A laser radar wind speed correction method, device and equipment
By collecting and calculating wind speed data from lidar, and combining it with flow field and terrain data, a wind speed correction model was used to solve the error problem of lidar in measuring wind speed in complex terrain, thus improving the measurement accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING GOLDWIND SCI & CREATION WINDPOWER EQUIP CO LTD
- Filing Date
- 2021-12-22
- Publication Date
- 2026-07-10
AI Technical Summary
When lidar measures wind speed in complex terrain conditions, errors occur, leading to a decrease in measurement accuracy.
Wind speed data at observation points is collected using lidar. Combined with flow field calculations and terrain data, a wind speed correction model is used to correct the data and establish the correspondence between the wind speed at the observation points and the standard wind speed of the wind measuring tower.
The accuracy of lidar in wind speed measurement under complex terrain conditions has been improved, and the corrected wind speed data is closer to the standard wind speed data of the wind measuring tower.
Smart Images
Figure CN116362144B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, specifically to a lidar wind speed correction method, apparatus, and equipment. Background Technology
[0002] LiDAR, such as ground-based lidar, is widely used in the wind energy industry for measuring wind field information. Compared with traditional wind measurement towers, lidar is easy to install, has low measurement costs, and has good performance both on land and at sea.
[0003] While lidar offers high accuracy in flat terrain measurements, its inherent assumptions limit its accuracy in complex terrain and non-uniform flow fields, leading to inherent errors in wind speed measurements. Therefore, improving the accuracy of lidar wind speed measurements through correction is a pressing technical challenge. Summary of the Invention
[0004] In view of this, embodiments of this application provide a lidar wind speed correction method, apparatus, and device to improve the accuracy of lidar wind speed measurement.
[0005] To address the above problems, the technical solutions provided in this application are as follows:
[0006] This application provides a lidar wind speed correction method, the method comprising:
[0007] Wind speed data at observation points are collected using lidar.
[0008] The flow field is calculated based on the location of the observation point to obtain the flow field data of the observation point;
[0009] Acquire the terrain data and wind measurement height corresponding to the observation point;
[0010] The wind speed data, flow field data, terrain data, and wind measurement height of the observation point are input into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0011] In one possible implementation, the step of inputting the wind speed data, flow field data, terrain data corresponding to the observation point, and wind measurement height corresponding to the observation point into the wind speed correction model to obtain the corrected wind speed data includes:
[0012] The terrain data corresponding to the observation point is input into the feature extraction model to obtain the terrain vector;
[0013] The wind speed data, flow field data, terrain vector, and wind measurement height corresponding to the observation point are concatenated to obtain the input feature vector.
[0014] The input feature vector is input into the wind speed correction model to obtain the corrected wind speed data.
[0015] In one possible implementation, the acquisition of wind speed data at the observation point via lidar includes:
[0016] The horizontal wind speed at the observation point and the radial wind speed at the reference point are collected by a lidar as wind speed data for the observation point, where the reference point is on the path of the beam emitted by the lidar.
[0017] In one possible implementation, the step of calculating the flow field based on the location of the observation point to obtain the flow field data of the observation point includes:
[0018] The flow field is calculated based on the location of the observation point and the location of the reference point to obtain the simulated horizontal wind speed at the observation point and the simulated horizontal wind speed at the reference point as the flow field data of the observation point.
[0019] In one possible implementation, the step of inputting the wind speed data, flow field data, and corresponding terrain data of the observation point into a wind speed correction model to obtain corrected wind speed data includes:
[0020] The terrain data corresponding to the observation point is input into the feature extraction model to obtain the terrain vector;
[0021] The horizontal wind speed at the observation point, the radial wind speed at the reference point, the fluid simulation horizontal wind speed at the observation point, the fluid simulation horizontal wind speed at the reference point, the terrain vector, and the wind measurement height corresponding to the observation point are concatenated to obtain the input feature vector.
[0022] The input feature vector is input into the wind speed correction model to obtain the corrected wind speed data.
[0023] In one possible implementation, the modified wind speed data is used for wind turbine power curve testing, wind resource assessment, and / or wind power project assessment.
[0024] In one possible implementation, the method further includes:
[0025] Wind speed data at the observation points to be trained are collected using lidar.
[0026] Obtain standard wind speed data of the observation points to be trained, as measured by the wind measurement tower;
[0027] The flow field is calculated based on the location of the observation point to be trained, and the flow field data of the observation point to be trained is obtained.
[0028] Acquire the terrain data and wind height corresponding to the observation point to be trained;
[0029] The wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained are used as training data, and the standard wind speed data of the observation point to be trained are used as the label of the training data to train the wind speed correction model.
[0030] In one possible implementation, the step of training a wind speed correction model by using the wind speed data, flow field data, terrain data, and wind height of the observation point to be trained as training data, and using the standard wind speed data of the observation point to be trained as the label of the training data, includes:
[0031] The terrain data corresponding to the observation points to be trained is input into the feature extraction model to obtain the terrain vector to be trained;
[0032] The wind speed data, flow field data, terrain vector, and wind measurement height corresponding to the observation point to be trained are concatenated to obtain the feature vector of the model to be trained.
