Meteorological state prediction method and system based on big data

Abstract

The embodiment of the invention discloses a meteorological state prediction method and system based on big data, and belongs to the technical field of meteorological data analysis. By acquiring the observation data of the multi-source wind power plant and identifying the state trend description event of the target meteorological observation data, information in the data can be fully mined, and a detailed description distribution label is obtained. The scale coupling evolution weight is determined based on the labels, so that the influence degree of meteorological elements on the overall meteorological state under different scales can be accurately mastered. Global trend evolution parameters and multi-source coupling evolution discrete indexes are obtained, the overall characteristics of various meteorological elements and multi-source data can be comprehensively considered, and the comprehensiveness of meteorological state analysis is improved. And state prediction viewpoints and confidence features are determined according to the parameters, so that a prediction result has a clear direction and credibility evaluation. The finally determined meteorological state prediction result is higher in accuracy, and reliable support can be provided for operation management of the wind power plant.

Description

Technical Field

[0001] The embodiments of the present application belong to the technical field of meteorological data analysis, and specifically relate to a meteorological state prediction method and system based on big data. Background Art

[0002] In the operation and management of wind farms, accurate weather forecasting is crucial. Traditional weather forecasting methods often fail to meet the high-precision requirements under complex conditions of wind farms. Meteorological observation data from a single data source has limitations and cannot fully reflect the meteorological conditions of wind farms. In addition, when analyzing meteorological data, there is a lack of effective means to comprehensively consider the impact of meteorological elements at different scales and the differences between different data sources. This leads to inaccurate weather forecast results and cannot provide a reliable basis for the operation and management of wind farms. For example, the layout of wind turbines may be unreasonable, the power generation efficiency cannot be optimized, and the safety of wind turbines under complex meteorological conditions is difficult to guarantee. Summary of the invention

[0003] The embodiments of the present application provide a method and system for weather condition prediction based on big data, which can solve or partially solve the technical problems involved in the above-mentioned background technology.

[0004] The embodiment of the present application provides a meteorological state prediction method based on big data, which is applied to a meteorological state prediction system. The method includes: obtaining multi-source wind farm observation data, and performing state trend description event recognition on target meteorological observation data included in the multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data; determining the scale coupling evolution weight of at least one set target scale coupling evolution vector in the target meteorological observation data through the description distribution labels of the several target state trend description events; obtaining the global trend evolution parameter and multi-source coupling evolution discrete index of each set target scale coupling evolution vector, wherein the global trend evolution parameter and the multi-source coupling evolution discrete index are obtained by processing the scale coupling evolution weight of each set target scale coupling evolution vector of several different target meteorological observation data; determining the state prediction viewpoint of each set target scale coupling evolution vector and the confidence feature of each state prediction viewpoint according to the global trend evolution parameter, the multi-source coupling evolution discrete index and the scale coupling evolution weight of each set target scale coupling evolution vector; determining the meteorological state prediction result of the multi-source wind farm observation data through the state prediction viewpoint of each set target scale coupling evolution vector.

[0005] An embodiment of the present application provides a meteorological status prediction system, comprising at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the above method.

[0006] An embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above method are implemented.

[0007] The embodiment of the present application can fully mine the information in the data and obtain detailed descriptive distribution labels by acquiring multi-source wind farm observation data and performing state trend description event identification of target meteorological observation data. Determining the scale coupling evolution weights based on these labels helps to accurately grasp the degree of influence of meteorological elements on the overall meteorological state at different scales. Obtaining global trend evolution parameters and multi-source coupling evolution discrete indicators can comprehensively consider the overall characteristics of various meteorological elements and multi-source data to improve the comprehensiveness of meteorological state analysis. Based on these parameters, the state prediction viewpoint and confidence characteristics are determined, so that the prediction results have both a clear direction and a credibility assessment. The final meteorological state prediction results are more accurate and can provide reliable support for wind farm operation management (optimizing wind turbine layout, improving power generation efficiency, and ensuring safe operation of wind turbines). BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flowchart of a meteorological condition prediction method based on big data provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the structure of a meteorological condition prediction system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0011] The terms "first", "second", etc. in the embodiments of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally a class, and the number of objects is not limited. For example, the first object can be one or more. In addition, in the embodiments of the present application, "and / or" represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0012] Figure 1 A meteorological state prediction method based on big data is shown, which is applied to a meteorological state prediction system. The method includes the following steps 101-105.

