A wind farm wind resource assessment method, device, electronic device and storage medium

Abstract

The present application provides a method, device, electronic device and storage medium for evaluating wind resources in a wind farm, the method comprising: detecting the maximum closed-loop area of ​​a triple whose entity relationship value is a first preset value and adding multiple nodes; detecting the closed-loop area of ​​a triple whose entity relationship value is a second preset value and deleting all nodes; detecting the nodes in the connected area of ​​a triple whose entity relationship value is a third preset value and performing degradation processing; and downsampling the nodes in the maximum closed-loop area of ​​a triple whose entity relationship value is a fourth preset value according to the relationship between the current number of nodes and the preset number of nodes. The method, device, electronic device and storage medium for evaluating wind resources in a wind farm provided by the embodiments of the present invention, based on the data structure of the knowledge graph, ensure that wind resources can be effectively evaluated in various terrains in the target area, and ensure that the overall number of wind towers is consistent with expectations, and also reduce the work of manual calibration of the orientation in the later stage.

Description

Technical Field

[0001] The present application relates to the field of new energy technologies, and in particular to a method, device, electronic device and storage medium for evaluating wind resources in a wind farm. Background Art

[0002] Before conducting a wind resource assessment, a wind farm usually requires professionals to select a representative location to install a wind tower based on the terrain, historical meteorological data, and preliminary wind field simulation. Subsequently, high-precision wind measuring instruments are used to continuously collect key data such as wind speed, wind direction, and temperature for at least one year to fully capture seasonal changes and wind conditions under extreme weather conditions. This process provides a solid data foundation for feasibility analysis, unit selection, and layout optimization of wind power projects. However, the site selection of wind towers is heavily dependent on manual experience, and often requires a lot of cost in implementation.

[0003] The prior art has also considered using maps to assist in the site selection of wind towers. However, the maps are only used to simply divide the wind farm area into grids, and then assign certain attribute values ​​to the grids to assist in determining whether to set up wind towers in the grids. This type of method has the following defects: the site selection of wind towers generally presets a certain number of wind towers for cost and workload considerations. However, the site selection method in the prior art cannot control the number of wind towers that are finally determined, resulting in the cost of wind resource assessment work often being much higher than expected; when the prior art divides the grids through the map, the grids are often divided evenly or approximately evenly, making the local density of the wind tower unable to adapt to the needs of different terrains such as flat land, sloping land, and rough ground; when the prior art selects a site for a wind tower, it often only considers the location of the wind tower, but cannot automatically determine the best orientation of the wind tower, resulting in the need for additional manual calibration of the orientation of the wind tower. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, electronic device and storage medium for evaluating wind resources in a wind farm.

[0005] In a first aspect, an embodiment of the present invention provides a method for evaluating wind resources in a wind farm, the method comprising:

[0006] Initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm;

[0007] Detecting and obtaining multiple maximum closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset values, and adding multiple nodes in each maximum closed-loop region;

[0008] Detecting and obtaining multiple closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset values, and deleting all nodes in each closed-loop region;

[0009] Detecting and obtaining multiple connected regions formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset values, and performing degeneration processing on the nodes in each of the connected regions;

[0010] Determine the current number of nodes in the wind farm wind measurement knowledge graph, and based on the relationship between the current number of nodes and the preset number of nodes, perform downsampling processing on the nodes in multiple maximum closed-loop areas formed by triples of all entity relationship values ​​in the wind farm wind measurement knowledge graph being a fourth preset value;

[0011] Wind towers are set at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and the wind resource assessment result of the target wind farm is obtained by measuring with the wind towers.

[0012] Optionally, the initializing the wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm specifically includes:

[0013] Randomly sampling a predetermined number of points in the target wind farm area as initial nodes of the wind farm wind measurement knowledge graph, wherein the data stored in each node includes the ground roughness and slope corresponding to the location of the node;

[0014] For each of the initial nodes, detect whether there are other initial nodes within a preset distance. If so, use the two initial nodes as the head entity and the tail entity respectively to construct a triple of the wind farm wind measurement knowledge graph;

[0015] For each triple, according to the ground roughness and slope data stored in the head entity and the tail entity of the triple, the value stored in the entity relationship of the triple is determined as the first preset value, the second preset value, the third preset value or the fourth preset value.

[0016] Optionally, a plurality of nodes are added in each maximum closed-loop region, wherein each maximum closed-loop region specifically includes:

[0017] Determine the area within the region and the average value of the ground roughness among all nodes within the region;

[0018] Determine the frequency of supersampling according to the area and the average value of the ground roughness and perform supersampling of newly added nodes in the maximum closed-loop area;

[0019] Delete all existing nodes in the region.