[0033] The feature vector of the model to be trained is used as training data, and the standard wind speed data of the observation point to be trained is used as the label of the training data to train the wind speed correction model.
[0034] In one possible implementation, the acquisition of wind speed data of the observation points to be trained via lidar includes:
[0035] The horizontal wind speed at the observation point to be trained and the radial wind speed at the reference point to be trained are collected by lidar as wind speed data of the observation point to be trained. The reference point to be trained is on the path of the beam emitted by the lidar.
[0036] In one possible implementation, the step of calculating the flow field based on the location of the observation point to be trained, to obtain the flow field data of the observation point to be trained, includes:
[0037] The flow field is calculated based on the location of the observation point to be trained and the location of the reference point to be trained, and the simulated horizontal wind speed of the fluid at the observation point to be trained and the simulated horizontal wind speed of the fluid at the reference point to be trained are obtained as the flow field data of the observation point to be trained.
[0038] In one possible implementation, the wind speed data, flow field data, terrain data, and wind height of the observation point to be trained are used as training data, and the standard wind speed data of the observation point to be trained are used as the label of the training data to train a wind speed correction model, including:
[0039] The terrain data corresponding to the observation points to be trained is input into the feature extraction model to obtain the terrain vector to be trained;
[0040] The horizontal wind speed of the observation point to be trained, the radial wind speed of the reference point to be trained, the fluid simulation horizontal wind speed of the observation point to be trained, the fluid simulation horizontal wind speed of the reference point to be trained, the terrain vector to be trained, and the wind measurement height corresponding to the observation point to be trained are spliced together to obtain the feature vector of the model to be trained.
[0041] The feature vector of the model to be trained is used as training data, and the standard wind speed data of the observation point to be trained is used as the label of the training data to train the wind speed correction model.
[0042] This application embodiment also provides a lidar wind speed correction device, the device comprising:
[0043] The first acquisition unit is used to acquire wind speed data at the observation point via lidar.
[0044] The first calculation unit is used to perform flow field calculations based on the location of the observation point to obtain the flow field data of the observation point.
[0045] The first acquisition unit is used to acquire the terrain data corresponding to the observation point and the wind measurement height corresponding to the observation point;
[0046] The second acquisition unit is used to input the wind speed data, flow field data, terrain data, and wind measurement height of the observation point into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0047] This application also provides an electronic device, the electronic device comprising:
[0048] processor;
[0049] A memory configured to store processor-executable instructions;
[0050] The processor is configured to execute the instructions to implement the steps of the lidar wind speed correction method as described above.
[0051] This application embodiment also provides a storage medium, which, when the instructions in the storage medium are executed by the processor of a lidar wind speed correction device or an electronic device, enables the lidar wind speed correction device or the electronic device to implement the steps of the lidar wind speed correction method as described above.
[0052] This application also provides a computer program product, which includes a computer program stored in a readable storage medium. At least one processor of the device reads and executes the computer program from the storage medium, causing the device to perform the steps of the lidar wind speed correction method as described above.
[0053] Therefore, the embodiments of this application have the following beneficial effects:
[0054] This application provides a method, apparatus, and device for correcting wind speed using a lidar system. Wind speed data at observation points is collected using a lidar system. Flow field calculations are performed based on the location of the observation points to obtain flow field data. Topographic data and the corresponding wind measurement height for each observation point are acquired. Then, the wind speed data, flow field data, topographic data, and wind measurement height are input into a wind speed correction model to obtain corrected wind speed data. The corrected wind speed data represents a more standard wind speed at the observation point location. The wind speed correction model characterizes the correspondence between a target data set and the standard wind speed data collected by the wind tower at the observation points. The target data set includes the wind speed data, flow field data, topographic data, and wind measurement height at the observation points. Thus, by considering the terrain features of the lidar location and based on the wind speed data collected by the lidar and the calculated flow field data, the wind speed measured by the lidar can be corrected using the wind speed correction model, improving the accuracy of lidar wind speed measurements. Attached Figure Description
[0055] Figure 1 A scatter plot comparing wind speed measured by lidar and wind speed measured by a wind tower is provided in an embodiment of this application.
[0056] Figure 2A schematic diagram illustrating an exemplary application scenario provided in this application embodiment;
[0057] Figure 3 A flowchart of a lidar wind speed correction method provided in this application embodiment;
[0058] Figure 4 A schematic diagram of a lidar method for measuring wind speed, provided as an embodiment of this application;
[0059] Figure 5 A schematic diagram of a lidar wind speed correction method provided in an embodiment of this application;
[0060] Figure 6 A flowchart illustrating a wind speed correction model training method provided in this application embodiment;
[0061] Figure 7 A schematic diagram of a lidar wind speed correction device provided in an embodiment of this application;
[0062] Figure 8 A schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0063] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the embodiments of this application will be further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0064] To facilitate understanding and explanation of the technical solutions provided in the embodiments of this application, the background technology of the embodiments of this application will be described below.
[0065] LiDAR, such as ground-based lidar, is widely used in the wind energy industry for measuring wind field information. Compared to traditional wind measurement towers, lidar is easier to install, has lower measurement costs, and performs well both onshore and offshore. Furthermore, lidar can not only acquire wind resource data but also provide model input and validation for numerical simulations of wind fields.