[0013] Step 101: The meteorological state prediction system obtains multi-source wind farm observation data, and performs state trend description event recognition on target meteorological observation data included in the multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data.

[0014] Step 102: The meteorological state prediction system determines the scale coupling evolution weight of at least one set target scale coupling evolution vector in the target meteorological observation data through the description distribution labels of the plurality of target state trend description events.

[0015] Step 103, the meteorological state prediction system obtains the global trend evolution parameters and multi-source coupling evolution discrete indicators of each set target scale coupling evolution vector, wherein the global trend evolution parameters and multi-source coupling evolution discrete indicators are obtained by processing the scale coupling evolution weights of each set target scale coupling evolution vector of a number of different target meteorological observation data.

[0016] Step 104: The meteorological state prediction system determines the state prediction viewpoints of each set target scale coupling evolution vector and the confidence characteristics of each state prediction viewpoint based on the global trend evolution parameters, multi-source coupling evolution discrete indicators and scale coupling evolution weights of each set target scale coupling evolution vector.

[0017] Step 105: The meteorological state prediction system determines the meteorological state prediction result of the multi-source wind farm observation data through the state prediction viewpoints of the coupled evolution vectors of the set target scales.

[0018] First of all, the first step of the meteorological state prediction system is to obtain multi-source wind farm observation data. The sources of wind farm observation data are diverse, such as direct observation data from meteorological stations, wind profile radar data, satellite remote sensing data, etc. These multi-source data contain rich meteorological information, and the system needs to identify the state trend description events of the target meteorological observation data. For example, if the target meteorological observation data is wind speed and wind direction data, the state trend description event recognition may identify the continuous increase in wind speed (for example, the wind speed in each period increases by more than 2 meters per second compared with the previous period in 5 consecutive observation periods), stable wind speed (the wind speed fluctuates within ±1 meter per second in 10 consecutive observation periods), and reduced wind speed (the wind speed in each period decreases by more than 1.5 meters per second compared with the previous period in 3 consecutive observation periods) as different events, and the clockwise change of wind direction (for example, the wind direction rotates clockwise by more than 10 degrees each time in 8 consecutive observation periods), counterclockwise change (the wind direction rotates counterclockwise by more than 15 degrees each time in 6 consecutive observation periods) or stable (the wind direction fluctuates within ±5 degrees in 12 consecutive observation periods) as different events. Through such detailed recognition, the system obtains the descriptive distribution labels of several target state trend description events in the target meteorological observation data. For example, in a certain time period, the frequency of the event of continuous increase in wind speed is relatively high (for example, the frequency is 0.4, that is, the wind speed continues to increase in 40% of the observation periods), then its descriptive distribution label will reflect the characteristics of this high frequency; and the wind direction stable event may have a unique descriptive distribution label in another time period. This label can reflect the proportion of the occurrence of the wind direction stable event in the time period relative to other events and other information (for example, the wind direction stable event accounts for 0.6 of all events in the time period, that is, 60%).