[0020] Optionally, the data stored in each node in the wind farm wind measurement knowledge graph also includes the direction corresponding to the node's position. Correspondingly, the direction corresponding to each newly added node obtained by supersampling is determined by randomly perturbing the dominant wind direction of the target wind farm.

[0021] Optionally, the step of performing degeneration processing on the nodes in each of the connected regions may include, for each of the connected regions, specifically:

[0022] Determining the average position of all nodes in the connected region;

[0023] The node closest to the average position among all the nodes in the connected area is retained, and other nodes are deleted.

[0024] Optionally, according to the relationship between the current number of nodes and the preset number of nodes, downsampling processing is performed on the nodes in multiple maximum closed-loop areas constituted by triples of all entity relationship values ​​in the wind farm wind measurement knowledge graph being a fourth preset value, wherein for each maximum closed-loop area, specifically including:

[0025] Determine the difference between the current number of nodes and the preset number of nodes;

[0026] The frequency of downsampling is determined according to the difference and the number of nodes in the maximum closed-loop area, downsampling of newly added nodes is performed in the maximum closed-loop area, and all original nodes in the area are deleted.

[0027] Optionally, the data stored in each node in the wind farm wind measurement knowledge graph also includes the direction corresponding to the node's position. Correspondingly, the direction corresponding to each newly added node obtained by downsampling is determined according to the slope corresponding to the node based on the dominant wind direction of the target wind farm.

[0028] In a second aspect, an embodiment of the present invention provides a wind resource assessment device for a wind farm, the device comprising:

[0029] An initialization module, used to initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to a target wind farm;

[0030] A rough ground processing module is used to detect multiple maximum closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset value, and add multiple nodes in each maximum closed-loop area;

[0031] An abnormal terrain processing module is used to detect multiple closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset value, and delete all nodes in each closed-loop area;

[0032] A flat ground processing module is used to detect multiple connected areas formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are third preset values, and perform degeneration processing on the nodes in each of the connected areas;

[0033] A slope processing module is used to determine the current number of nodes in the wind farm wind measurement knowledge graph, and according to the relationship between the current number of nodes and the preset number of nodes, downsample the nodes in multiple maximum closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the fourth preset value (slope);

[0034] The wind resource assessment module is used to set up wind measurement towers at the positions of the remaining nodes in the wind farm wind measurement knowledge graph, and use the wind measurement towers to measure and obtain the wind resource assessment results of the target wind farm.

[0035] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0036] one or more processors;

[0037] A memory for storing one or more programs;

[0038] When the one or more programs are executed by the one or more processors, the one or more processors execute the method as described in the first aspect.

[0039] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having executable instructions stored thereon, wherein when the executable instructions are executed by a processor, the processor executes the method as described in the first aspect.

[0040] The wind resource assessment method, device, electronic device and storage medium for a wind farm provided by the embodiments of the present invention customize the number of wind towers in different terrain areas of a wind farm based on the data structure of a knowledge graph to ensure that wind resources can be effectively assessed in various terrains in the target area; by dynamically adjusting the number of wind towers in different terrains, the overall number of wind towers is ensured to be consistent with expectations, thereby controlling the cost of wind resource assessment work; in addition, the orientation of wind farms in different terrains is also dynamically adjusted, so that the position and orientation basis of the wind towers are finally obtained at the same time, greatly reducing the work of manual calibration of the orientation in the later stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solution of the embodiments of the present application, the drawings required for use in the embodiments of the present application are briefly introduced below.

[0042] Figure 1 A schematic diagram of a flow chart of a method for evaluating wind resources in a wind farm provided by an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a process for initializing a wind farm wind measurement knowledge graph provided by an embodiment of the present invention;

[0044] Figure 3A schematic diagram of a flow chart of a rough terrain supersampling method provided by an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of the structure of a wind resource assessment device for a wind farm provided by an embodiment of the present invention;

[0046] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0048] Similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0049] Before conducting a wind resource assessment, a wind farm usually requires professionals to select a representative location to install a wind tower based on the terrain, historical meteorological data, and preliminary wind field simulation. Subsequently, high-precision wind measuring instruments are used to continuously collect key data such as wind speed, wind direction, and temperature for at least one year to fully capture seasonal changes and wind conditions under extreme weather conditions. This process provides a solid data foundation for feasibility analysis, unit selection, and layout optimization of wind power projects. However, the site selection of wind towers is heavily dependent on manual experience, and often requires a lot of cost in implementation.