[0066] Because lidar, especially ground-based lidar, measures wind speed based on the assumption of a uniform flow field. That is, ground-based lidar emits a radar beam upwards, assuming that the flow field at each laser beam at the same height is uniform, and the wind field conditions are consistent at all points. Therefore, ground-based lidar has high accuracy in measuring wind speed on flat terrain. However, due to the limitations of lidar measurement assumptions, when lidar is set up in complex terrain, the flow field is non-uniform, and the uniform flow field assumption does not hold. This will cause a certain deviation between the wind speed measured by lidar and the wind speed measured by the anemometer. Therefore, the embodiments of this application can correct the wind speed measured by lidar, especially ground-based lidar, when set up in complex terrain.
[0067] See Figure 1 , Figure 1 This is a scatter plot comparing wind speed measured by lidar and wind speed measured by a wind tower, provided as an embodiment of this application. Figure 1 The horizontal axis represents the horizontal wind speed measured by the wind tower, and the vertical axis represents the horizontal wind speed measured by the lidar. The relationship is y = 0.957x + 0.054, where y is the horizontal wind speed measured by the lidar and x is the horizontal wind speed measured by the wind tower. 2 For the degree of dispersion, R 2 A larger value indicates a better linearity of the curve and a better clustering of scatter points. Figure 1 In the diagram, the horizontal wind speed measured by the wind tower and the corresponding horizontal wind speed measured by the lidar are: Figure 1 The scatter plots corresponding to the original data. The difference between the horizontal wind speed measured by the wind tower and the horizontal wind speed measured by the lidar is... Figure 1 The scatter plots correspond to the original data with medium deviation. It is evident that there is a certain error between the horizontal wind speed measured by the wind tower and the horizontal wind speed measured by the lidar. Therefore, corrections are needed for the radar wind speed to improve the accuracy of lidar measurements in complex terrain.
[0068] Therefore, how to correct for lidar wind speed measurements and improve the accuracy of lidar measurements is an urgent technical problem to be solved.
[0069] Based on this, embodiments of this application provide a lidar wind speed correction method, apparatus, and device. To facilitate understanding of the lidar wind speed correction method provided in this application, the following is a detailed explanation. Figure 2 Exemplary application scenarios are described. Figure 2 This is a schematic diagram illustrating an exemplary application scenario provided by an embodiment of this application. The method can be implemented by a terminal device.
[0070] In practical applications, some data needs to be prepared in advance. For example, wind speed data at observation points can be collected using lidar. The wind speed data collected by lidar at the observation points may deviate from the wind speed data measured by the anemometer tower. In this embodiment, the wind speed data measured by the anemometer tower is considered standard wind speed data. Therefore, the wind speed data collected by lidar at the observation points needs to be corrected to be as close as possible to the wind speed data measured by the anemometer tower.
[0071] Next, flow field calculations are performed based on the location of the observation point to obtain the flow field data for that point. This flow field data serves as the input data for the wind speed correction model. Furthermore, the terrain data and wind measurement height corresponding to the observation point are acquired. These data also serve as input data for the wind speed correction model.
[0072] After acquiring the above data, the wind speed data, flow field data, corresponding terrain data, and corresponding wind measurement height at the observation points are input into the wind speed correction model to obtain corrected wind speed data. The corrected wind speed data then represents more standard wind speed data. The wind speed correction model characterizes the correspondence between the target dataset and the standard wind speed data collected by the wind tower at the observation points. The target dataset includes the wind speed data, flow field data, corresponding terrain data, and corresponding wind measurement height at the observation points. Those skilled in the art will understand that… Figure 2 The schematic diagram shown is merely one example in which embodiments of this application can be implemented. The scope of application of the embodiments of this application is not limited by any aspect of this framework.
[0073] Based on the above description, the lidar wind speed correction method provided in this application will be described in detail below with reference to the accompanying drawings.
[0074] See Figure 3 The figure is a flowchart of a lidar wind speed correction method provided in an embodiment of this application, which can be executed by a terminal device. Figure 3 As shown, the method may include S301-S304:
[0075] S301: Collects wind speed data at observation points using lidar.
[0076] A lidar system is a radar system that uses emitted laser beams to detect the position, velocity, and other characteristics of a target. The working principle of lidar is to emit a detection signal (i.e., a laser beam) towards the target, and then compare and process the received signal reflected back from the target (i.e., the target echo) with the emitted detection signal to obtain relevant information about the target. There are various types of lidar, such as ground-based lidar, which is a type of lidar installed on the ground and emits a radar beam upwards.
[0077] LiDAR is deployed on the ground to collect wind speed data at different locations, such as flat or complex terrain. In this embodiment, the LiDAR is deployed on complex terrain.
[0078] Wind speed data at observation points is collected using lidar. In one possible implementation, this application provides a specific method for collecting wind speed data at observation points using lidar, including:
[0079] The horizontal wind speed at the observation point and the radial wind speed at the reference point are collected by lidar as wind speed data for the observation point, with the reference point located on the path of the lidar beam.
[0080] For example, see Figure 4 , Figure 4 This is a schematic diagram illustrating a lidar method for measuring wind speed, provided as an embodiment of this application. The lidar is placed in complex terrain for observation. Figure 4 Taking the 2-beam lidar as an example ( Figure 1 The lidar emits beams 1 and 2, with reference points A and B, and an observation point C. Reference points A and B are located on the paths of beam 1 and beam 2, respectively. In one or more embodiments, the observation point and the reference point are at the same height.