[0019] After obtaining these descriptive distribution labels, the meteorological state prediction system can determine the scale coupling evolution weight of at least one set target scale coupling evolution vector in the target meteorological observation data through these labels. Setting a target scale coupling evolution vector is a concept set to accurately describe the evolution process of the meteorological state. For example, for the meteorological element of wind speed, the evolution vectors of wind speed over time at different spatial scales (such as small scale near the ground and medium scale in a larger range) may be set. These vectors can comprehensively reflect the changing trend of wind speed at different scales. The scale coupling evolution weight reflects the relative importance of each scale coupling evolution vector in the evolution process of the entire meteorological state. For example, if the wind speed change at a small scale near the ground has a greater impact on the overall meteorological state in a certain wind farm area, then the weight of the scale coupling evolution vector related to the small scale near the ground will be relatively high (for example, the weight of the wind speed evolution vector at a small scale near the ground is 0.7, while the weight of the wind speed evolution vector at a large scale is 0.3). The determination of this weight is based on the analysis of the descriptive distribution labels of the target state trend description event obtained previously. If a state trend describes an event that occurs frequently at a certain scale and has an important impact on the overall meteorological state, the weight of the scale-coupled evolution vector corresponding to that scale will increase accordingly.

[0020] Next, the meteorological state prediction system obtains the global trend evolution parameter and multi-source coupling evolution discrete index of each set target scale coupling evolution vector. The global trend evolution parameter is a parameter obtained by comprehensively processing the scale coupling evolution weight of each set target scale coupling evolution vector of several different target meteorological observation data. It can reflect the overall evolution trend of the set target scale coupling evolution vector in the entire wind farm area and considering multiple meteorological elements. For example, for the two target meteorological observation data of wind speed and wind direction, under different set target scale coupling evolution vectors, a global trend evolution parameter that can reflect the comprehensive evolution trend of wind speed and wind direction in the entire wind farm is obtained through comprehensive analysis of their scale coupling evolution weights. The multi-source coupling evolution discrete index measures the degree of discreteness of the set target scale coupling evolution vector between different data sources from the perspective of multi-source data. For example, there may be certain differences between the wind speed data observed by the meteorological station and the wind speed data obtained by satellite remote sensing. The multi-source coupling evolution discrete index can reflect the embodiment of this difference in the set target scale coupling evolution vector. If the performance of a certain set target scale coupling evolution vector in the meteorological station data and satellite remote sensing data is significantly different, then its multi-source coupling evolution discrete index will be higher (for example, the average wind speed at a set target scale observed by the meteorological station is 8 m / s, and the average wind speed at the same scale obtained by satellite remote sensing is 10 m / s. This large difference will cause the multi-source coupling evolution discrete index of the set target scale coupling evolution vector to be higher, such as 0.6. If the difference between the two is small, such as the average value observed by the meteorological station is 9 m / s and the average value observed by satellite remote sensing is 9.5 m / s, the multi-source coupling evolution discrete index may be 0.2).

[0021] Afterwards, the meteorological state prediction system determines the state prediction viewpoints and confidence features of each set target scale coupling evolution vector based on the global trend evolution parameters, multi-source coupling evolution discrete indicators and scale coupling evolution weights of each set target scale coupling evolution vector. The state prediction viewpoint is a predictive judgment on the future state of the set target scale coupling evolution vector based on the various parameters and indicators obtained previously. For example, for the set target scale coupling evolution vector of wind speed, according to the global trend evolution parameters, the wind speed has an increasing trend (for example, the global trend evolution parameters indicate that the wind speed has a high probability of increasing in the next few observation periods), the multi-source coupling evolution discrete indicators indicate that the observation difference of wind speed between different data sources is small (for example, the multi-source coupling evolution discrete indicator is 0.1), and the scale coupling evolution weight also indicates that the wind speed evolution at this scale has a greater impact on the overall meteorological state (for example, the scale coupling evolution weight is 0.8), then its state prediction viewpoint may be that the wind speed will continue to increase in the future. The confidence feature is a description of the credibility of this state prediction viewpoint. If the stability of the global trend evolution parameters is high (for example, the global trend evolution parameters fluctuate little in multiple historical periods), the multi-source coupling evolution discrete index is low (indicating good consistency between data sources), and the scale coupling evolution weight is also reasonable, then the confidence feature of this state prediction view will be high (for example, the confidence feature is 0.9, indicating that the credibility of this prediction is high). On the contrary, if there are some uncertainties or contradictions in these parameters and indicators, then the confidence feature will be low (for example, if the global trend evolution parameters fluctuate greatly, the multi-source coupling evolution discrete index is 0.5, and the scale coupling evolution weight is not reasonable, the confidence feature may be 0.4).