[0050] The prior art has also considered using maps to assist in the site selection of wind towers. However, the maps are only used to simply divide the wind farm area into grids, and then assign certain attribute values ​​to the grids to assist in determining whether to set up wind towers in the grids. This type of method has the following defects: the site selection of wind towers generally presets a certain number of wind towers for cost and workload considerations. However, the site selection method in the prior art cannot control the number of wind towers that are finally determined, resulting in the cost of wind resource assessment work often being much higher than expected; when the prior art divides the grids through the map, the grids are often divided evenly or approximately evenly, making the local density of the wind tower unable to adapt to the needs of different terrains such as flat land, sloping land, and rough ground; when the prior art selects a site for a wind tower, it often only considers the location of the wind tower, but cannot automatically determine the best orientation of the wind tower, resulting in the need for additional manual calibration of the orientation of the wind tower.

[0051] Based on this, an embodiment of the present invention provides a method for evaluating wind resources in a wind farm. Figure 1The flowchart of the wind resource assessment method for a wind farm provided by an embodiment of the present invention is shown, which specifically includes steps S110 to S160.

[0052] Step S110, initializing a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm.

[0053] When conducting a wind resource assessment for a wind farm, it is necessary to first determine the location of the wind tower, so as to ensure that the result of the wind resource assessment can accurately reflect the regional characteristics of the target wind farm. The embodiment of the present invention determines the location of the wind tower based on the wind farm wind measurement knowledge graph. The size of the wind farm wind measurement knowledge graph can be understood as being the same as the physical size of the target wind farm area, that is, the node position in the wind farm wind measurement knowledge graph corresponds to the actual location of the wind tower in the target wind farm area. The node finally retained in the wind farm wind measurement knowledge graph is the actual location of the wind tower finally determined to be set.

[0054] In this step, the wind farm wind measurement knowledge graph needs to be initialized first. Figure 2 A flow chart of a method for initializing a wind farm wind measurement knowledge graph is shown, which specifically includes steps S111 to S113.

[0055] Step S111, randomly sampling a predetermined number of points in the target wind farm area as initial nodes of the wind farm wind measurement knowledge graph, and the data stored in each node includes the ground roughness and slope corresponding to the location of the node.

[0056] In this step, the initial nodes in the wind farm wind measurement knowledge graph are first determined. The number of initial nodes is determined by the number of wind towers predetermined by the user when performing wind measurement work. Specifically, the predetermined number of points are randomly sampled in the target wind farm area, and the position of the sampling point is the initial position of the node. In practical applications, the random sampling points can cover various types of areas in the target wind farm, such as flat land, sloping land, rough ground, abnormal terrain, etc. These nodes are the basic elements for constructing the wind farm wind measurement knowledge graph.

[0057] As a node in the knowledge graph, the initial node will store data such as ground roughness and slope corresponding to the node's location. Ground roughness and slope data at different locations can be obtained through remote sensing technology, digital elevation model, GIS analysis, manual survey, etc.

[0058] Step S112: for each of the initial nodes, detect whether there are other initial nodes within a preset distance. If so, use the two initial nodes as the head entity and the tail entity respectively to construct a triple of the wind farm wind measurement knowledge graph.

[0059] The basic structure for constructing a knowledge graph is a triple, and a triple consists of three parts: a head entity, a tail entity, and an entity relationship. The head entity and the tail entity in the embodiment of the present invention are the nodes in the wind farm wind measurement knowledge graph. This step needs to determine which node pairs in the initial nodes will be used as the head entity and the tail entity to construct the triple.

[0060] In this step, for each of the initial nodes, it is necessary to detect whether there are other initial nodes within a preset distance. The preset distance is a predefined spatial or logical distance threshold. It is used to determine whether two initial nodes are close enough to have some kind of association or influence. This distance can be physical, such as meters or kilometers; it can also be logical, such as calculated based on some similarity or association parameters.

[0061] In specific implementation, the system or algorithm of the embodiment of the present invention will first traverse or select all the initial nodes in the wind farm wind measurement knowledge graph as processing objects. For each selected initial node, the system will search whether there are other initial nodes within a preset distance range. If other initial nodes are found within the preset distance, then the two nodes are considered to have some kind of association or mutual influence. At this time, the system will designate the two nodes as the head entity and tail entity of the triple, respectively, and the entity relationship value is temporarily vacant. Thus, one or more triples are constructed, and these triples are then added to the data structure of the wind farm wind measurement knowledge graph. The above process is repeated for each initial node until all nodes have been processed.

[0062] Step S113, for each triple, according to the ground roughness and slope data stored in the head entity and the tail entity of the triple, determine the value stored in the entity relationship of the triple as the first preset value, the second preset value, the third preset value or the fourth preset value.