[0081] A lidar system acquires the radial wind speed at two reference points, A and B, by emitting a laser beam. The radial wind speed is the projection of the actual wind speed onto the direction of the laser beam. The radial wind speeds at reference points A and B are obtained by the lidar. Based on this, and according to the uniformity assumption of lidar wind speed synthesis, the horizontal wind speed at observation point C can be synthesized from the radial wind speeds at reference points A and B. The synthesized horizontal wind speed at observation point C is the same as the horizontal wind speed at observation point C acquired by the lidar. It can be understood that the radial wind speeds at reference points A and B, and the horizontal wind speed at observation point C, are all acquired by the lidar. That is, the horizontal wind speed at the observation point acquired by the lidar, along with the radial wind speeds at the reference points, are used as the wind speed data for the observation point.
[0082] In one or more embodiments, the lidar beam may also be 4 beams or 8 beams, etc.
[0083] S302: Calculate the flow field based on the location of the observation point to obtain the flow field data of the observation point.
[0084] Computational Fluid Dynamics (CFD) is used to solve the governing equations of fluid dynamics using computers and numerical methods, simulating and analyzing fluid dynamics problems. Flow field data is obtained through simulation, such as simulated wind speed data. In one or more embodiments, flow field data for the observation point can be obtained by combining the location of the observation point with CFD calculations.
[0085] In one possible implementation, this application provides a specific method for calculating the flow field based on the location of an observation point to obtain the flow field data of that observation point, including:
[0086] The flow field is calculated based on the location of the observation point and the location of the reference point. For example, the flow field is calculated based on the height of the observation point and the height of the reference point to obtain the simulated horizontal wind speed of the fluid at the observation point and the simulated horizontal wind speed of the fluid at the reference point as the flow field data of the observation point.
[0087] For example, by calculating using CFD software to obtain Figure 4 The horizontal wind speeds at reference points A, B, and C are used as the flow field data for the observation points.
[0088] S303: Obtain the terrain data and wind measurement height corresponding to the observation point.
[0089] In this embodiment, the lidar is positioned in complex terrain. The terrain data corresponding to the complex terrain affects the accuracy of the lidar's wind speed measurement. Therefore, it is necessary to acquire the terrain data corresponding to the observation point. For example, the terrain data includes ground slope, terrain complexity, etc. The terrain data corresponding to the observation point can be preset according to the actual situation. In addition, it is also necessary to acquire the wind measurement height corresponding to the observation point, for example, the wind measurement height corresponding to the observation point is... Figure 4 The vertical height of observation point C.
[0090] S304: Input the wind speed data, flow field data, terrain data, and wind measurement height of the observation point into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0091] The wind speed correction model is a pre-trained model used to characterize the correspondence between the target dataset and the standard wind speed data of the observation points collected by the wind tower. The target dataset includes the wind speed data of the observation points, the flow field data of the observation points, the corresponding topographic data of the observation points, and the corresponding wind measurement height of the observation points. The purpose of the wind speed correction model is to correct the wind speed dataset collected by lidar to the standard wind speed data of the observation points collected by the wind tower.
[0092] Based on this, after acquiring the wind speed data, flow field data, corresponding terrain data, and corresponding wind measurement height at the observation points, these data are input into the wind speed correction model to obtain the corrected wind speed data output by the model. The corrected wind speed data output by the model can be understood as closely approximating the standard wind speed data collected by the corresponding wind tower at the observation point, thus obtaining relatively standard wind speed data. In one or more embodiments, the wind speed correction model is a machine learning algorithm model, such as a convolutional neural network prediction model, an XGB prediction model, or a random forest prediction model.
[0093] In one or more embodiments, the corrected wind speed data can be used for applications such as wind turbine power curve testing, wind resource assessment, and / or wind power project assessment.
[0094] Wind resource assessment refers to the preliminary assessment of wind resources during wind farm site selection. In practice, corrected wind speed data can be input into a computer's corresponding algorithm model to obtain the wind resource assessment results. Wind power project assessment can evaluate wind power indicators in wind power projects. Similarly, in practice, corrected wind speed data can be input into a computer's corresponding algorithm model to obtain the wind power project assessment results.
[0095] In one possible implementation, this application provides a specific implementation method for inputting wind speed data, flow field data, terrain data corresponding to the observation point, and wind measurement height corresponding to the observation point into a wind speed correction model to obtain corrected wind speed data. Please refer to A1-A3 below.
[0096] Based on the content of S301-S304, this application provides a lidar wind speed correction method, apparatus, and device. Wind speed data at observation points is collected using lidar. Flow field calculations are performed based on the location of the observation points to obtain flow field data. Topographic data and wind measurement height corresponding to the observation points are acquired. Then, the wind speed data, flow field data, topographic data, and wind measurement height are input into a wind speed correction model to obtain corrected wind speed data. The corrected wind speed data represents a more standard wind speed at the observation point location. The wind speed correction model characterizes the correspondence between the target data set and the standard wind speed data collected by the wind tower at the observation points. The target data set includes the wind speed data, flow field data, topographic data, and wind measurement height at the observation points. Therefore, considering the terrain features where the lidar is located and the wind measurement height corresponding to the observation point, and based on the wind speed data collected by the lidar and the calculated flow field data, the wind speed measured by the lidar can be corrected through the wind speed correction model, which can improve the accuracy of lidar wind speed measurement.