[0022] Finally, the meteorological state prediction system determines the meteorological state prediction results of the multi-source wind farm observation data through the state prediction viewpoints of the coupled evolution vectors of each set target scale. For example, considering the state prediction viewpoints of the coupled evolution vectors of the set target scale corresponding to multiple meteorological elements such as wind speed and wind direction, if the wind speed is predicted to increase (for example, the wind speed is predicted to increase from the current 8 m / s to 12 m / s in the next 3 hours), and the wind direction is predicted to be stable (for example, the wind direction fluctuation angle is within ±3 degrees in the next 3 hours), then the meteorological state prediction results will reflect that the wind farm will be in a meteorological state of increasing wind speed and stable wind direction in the future. This prediction result can provide an important reference for the operation and management of the wind farm. In the operation of the wind farm, the accurate prediction of wind speed and wind direction has a vital impact on the power generation efficiency and safety of the wind turbine. If the wind speed is predicted to increase, the wind farm can adjust and maintain the wind turbine in advance to ensure safe and efficient operation under high wind speed conditions; if the wind direction is stable, it will also help optimize the layout and orientation of the wind turbine and other management decisions.

[0023] In addition, during the entire process, the meteorological state prediction system fully utilized the advantages of different data sources when assimilating multi-source wind field observation data. For example, meteorological station data has the characteristics of high precision and high temporal and spatial resolution, and can accurately reflect the meteorological conditions in local areas; wind profile radar data can provide vertical wind speed and direction information, which is very important for understanding the vertical structure of the wind field; satellite remote sensing data can cover a larger range and provide macroscopic meteorological information. Through the comprehensive processing of these multi-source data, the meteorological state prediction system can more comprehensively reflect the wind field information.

[0024] In terms of coupled large eddy models, the meteorological state prediction system performs specific optimizations on them. In the mesoscale-microscale transition scheme, for example, by adjusting some parameter settings in the process of converting the mesoscale model to the microscale model, the model can better adapt to the complex terrain and meteorological conditions of the wind farm. For improvements in the configuration of the physical process parameterization scheme, such as optimizing the parameterization of physical processes such as turbulent exchange and radiation transmission. Taking turbulent exchange as an example, if the turbulent exchange process can be more accurately parameterized, the model's ability to simulate the energy and material exchange within the wind farm can be improved, thereby generating meteorological information with a resolution of 100 meters or even higher. This high-resolution meteorological information is very valuable for the refined management of wind farms. For example, within a wind farm, the wind speed and direction in different areas may vary greatly due to the influence of terrain and wind turbine layout. Meteorological information with a resolution of 100 meters can accurately reflect these differences, which helps wind farms optimize the site selection, layout and operation strategies of wind turbines.

[0025] The meteorological state prediction system mines and analyzes the processed multi-source data and large eddy model output results to build an accurate meteorological state prediction model. In this process, the system will conduct in-depth analysis of various meteorological elements in the multi-source data and large eddy model output results. For example, analyze the correlation between different meteorological elements. For example, there may be a certain correlation between wind speed and temperature (for example, when the temperature rises by 5 degrees, the wind speed may increase by 2-3 meters per second). In some cases, changes in temperature will affect the size of the wind speed; there is also the relationship between humidity and precipitation, etc. By mining these relationships, the system can build a more accurate meteorological state prediction model. This model can not only take into account the changing trends of the meteorological elements themselves, but also take into account the interactions and influences between them, thereby improving the accuracy of meteorological state prediction.

[0026] Therefore, the meteorological state prediction system goes through a series of complex and orderly steps, from the acquisition of multi-source wind farm observation data to the determination of the final meteorological state prediction results, and combines technical means such as multi-source data assimilation and large eddy model optimization. It effectively improves the accuracy of meteorological state prediction under complex conditions of wind farms and provides more reliable meteorological support for the operation and management of wind farms.