[0063] Each triple is composed of a head entity, a relationship, and a tail entity. The previous step determines a series of attributes carried by the head entity and the tail entity, including but not limited to ground roughness and slope data. This step needs to determine the entity relationship between the head entity and the tail entity. The entity relationship in the embodiment of the present invention is used to describe the relationship between the terrain types in which the head entity and the tail entity are located. Specifically, the terrain types can be divided into rough ground, abnormal terrain, flat land, and sloping land, and the corresponding values ​​stored in the entity relationship are determined as the first preset value, the second preset value, the third preset value, or the fourth preset value. Among them, abnormal terrain includes transitional terrain, obstructed terrain, and other abnormal conditions that are not suitable for setting up a wind tower.

[0064] The value stored in the entity relationship of the triple is determined by the ground roughness and slope values ​​stored in the head entity and the tail entity of the triple. When the slope values ​​in the head entity and the tail entity are both close to 0 (less than the preset slope constant), and the ground roughness values ​​are both significantly greater than 0 (greater than the preset roughness constant), the value stored in the entity relationship of the triple is determined as the first preset value, representing a rough ground; when the slope values ​​in the head entity and the tail entity are both close to 0 (less than the preset slope constant), and the ground roughness values ​​are both close to 0 (less than the preset roughness constant), the value stored in the entity relationship of the triple is determined as the third preset value, representing a flat ground; when the slope values ​​in the head entity and the tail entity are both significantly greater than 0 (greater than the preset slope constant), and the ground roughness values ​​are both close to 0 (less than the preset roughness constant), the value stored in the entity relationship of the triple is determined as the fourth preset value, representing a sloping ground; in other cases, the value stored in the entity relationship of the triple is determined as the second preset value.

[0065] So far, multiple triple structures have been constructed in the wind farm wind measurement knowledge graph, and the initialization step of the wind farm wind measurement knowledge graph is completed.

[0066] Step S120, detecting and obtaining a plurality of maximum closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset values, and adding a plurality of nodes in each maximum closed-loop region.

[0067] After initializing the wind farm wind measurement knowledge graph, this step first processes the knowledge graph part of the rough ground area. Rough ground is one of the common terrains of wind farms, such as woods, buildings, hills, etc., which will hinder and disturb the flow of wind, resulting in significant changes in wind resource parameters such as wind speed and wind direction in the horizontal direction. This change makes the wind resource conditions at different locations in the same wind farm area likely to be quite different. The main function of the wind tower is to collect meteorological data such as wind speed and wind direction at different locations, providing an important basis for the design, construction and operation of wind farms. The data obtained by the wind tower can evaluate the wind resource conditions of the wind farm, select the appropriate wind turbine type and layout, and optimize the power generation efficiency of the wind farm. Therefore, under rough ground conditions, in order to more accurately reflect the wind resource distribution characteristics in the wind farm area, it is necessary to increase the number of wind towers or improve the representativeness of wind towers. For example, by encrypting the setting of wind towers, more wind resource change information in more areas can be captured, and the accuracy of wind farm design can be improved.

[0068] In this step, the system will use graph theory algorithms (such as depth-first search DFS, breadth-first search BFS or loop detection algorithms) based on the selected triplets to find and construct all possible closed loop regions formed by these triplets. A closed loop region refers to a closed path in the graph that starts from a certain node, moves along the edge and finally returns to the starting node. The maximum closed loop region refers to a closed loop that no longer includes other closed loop regions.

[0069] The above process may detect multiple maximum closed-loop areas representing rough ground. For each of these maximum closed-loop areas, oversampling is required in these areas to meet the requirements of rough ground for the density of wind towers. Figure 3 A schematic flow chart of a rough surface supersampling method provided in real time by the present invention is shown, which specifically includes steps S121 to S123.

[0070] Step S121, determining the area of ​​the region and the average value of the ground roughness of all nodes in the region.

[0071] This step first requires that the boundaries of the area to be processed be clearly defined, and then the area of ​​the area is calculated using appropriate measurement tools (such as GIS software, remote sensing technology, or field measurements). This area will serve as the basis for subsequent calculations and analysis. The surface roughness is stored in all nodes within the maximum closed-loop area, and the average surface roughness of the entire area is obtained by calculating the arithmetic mean or weighted mean of these values.

[0072] Step S122: determining the frequency of super sampling according to the area and the average value of the ground roughness, and performing super sampling of newly added nodes in the maximum closed-loop area.

[0073] Supersampling is a technique that increases the density of data points to improve the accuracy of a simulation or analysis. In this step, the frequency of supersampling in an area is determined based on the area and average ground roughness calculated previously. Generally speaking, areas with larger areas or higher ground roughness may require more intensive supersampling to improve the accuracy and reliability of wind tower site selection.