[0097] In one possible implementation, this application provides a specific implementation method for inputting the wind speed data, flow field data, terrain data corresponding to the observation point, and wind measurement height corresponding to the observation point into the wind speed correction model in step S304 to obtain the corrected wind speed data, including:
[0098] A1: Input the terrain data corresponding to the observation point into the feature extraction model to obtain the terrain vector.
[0099] Because the acquired terrain data may have different data dimensions than the other input wind speed correction model data, it is necessary to input the terrain data corresponding to the observation point into the feature extraction model to obtain a terrain vector. The obtained terrain vector can be matched with the other input wind speed correction model data, such as the wind speed data and flow field data of the observation point.
[0100] In one or more embodiments, the feature extraction model is a neural network model.
[0101] A2: The wind speed data, flow field data, terrain vector, and wind measurement height at the observation point are concatenated to obtain the input feature vector.
[0102] Then, the wind speed data, flow field data, terrain vector, and wind measurement height corresponding to the observation point are spliced together to obtain the input feature vector that can be input into the wind speed correction model.
[0103] A3: Input the input feature vector into the wind speed correction model to obtain the corrected wind speed data.
[0104] By inputting the feature vector into the wind speed correction model, the corrected wind speed data can be obtained. It can be understood that the corrected wind speed data can be used as the standard wind speed data for the observation point.
[0105] See Figure 5 , Figure 5 This is a schematic diagram of a lidar wind speed correction method provided in an embodiment of this application. Figure 5 As shown, based on A1-A3, this application embodiment provides another specific implementation method in S304 for inputting the wind speed data, flow field data, terrain data corresponding to the observation point, and wind measurement height corresponding to the observation point into the wind speed correction model to obtain the corrected wind speed data, including:
[0106] B1: Input the terrain data corresponding to the observation point into the feature extraction model to obtain the terrain vector.
[0107] Because the acquired terrain data may have different data dimensions than the other input wind speed correction model data, it is necessary to input the terrain data into a feature extraction model to obtain a terrain vector. The obtained terrain vector can be matched with the other input wind speed correction model data, such as the wind speed data and flow field data at the observation points.
[0108] For example, terrain data includes parameters such as ground slope and ground complexity.
[0109] B2: The horizontal wind speed at the observation point, the radial wind speed at the reference point, the fluid simulation horizontal wind speed at the observation point, the fluid simulation horizontal wind speed at the reference point, the terrain vector, and the wind measurement height are concatenated to obtain the input feature vector.
[0110] Based on the wind speed data at the observation point, including the horizontal wind speed at the observation point and the radial wind speed at the reference point, and the flow field data at the observation point, including the fluid simulation horizontal wind speed at the observation point and the fluid simulation horizontal wind speed at the reference point, the horizontal wind speed at the observation point, the radial wind speed at the reference point, the fluid simulation horizontal wind speed at the observation point, the fluid simulation horizontal wind speed at the reference point, the terrain vector, and the wind measurement height are concatenated to obtain the input feature vector that can be input into the wind speed correction model.
[0111] B3: Input the input feature vector into the wind speed correction model to obtain the corrected wind speed data.
[0112] By inputting the feature vector into the wind speed correction model, the corrected wind speed data can be obtained. It can be understood that the corrected wind speed data is the standard wind speed data for the observation point.
[0113] Based on the content of B1-B3, it can be seen that by using the terrain data of the terrain where the lidar is located, the wind measurement height corresponding to the observation point, the wind speed data collected by the lidar, and the corresponding flow field data calculated, the wind speed measured by the lidar can be corrected through the wind speed correction model, thereby improving the accuracy of the lidar wind speed measurement.
[0114] In one possible implementation, embodiments of this application also provide a method for training a wind speed correction model. See [link to relevant documentation]. Figure 6 , Figure 6 A flowchart of a wind speed correction model training method provided in this application embodiment is shown below. Figure 6 As shown, the method includes S601-S605:
[0115] S601: Collects wind speed data of the observation points to be trained using lidar.
[0116] To train a wind speed correction model, a large amount of training data is needed. For example, wind speed data from observation points to be trained, collected using lidar in complex terrain.
[0117] In one possible implementation, this application provides a specific method for acquiring wind speed data of observation points to be trained using lidar, including:
[0118] The horizontal wind speed at the observation point to be trained and the radial wind speed at the reference point to be trained are collected by lidar as wind speed data for the observation point to be trained. The reference point to be trained is located on the beam path emitted by the lidar.
[0119] S602: Obtain standard wind speed data of the training observation points measured by the anemometer tower.
[0120] To train the wind speed correction model, the desired standard wind speed data is required. The standard wind speed data for the observation points to be trained can be obtained by measuring the observation points using a wind tower.
[0121] For example, such as Figure 4 As shown, when the observation point to be trained is point C, the standard wind speed data of the observation point to be trained measured by the wind tower is the standard wind speed data of point C.
