance index is obtained by randomly generating multiple points in the area and calculating the average distance from each point to the coastline.

7. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 4, characterized in that: The wind resource power density, effective wind speed, and wind energy abundance are positive indicators, and the wind energy variability, monthly change rate, extreme wind speed, water depth index, and offshore distance index are negative indicators. All positive and negative indicators are normalized and unified within the range of 0-1. The larger the normalized indicator value, the more favorable the conditions for wind farm construction.

8. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 4, characterized in that: The calculation formula of the comprehensive ocean wind energy index is as follows: COWEI=0.22×WPD+0.219×EWSO+0.102×RLO+0.096×CV+0.047×MVI+EWS×0.145+WD×0.0065+DC×0.106-0.1×FW-WF Among them, WPD is the wind resource power density, EWSO is the effective wind speed, RLO is the wind energy abundance, CV is the wind energy variability, MVI is the monthly change rate, EWS is the extreme wind speed, WD is the water depth index, DC is the offshore distance index, FW is the wake index, and WF is the wind farm location index.

9. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The method of obtaining the wind resource situation at a specified height by using the wind shear index of the candidate area is as follows: Using the wind shear exponent law, Z r The wind speed at altitude is converted to the wind speed at Z altitude: Where U represents the wind speed at height Z, U r Represents Z r Wind speed at height, wind speed unit is m / s, height unit is m, α is the wind shear coefficient.

10. A method for selecting a wind farm site based on multi-level offshore wind energy resource assessment according to claim 1, characterized in that: The Weibull probability distribution is used to statistically analyze the wind resource conditions in the candidate area to obtain the average wind power density, which is specifically: The Weibull probability distribution follows: Among them, V is the wind speed, k is the scale parameter, and A is the shape parameter, which are calculated by the following formulas: Among them, δ is the variance of a set of wind speeds, is the average value of a set of wind speed data, г() is the Gamma function; The average wind power density AWPD is expressed as: Where ρ is the regional air density.

Images