[0074] Specifically, a lookup table of two variables, regional area and average ground roughness, and the oversampling frequency can be preset. The lookup table can be preset based on the statistical values ​​of the rough ground wind farm site selection in history or artificial experience values. By using the lookup table, the frequency of oversampling or the number of newly added nodes for oversampling can be determined based on the two variables of regional area and average ground roughness, and finally random sampling is performed within the maximum closed-loop area with this frequency or number of nodes. The lookup table in the embodiment of the present invention can also adjust the values ​​in the lookup table as a whole according to the preset number of wind towers for the project, so as to avoid the situation where the total number of nodes in the entire target wind farm far exceeds the preset number of wind towers after oversampling of newly added nodes, thereby causing the cost of wind resource assessment to be often much higher than expected.

[0075] Step S123, deleting all existing nodes in the area.

[0076] After completing the oversampling and adding enough nodes, in order to maintain data consistency and reduce redundancy, you may need to delete all the original nodes in the area. In this process, you also need to consider data backup and recovery strategies to prevent unexpected situations that may cause data loss when the original node data is deleted by mistake.

[0077] Step S130, detecting and obtaining multiple closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are within a second preset value, and deleting all nodes in each closed-loop region.

[0078] In wind farm planning and operation, wind measurement data is an important basis for evaluating wind energy resources and designing wind farm layout. The wind measurement knowledge graph is a database that structuredly represents entities such as wind stations, terrain, wind direction, wind speed, and their relationships. However, certain terrain conditions (such as transitional terrain, obstructed terrain, etc.) may cause abnormal wind measurement data and affect the effectiveness of wind farm design and operation. Therefore, it is necessary to identify and process these abnormal terrain areas to improve the accuracy and reliability of the data.

[0079] The specific meaning represented by the second preset value in the embodiment of the present invention is the value stored in the entity relationship of the triples distributed in the abnormal terrain. In this step, all triples in the wind farm wind measurement knowledge graph are traversed to find those triples whose relationship values ​​are the second preset values. Based on these triples, a closed-loop area connected by multiple such triples is identified and constructed. These closed-loop areas represent areas with similar abnormal terrain features. Once it is confirmed that a closed-loop area does correspond to abnormal terrain, all nodes in the area need to be deleted from the knowledge graph. For areas identified as abnormal terrain, wind farm planners do not need to consider the layout of wind towers in the area, and may also take other measures to overcome the impact of terrain on wind measurement data and manually set up wind towers.

[0080] Step S140: Detect and obtain multiple connected regions consisting of triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset value, and perform degeneration processing on the nodes in each of the connected regions.

[0081] This step involves identifying the connected areas formed by triplets of specific relationship values ​​(third preset values) in the wind farm wind measurement knowledge graph, and simplifying each such connected area into a single node, while deleting other nodes in the area. This process is intended to reduce data redundancy while maintaining the ability to identify flat land. The position of the single node can be the node with the highest degree in the connected area, or it can be the node closest to the center position in the connected area. Another implementation method is to first determine the average position of all nodes in the connected area, then retain the nodes closest to the average position among all nodes in the connected area, and delete other nodes.

[0082] In the knowledge graph, if there is a triple relationship between two nodes and the entity relationship value is equal to a third preset value, then the two nodes belong to the same connected area. If any node in a set of multiple nodes belongs to the same connected area as at least one other node in the set, then the multiple nodes also constitute a connected area. After the wind farm wind measurement knowledge graph is initialized, there are multiple such connected areas in the knowledge graph, which are used to indicate multiple flat areas in the target wind farm area. Since the wind measurement properties of each location in a flat area are basically the same, only one wind measurement tower is retained in a flat area to meet the needs of wind resource assessment.

[0083] Step S150, determine the current number of nodes in the wind farm wind measurement knowledge graph, and based on the relationship between the current number of nodes and the preset number of nodes, downsample the nodes in multiple maximum closed-loop areas formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the fourth preset value.

[0084] This step first needs to traverse the entire wind farm wind measurement knowledge graph, count and determine the total number of all nodes in the current graph. After the above steps, the number of nodes in the wind farm wind measurement knowledge graph is different from the number of nodes when initialization is completed. Generally speaking, due to the increase and decrease of the number of nodes in the above steps, especially the oversampling of a large number of new nodes for rough ground in step S120, the total number of nodes at this time will generally be more than the total number of nodes in the wind farm wind measurement knowledge graph after initialization, even if some nodes are degraded or downsampled in steps S130 and S140.