[0122] S603: Calculate the flow field based on the location of the observation point to be trained, and obtain the flow field data of the observation point to be trained.
[0123] The flow field data at the location of the observation point to be trained is simulation data. For example, the flow field data of the observation point to be trained can be obtained by performing flow field calculations at the location of the observation point.
[0124] In one possible implementation, this application provides a specific method for calculating the flow field based on the location of the observation point to be trained, thereby obtaining the flow field data of the observation point, including:
[0125] The flow field is calculated based on the location of the observation point to be trained and the location of the reference point to be trained, and the simulated horizontal wind speed of the fluid at the observation point to be trained and the simulated horizontal wind speed of the fluid at the reference point to be trained are used as the flow field data of the observation point to be trained.
[0126] S604: Obtain the terrain data and wind height corresponding to the observation point to be trained.
[0127] Since terrain data and other factors affect the accuracy of wind speed measurements obtained by lidar, it is necessary to collect terrain data to train the wind speed correction model. Simultaneously, the wind measurement height corresponding to the observation points to be trained needs to be obtained.
[0128] S605: The wind speed data, flow field data, terrain data, and wind measurement height of the observation points to be trained are used as training data, and the standard wind speed data of the observation points to be trained are used as the labels of the training data to train the wind speed correction model.
[0129] The wind speed data, flow field data, terrain data, and wind measurement height of the observation points to be trained are used as training data, and the standard wind speed data of the observation points to be trained measured by the wind tower are used as the labels of the training data to train the wind speed correction model, so as to obtain the trained wind speed correction model.
[0130] Based on the trained wind speed correction model, the corrected wind speed data output by the model can be obtained from the input wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0131] Based on the content of S601-S605, this application provides a wind speed correction model training method. Wind speed data of the observation point to be trained is collected using a lidar, and standard wind speed data of the observation point to be trained is obtained from a wind tower. Flow field calculations are performed based on the location of the observation point to be trained, resulting in flow field data for that observation point. Additionally, the terrain data and wind measurement height corresponding to the observation point to be trained are acquired. The wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained are used as training data, and the standard wind speed data of the observation point to be trained are used as the label for the training data, thus training a wind speed correction model. In this way, considering the terrain features where the lidar is located, the wind speed correction model is trained based on the wind speed data collected by the lidar, the calculated flow field data, and the standard wind speed data measured by the wind tower. The trained wind speed correction model is then used to correct the wind speed data measured by the lidar, improving the accuracy of the lidar's wind speed measurements.
[0132] In one possible implementation, this application provides a specific implementation method for training a wind speed correction model in step S605, which uses wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained as training data, and standard wind speed data of the observation point to be trained as the label of the training data. The method includes:
[0133] Input the terrain data corresponding to the observation points to be trained into the feature extraction model to obtain the terrain vector to be trained;
[0134] The wind speed data, flow field data, terrain vector, and wind measurement height of the observation point to be trained are concatenated to obtain the feature vector of the model to be trained.
[0135] The wind speed correction model is trained by using the feature vector of the model to be trained as the training data and the standard wind speed data of the observation points to be trained as the labels of the training data.
[0136] It should be noted that some technical details can be found in A1-A3, and will not be repeated here.
[0137] This application embodiment provides another step S605, in which the wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained are used as training data, and the standard wind speed data of the observation point to be trained are used as the label of the training data to train a wind speed correction model, including:
[0138] Input the terrain data corresponding to the observation points to be trained into the feature extraction model to obtain the terrain vector to be trained;
[0139] The horizontal wind speed at the observation point to be trained, the radial wind speed at the reference point to be trained, the horizontal wind speed of the fluid simulation at the observation point to be trained, the horizontal wind speed of the fluid simulation at the reference point to be trained, the terrain vector to be trained, and the wind measurement height corresponding to the observation point to be trained are spliced together to obtain the feature vector of the model to be trained.
[0140] The wind speed correction model is trained by using the feature vector of the model to be trained as the training data and the standard wind speed data of the observation points to be trained as the labels of the training data.
[0141] It should be noted that some technical details can be found in B1-B3, and will not be repeated here.
[0142] Based on the lidar wind speed correction method provided in the above method embodiments, this application also provides a lidar wind speed correction device. The lidar wind speed correction device will be described below with reference to the accompanying drawings. For technical details of the device, please refer to the above method embodiments.
[0143] See Figure 7 ,Should Figure 7 This is a schematic diagram of a lidar wind speed correction device provided in an embodiment of this application. Figure 7 As shown, the lidar wind speed correction device includes:
[0144] The first acquisition unit 701 is used to acquire wind speed data at the observation point via lidar.
[0145] The first calculation unit 702 is used to perform flow field calculations based on the location of the observation point to obtain the flow field data of the observation point.
[0146] The first acquisition unit 703 is used to acquire the terrain data corresponding to the observation point and the wind measurement height corresponding to the observation point.
[0147] The second acquisition unit 704 is used to input the wind speed data, flow field data, terrain data, and wind measurement height of the observation point into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0148] In one possible implementation, the second acquisition unit 704 includes:
[0149] The first acquisition subunit is used to input the terrain data corresponding to the observation point into the feature extraction model to obtain the terrain vector;
[0150] The first stitching subunit is used to stitch together the wind speed data, the flow field data, the terrain vector, and the wind measurement height corresponding to the observation point to obtain the input feature vector.