[0085] After determining the current number of nodes and the preset number of nodes, the relationship between the two values ​​needs to be compared. If the current number of nodes is less than or equal to the preset number of nodes, then no processing may be required because the scale of nodes used to characterize the location of the wind tower is beyond expectations. However, if the current number of nodes exceeds the preset number of nodes, then measures need to be taken to reduce the number of slope nodes represented by the fourth preset value so that the node scale remains at the expected number.

[0086] This step needs to identify all triples whose entity relationship values ​​are the fourth preset value. The fourth preset value is a specific relationship value, which defines the area near the triple as a sloped terrain. Then, based on these triples, it is necessary to find the closed-loop area formed by them. A closed-loop area refers to a series of interconnected triplets whose entity relationship values ​​are the fourth preset value, which form a closed-loop structure, that is, starting from a node, through a series of relationships and finally returning to the node itself. The maximum closed-loop area refers to such a closed loop, which no longer includes other closed-loop areas.

[0087] For each identified maximum closed-loop area, it represents a continuous slope area in the target wind farm. There are two phenomena in the slope area: in the gently rising or falling slope area, parameters such as wind speed are different from those in the flat area, but the data consistency is strong. At this time, the density of wind tower settings can be reduced; in the non-uniformly rising or falling slope area, terrain changes will produce turbulence and wake effects, which will significantly reduce the wind speed and increase the volatility of the wind speed. The wind speed data in the turbulence and wake area has a low reference value for wind farm development because the wind energy utilization efficiency in these areas is also low. Based on the above two situations, in the slope terrain, wind towers are usually suitable for being set up in these areas with a lower density. Therefore, in the original wind farm wind measurement knowledge graph, the node density of the slope area needs to be downsampled.

[0088] Specifically, assuming that the total number of nodes in the wind farm wind measurement knowledge graph is M, and the preset number of nodes is N, where M>N. Then, for each of the multiple maximum closed-loop regions formed by the triples whose entity relationship value is the fourth preset value, the nodes are downsampled at a ratio of N / M. That is, if the number of existing nodes in a certain slope area is X, the specific operation is to delete the X nodes and perform downsampling in the slope area. The user can use random sampling or uniform sampling to resample nodes.

[0089] Step S160, setting up wind towers at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and using the wind towers to measure to obtain wind resource assessment results of the target wind farm.

[0090] In this step, it is necessary to identify the remaining nodes in the current wind measurement knowledge graph and install a wind tower. The height, structure and sensor configuration of the wind tower should be customized according to specific needs to ensure that the wind resource characteristics of the target wind farm can be accurately captured. After the wind tower is installed, we start collecting data. This includes meteorological parameters such as wind speed, wind direction, temperature, humidity, and possible more complex parameters such as turbulence intensity and wind shear. The data collection period and frequency should be set according to the evaluation requirements. Using the collected data, we conduct wind resource assessment. This usually involves data preprocessing (such as denoising, calibration, etc.), statistical analysis (such as wind speed distribution fitting, wind direction frequency calculation, etc.) and more advanced analysis methods (such as wind resource simulation, power generation potential prediction, etc.). Finally, we generate detailed reports and recommendations based on the results of the wind resource assessment. These reports may include the overall wind resource status of the wind farm, the wind energy potential of each region, and optimization recommendations for specific regions. This information is of great guiding significance for the planning, construction and operation of wind farms.

[0091] Based on the above embodiments, the data stored in each node in the wind farm wind measurement knowledge graph in the embodiment of the present invention includes not only the ground roughness and slope corresponding to the node's position, but also the direction corresponding to the node's position, which specifically represents the clockwise angle between the direction of the wind tower located at this position and the dominant wind direction of the target wind farm area. In step S110, the direction values ​​in all initialized nodes are initialized to 0. In step S120, the direction corresponding to each newly added node obtained by oversampling is determined by randomly perturbing the dominant wind direction of the target wind farm. In step S140, the direction value in the node obtained after degradation is still 0, which remains consistent with the dominant wind direction. In step S150, the direction corresponding to each newly added node obtained by downsampling is determined based on the slope corresponding to the node on the basis of the dominant wind direction of the target wind farm.

[0092] Therefore, the orientation values ​​stored in each node in the wind farm wind measurement knowledge graph are finally customized based on nodes of different terrains, providing a more effective basis for the site selection of wind farm wind towers and wind resource assessment.

[0093] The wind resource assessment method for a wind farm provided in an embodiment of the present invention customizes the number of wind towers in different terrain areas of a wind farm based on the data structure of a knowledge graph to ensure that wind resources can be effectively assessed in various terrains in the target area; by dynamically adjusting the number of wind towers in different terrains, the overall number of wind towers is ensured to be consistent with expectations, thereby controlling the cost of wind resource assessment work; in addition, the orientation of wind farms in different terrains is also dynamically adjusted, so that the position and orientation basis of the wind towers are ultimately obtained at the same time, greatly reducing the work of manual calibration of the orientation in the later stage.