[0151] The first input subunit is used to input the input feature vector into the wind speed correction model to obtain the corrected wind speed data.
[0152] In one possible implementation, the first acquisition unit 701 is specifically used for:
[0153] The horizontal wind speed at the observation point and the radial wind speed at the reference point are collected by a lidar as wind speed data for the observation point, where the reference point is on the path of the beam emitted by the lidar.
[0154] In one possible implementation, the first computing unit 702 is specifically used for:
[0155] The flow field is calculated based on the location of the observation point and the location of the reference point to obtain the simulated horizontal wind speed at the observation point and the simulated horizontal wind speed at the reference point as the flow field data of the observation point.
[0156] In one possible implementation, the second acquisition unit 704 includes:
[0157] The second acquisition subunit is used to input the terrain data corresponding to the observation point into the feature extraction model to obtain the terrain vector;
[0158] The second splicing subunit is used to splice the horizontal wind speed at the observation point, the radial wind speed at the reference point, the fluid simulation horizontal wind speed at the observation point, the fluid simulation horizontal wind speed at the reference point, the terrain vector, and the wind measurement height corresponding to the observation point to obtain an input feature vector.
[0159] The second input subunit is used to input the input feature vector into the wind speed correction model to obtain the corrected wind speed data.
[0160] In one possible implementation, the modified wind speed data is used for wind turbine power curve testing, wind resource assessment, and / or wind power project assessment.
[0161] In one possible implementation, the device further includes:
[0162] The second acquisition unit is used to acquire wind speed data of the observation points to be trained using lidar.
[0163] The third acquisition unit is used to acquire the standard wind speed data of the observation point to be trained measured by the wind tower;
[0164] The second calculation unit is used to perform flow field calculations based on the location of the observation point to be trained, and to obtain the flow field data of the observation point to be trained.
[0165] The fourth acquisition unit is used to acquire the terrain data corresponding to the observation point to be trained and the wind measurement height corresponding to the observation point to be trained.
[0166] The training unit is used to train a wind speed correction model by using the wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained, and the standard wind speed data of the observation point to be trained as the label of the training data.
[0167] In one possible implementation, the training unit includes:
[0168] The third input subunit is used to input the terrain data corresponding to the observation point to be trained into the feature extraction model to obtain the terrain vector to be trained.
[0169] The third splicing subunit is used to splice the wind speed data, flow field data, terrain vector, and wind measurement height of the observation point to be trained to obtain the feature vector of the model to be trained.
[0170] The first training subunit is used to train a wind speed correction model by using the feature vector of the model to be trained as training data and the standard wind speed data of the observation point to be trained as the label of the training data.
[0171] In one possible implementation, the second acquisition unit is specifically used for:
[0172] The horizontal wind speed at the observation point to be trained and the radial wind speed at the reference point to be trained are collected by lidar as wind speed data of the observation point to be trained. The reference point to be trained is on the path of the beam emitted by the lidar.
[0173] In one possible implementation, the second computing unit is specifically used for:
[0174] The flow field is calculated based on the location of the observation point to be trained and the location of the reference point to be trained, and the simulated horizontal wind speed of the fluid at the observation point to be trained and the simulated horizontal wind speed of the fluid at the reference point to be trained are obtained as the flow field data of the observation point to be trained.
[0175] In one possible implementation, the training unit includes:
[0176] The fourth input subunit is used to input the terrain data corresponding to the observation point to be trained into the feature extraction model to obtain the terrain vector to be trained.
[0177] The fourth splicing subunit is used to splice the horizontal wind speed of the observation point to be trained, the radial wind speed of the reference point to be trained, the fluid simulation horizontal wind speed of the observation point to be trained, the fluid simulation horizontal wind speed of the reference point to be trained, the terrain vector to be trained, and the wind height corresponding to the observation point to be trained to obtain the feature vector of the model to be trained.
[0178] The second training subunit is used to train a wind speed correction model by using the feature vector of the model to be trained as training data and the standard wind speed data of the observation point to be trained as the label of the training data.
[0179] See Figure 8 , Figure 8 A schematic diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown.
[0180] Reference Figure 8 An electronic device according to an exemplary embodiment of the present disclosure includes a processor 82 and a memory 81 configured to store processor-executable instructions, the processor being configured to execute the instructions to implement a lidar wind speed correction method according to an exemplary embodiment of the present disclosure, namely the following steps:
[0181] Wind speed data at observation points are collected using lidar.
[0182] The flow field is calculated based on the location of the observation point to obtain the flow field data of the observation point;
[0183] Obtain the terrain data and wind measurement height corresponding to the observation point;
[0184] The wind speed data, flow field data, terrain data, and wind measurement height of the observation point are input into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
[0185] In addition, this application embodiment also provides a storage medium, which, when the instructions in the storage medium are executed by the processor of the lidar wind speed correction device or the electronic device, enables the lidar wind speed correction device or the electronic device to implement the steps of the lidar wind speed correction method as described in any of the above embodiments.