[0094] Specifically, in the process of downsampling nodes in the slope area of ​​the wind farm, the orientation of each newly added node is not determined arbitrarily, but is adjusted based on the dominant wind direction of the target wind farm and the slope of the node. In other words, the orientation of each newly added node must take into account both the dominant wind direction of the wind farm to ensure that the wind turbine can capture wind energy to the maximum extent, and the slope of the node to cope with the impact of the terrain on the wind. Parameters such as the dominant wind direction and the slope corresponding to the node position can be input into the CFD (computational fluid dynamics) model for simulation, so as to calculate the orientation of the newly added node. Since the nodes have been downsampled in the slope area, the complexity of the calculation process is relatively low.

[0095] Based on any of the above embodiments, Figure 4 The following is a schematic diagram showing the structure of a wind resource assessment device for a wind farm provided by an embodiment of the present invention, and the specific contents are as follows:

[0096] An initialization module 410 is used to initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to a target wind farm;

[0097] The rough ground processing module 420 is used to detect multiple maximum closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset value, and add multiple nodes in each maximum closed-loop area;

[0098] The abnormal terrain processing module 430 is used to detect multiple closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset value, and delete all nodes in each closed-loop area;

[0099] The flat ground processing module 440 is used to detect multiple connected areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset value, and perform degeneration processing on the nodes in each of the connected areas;

[0100] The slope processing module 450 is used to determine the current number of nodes in the wind farm wind measurement knowledge graph, and according to the relationship between the current number of nodes and the preset number of nodes, downsample the nodes in multiple maximum closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the fourth preset value (slope);

[0101] The wind resource assessment module 460 is used to set up wind towers at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and use the wind towers to measure and obtain the wind resource assessment result of the target wind farm.

[0102] The wind resource assessment device for a wind farm provided by an embodiment of the present invention customizes the number of wind towers in different terrain areas of a wind farm based on the data structure of a knowledge graph to ensure that wind resources can be effectively assessed in various terrains in the target area; by dynamically adjusting the number of wind towers in different terrains, the overall number of wind towers is ensured to be consistent with expectations, thereby controlling the cost of wind resource assessment work; in addition, the orientation of wind farms in different terrains is also dynamically adjusted, so that the position and orientation basis of the wind towers are ultimately obtained at the same time, greatly reducing the work of manual calibration of the orientation in the later stage.

[0103] Based on any of the above embodiments, Figure 5 The physical structure diagram of the electronic device provided by the embodiment of the present invention is shown, and the electronic device may include: a processor (processor) 510, a communication interface (CommunicationsInterface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute the following method:

[0104] Initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm;

[0105] Detecting and obtaining multiple maximum closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset values, and adding multiple nodes in each maximum closed-loop region;

[0106] Detecting and obtaining multiple closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset values, and deleting all nodes in each closed-loop region;

[0107] Detecting and obtaining multiple connected regions formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset values, and performing degeneration processing on the nodes in each of the connected regions;

[0108] Determine the current number of nodes in the wind farm wind measurement knowledge graph, and based on the relationship between the current number of nodes and the preset number of nodes, perform downsampling processing on the nodes in multiple maximum closed-loop areas formed by triples of all entity relationship values ​​in the wind farm wind measurement knowledge graph being a fourth preset value;

[0109] Wind towers are set at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and the wind resource assessment result of the target wind farm is obtained by measuring with the wind towers.

[0110] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the embodiment of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0111] On the other hand, an embodiment of the present invention further provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method provided in each of the above embodiments is implemented, for example, including:

[0112] Initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm;

[0113] Detecting and obtaining multiple maximum closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset values, and adding multiple nodes in each maximum closed-loop region;

[0114] Detecting and obtaining multiple closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset values, and deleting all nodes in each closed-loop region;

[0115] Detecting and obtaining multiple connected regions formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset values, and performing degeneration processing on the nodes in each of the connected regions;

[0116] Determine the current number of nodes in the wind farm wind measurement knowledge graph, and based on the relationship between the current number of nodes and the preset number of nodes, perform downsampling processing on the nodes in multiple maximum closed-loop areas formed by triples of all entity relationship values ​​in the wind farm wind measurement knowledge graph being a fourth preset value;

[0117] Wind towers are set at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and the wind resource assessment result of the target wind farm is obtained by measuring with the wind towers.

[0118] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0119] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating wind resources in a wind farm, characterized in that: The method comprises: Initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm; Detecting and obtaining multiple maximum closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset values, and adding multiple nodes in each maximum closed-loop region; Detecting and obtaining multiple closed-loop regions formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset values, and deleting all nodes in each closed-loop region; Detecting and obtaining multiple connected regions formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the third preset values, and performing degeneration processing on the nodes in each of the connected regions; Determine the current number of nodes in the wind farm wind measurement knowledge graph, and based on the relationship between the current number of nodes and the preset number of nodes, perform downsampling processing on the nodes in multiple maximum closed-loop areas constituted by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the fourth preset value; wherein the downsampling processing is specifically: delete all nodes in the maximum closed-loop area, and resample in the maximum closed-loop area with the ratio of the preset number of nodes to the current number of nodes as the downsampling frequency; Wind towers are set at the locations of the remaining nodes in the wind farm wind measurement knowledge graph, and the wind resource assessment result of the target wind farm is obtained by measuring with the wind towers.

2. The wind resource assessment method for a wind farm according to claim 1, characterized in that: The initialization of the wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to the target wind farm specifically includes: Randomly sampling a predetermined number of points in the target wind farm area as initial nodes of the wind farm wind measurement knowledge graph, wherein the data stored in each node includes the ground roughness and slope corresponding to the location of the node; For each of the initial nodes, detect whether there are other initial nodes within a preset distance. If so, use the two initial nodes as the head entity and the tail entity respectively to construct a triple of the wind farm wind measurement knowledge graph; For each triple, according to the ground roughness and slope data stored in the head entity and the tail entity of the triple, the value stored in the entity relationship of the triple is determined as the first preset value, the second preset value, the third preset value or the fourth preset value.

3. The wind resource assessment method for a wind farm according to claim 2, characterized in that: A plurality of nodes are added in each maximum closed loop area, wherein each maximum closed loop area specifically includes: Determine the area within the region and the average value of the ground roughness among all nodes within the region; Determine the frequency of supersampling according to the area and the average value of the ground roughness and perform supersampling of newly added nodes in the maximum closed-loop area; Delete all existing nodes in the region.

4. The wind resource assessment method for a wind farm according to claim 3, characterized in that: The data stored in each node in the wind farm wind measurement knowledge graph also includes the direction corresponding to the node's position. Correspondingly, the direction corresponding to each newly added node obtained by supersampling is determined by randomly perturbing the dominant wind direction of the target wind farm.

5. The wind resource assessment method for a wind farm according to claim 2, characterized in that: The step of performing degeneration processing on the nodes in each of the connected regions may include, for each of the connected regions, specifically: Determining the average position of all nodes in the connected region; The node closest to the average position among all the nodes in the connected area is retained, and other nodes are deleted.

6. The wind resource assessment method for a wind farm according to claim 2, characterized in that: The data stored in each node in the wind farm wind measurement knowledge graph also includes the direction corresponding to the node's position. Correspondingly, the direction corresponding to each newly added node obtained by downsampling is determined based on the dominant wind direction of the target wind farm according to the slope corresponding to the node.

7. A wind resource assessment device for a wind farm, characterized in that: The device comprises: An initialization module, used to initialize a wind farm wind measurement knowledge graph including a predetermined number of nodes corresponding to a target wind farm; A rough ground processing module is used to detect multiple maximum closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the first preset value, and add multiple nodes in each maximum closed-loop area; An abnormal terrain processing module is used to detect multiple closed-loop areas formed by triples whose entity relationship values ​​in the wind farm wind measurement knowledge graph are the second preset value, and delete all nodes in each closed-loop area; A flat ground processing module is used to detect multiple connected areas formed by triplets whose entity relationship values ​​in the wind farm wind measurement knowledge graph are third preset values, and perform degeneration processing on the nodes in each of the connected areas; The slope processing module is used to determine the current number of nodes in the wind farm wind measurement knowledge graph, and according to the relationship between the current number of nodes and the preset number of nodes, perform downsampling processing on the nodes in multiple maximum closed-loop areas formed by triples of all entity relationship values ​​in the wind farm wind measurement knowledge graph being a fourth preset value; wherein the downsampling processing is specifically: deleting all nodes in the maximum closed-loop area, and resampling in the maximum closed-loop area with the ratio of the preset number of nodes to the current number of nodes as the downsampling frequency; The wind resource assessment module is used to set up wind measurement towers at the positions of the remaining nodes in the wind farm wind measurement knowledge graph, and use the wind measurement towers to measure and obtain the wind resource assessment results of the target wind farm.

8. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors execute the method as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium having executable instructions stored thereon, characterized in that: When the executable instructions are executed by a processor, the processor executes the method according to any one of claims 1 to 6.

Images