[0186] In addition, this application also provides a computer program product, which includes a computer program stored in a readable storage medium. At least one processor of the device reads from the storage medium and executes the computer program, causing the device to perform the steps of the lidar wind speed correction method as described in any of the above embodiments.
[0187] This application provides a lidar wind speed correction device and equipment. Wind speed data at observation points is collected using lidar. Flow field calculations are performed based on the location of the observation points to obtain flow field data. Topographic data and the corresponding wind measurement height for each observation point are acquired. Then, the wind speed data, flow field data, topographic data, and wind measurement height are input into a wind speed correction model to obtain corrected wind speed data. The corrected wind speed data represents a more standard wind speed at the observation point location. The wind speed correction model characterizes the correspondence between the wind speed data, flow field data, topographic data, and wind measurement height at the observation point, and the standard wind speed data collected by the wind tower at the observation point. Thus, considering the terrain features where the lidar is located, and based on the wind speed data collected by the lidar and the calculated flow field data, the wind speed correction model can correct the wind speed measured by the lidar, improving the accuracy of lidar wind speed measurements.
[0188] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.
[0189] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0190] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0191] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0192] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for correcting wind speed using a lidar system, characterized in that, The method includes: Wind speed data at observation points are collected using lidar. The flow field is calculated based on the location of the observation point to obtain the flow field data of the observation point; Acquire the terrain data and wind measurement height corresponding to the observation point; The wind speed data, flow field data, terrain data, and wind measurement height of the observation point are input into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
2. The method according to claim 1, characterized in that, The step of inputting the wind speed data, flow field data, terrain data, and wind measurement height at the observation point into the wind speed correction model to obtain the corrected wind speed data includes: The terrain data corresponding to the observation point is input into the feature extraction model to obtain the terrain vector; The wind speed data, flow field data, terrain vector, and wind measurement height corresponding to the observation point are concatenated to obtain the input feature vector. The input feature vector is input into the wind speed correction model to obtain the corrected wind speed data.
3. The method according to claim 1, characterized in that, The wind speed data collected at the observation point via lidar includes: The horizontal wind speed at the observation point and the radial wind speed at the reference point are collected by a lidar as wind speed data for the observation point, where the reference point is on the path of the beam emitted by the lidar.
4. The method according to claim 3, characterized in that, The step of calculating the flow field based on the location of the observation point to obtain the flow field data of the observation point includes: The flow field is calculated based on the location of the observation point and the location of the reference point to obtain the simulated horizontal wind speed at the observation point and the simulated horizontal wind speed at the reference point as the flow field data of the observation point.
5. The method according to claim 4, characterized in that, The step of inputting the wind speed data, flow field data, terrain data, and wind measurement height at the observation point into the wind speed correction model to obtain the corrected wind speed data includes: The terrain data corresponding to the observation point is input into the feature extraction model to obtain the terrain vector; The horizontal wind speed at the observation point, the radial wind speed at the reference point, the fluid simulation horizontal wind speed at the observation point, the fluid simulation horizontal wind speed at the reference point, the terrain vector, and the wind measurement height corresponding to the observation point are concatenated to obtain the input feature vector. The input feature vector is input into the wind speed correction model to obtain the corrected wind speed data.
6. The method according to claim 4, characterized in that, The corrected wind speed data is used for wind turbine power curve testing, wind resource assessment, and / or wind power project assessment.
7. The method according to claim 1, characterized in that, The method further includes: Wind speed data at the observation points to be trained are collected using lidar. Obtain standard wind speed data of the observation points to be trained, as measured by the wind measurement tower; The flow field is calculated based on the location of the observation point to be trained, and the flow field data of the observation point to be trained is obtained. Acquire the terrain data and wind height corresponding to the observation point to be trained; The wind speed data, flow field data, terrain data, and wind measurement height of the observation point to be trained are used as training data, and the standard wind speed data of the observation point to be trained are used as the label of the training data to train the wind speed correction model.
8. A lidar wind speed correction device, characterized in that, The device includes: The first acquisition unit is used to acquire wind speed data at the observation point via lidar. The first calculation unit is used to perform flow field calculations based on the location of the observation point to obtain the flow field data of the observation point. The first acquisition unit is used to acquire the terrain data corresponding to the observation point and the wind measurement height corresponding to the observation point; The second acquisition unit is used to input the wind speed data, flow field data, terrain data, and wind measurement height of the observation point into the wind speed correction model to obtain the corrected wind speed data. The wind speed correction model is used to characterize the correspondence between the target data set and the standard wind speed data of the observation point collected by the wind tower. The target data set includes the wind speed data, flow field data, terrain data, and wind measurement height of the observation point.
9. An electronic device, characterized in that, The electronic device includes: processor; A memory configured to store processor-executable instructions; The processor is configured to execute the instructions to implement the steps of the lidar wind speed correction method as described in any one of claims 1 to 7.
10. A storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the lidar wind speed correction device or electronic device, the lidar wind speed correction device or electronic device shall implement the steps of the lidar wind speed correction method as described in any one of claims 1 to 7.
11. A computer program product, characterized in that, The computer program product includes a computer program stored in a readable storage medium, and at least one processor of the device reads from the storage medium and executes the computer program, causing the device to perform the steps of the lidar wind speed correction method as described in any one of claims 1 to 7.