[0027] In the above technical solution, the following is a detailed introduction to setting the target scale coupling evolution vector, scale coupling evolution weight, global trend evolution parameter and multi-source coupling evolution discrete index based on the above content.

[0028] 1. Setting the target scale coupling evolution vector

[0029] Setting the target scale coupled evolution vector is a concept used to accurately describe the evolution of meteorological conditions. Taking the wind speed as an example, consider the evolution of wind speed over time at different spatial scales.

[0030] For example, when studying a wind farm area, the spatial scale is divided into a small scale near the ground and a larger mesoscale. For the small scale near the ground, a wind speed evolution vector over time is set. For example, in 10 consecutive time intervals (such as one interval every 10 minutes), the wind speed observation values ​​are [3 m / s, 3.5 m / s, 4 m / s, 4.2 m / s, 4.5 m / s, 4.3 m / s, 4.8 m / s, 5 m / s, 5.2 m / s, 5.5 m / s]. This set of data constitutes an evolution vector of wind speed over time at a small scale near the ground, indicating the changing trend of wind speed at this small scale.

[0031] Similarly, for the mesoscale range, in the same 10 time intervals, the wind speed observations may be [2.5 m / s, 2.8 m / s, 3 m / s, 3.2 m / s, 3.5 m / s, 3.3 m / s, 3.6 m / s, 3.8 m / s, 4 m / s, 4.2 m / s], which is the mesoscale wind speed evolution vector. These wind speed evolution vectors at different scales are an example of the set target scale coupled evolution vector, which comprehensively reflects the changing trend of wind speed at different scales over time, and the set target scale coupled evolution vectors of other meteorological elements such as wind direction can be constructed in a similar way.

[0032] 2. Scale Coupling Evolution Weight

[0033] The scale coupling evolution weight reflects the relative importance of each scale coupling evolution vector in the entire meteorological state evolution process.

[0034] Continuing with the above wind speed example, for example, after analyzing the meteorological state of the wind farm, it is found that the change of wind speed at a small scale near the ground has a greater impact on the overall meteorological state. For example, according to the previous identification and related analysis of the state trend description event of the target meteorological observation data, the change of wind speed at a small scale near the ground is more closely related to key indicators such as the power generation efficiency and stability of the wind turbines in the wind farm. In this case, the weight of the scale coupling evolution vector related to the small scale near the ground will be relatively high. For example, the weight of the wind speed evolution vector at a small scale near the ground is set to 0.7, while the weight of the wind speed evolution vector at a larger scale is set to 0.3. This means that when comprehensively considering the evolution of the meteorological state, the evolution of wind speed at a small scale near the ground occupies a more important position in the overall evaluation, and the determination of the weight is based on the descriptive distribution label analysis of the target state trend description event. For example, if the wind speed at a small scale near the ground continues to increase within a certain period of time (such as the continuous increase in wind speed within 10 consecutive time intervals mentioned above), the frequency of this event is relatively high (for example, the frequency of occurrence is 0.6), and this wind speed increase has a critical impact on the overall meteorological related impact of the wind farm (such as wind turbine power, stability, etc.), then the weight of the scale coupling evolution vector corresponding to this scale will increase accordingly.

[0035] 3. Global Trend Evolution Parameters

[0036] The global trend evolution parameter is a parameter obtained by comprehensively processing the scale coupling evolution weights of each set target scale coupling evolution vector of several different target meteorological observation data. It reflects the overall evolution trend of the set target scale coupling evolution vector in the entire wind farm area and considering multiple meteorological elements.

[0037] For example, consider the two target meteorological observation data of wind speed and wind direction. For wind speed, there are set target scale coupling evolution vectors at the near-ground small scale and medium scale, and their scale coupling evolution weights are 0.7 and 0.3 respectively (as in the above example); for wind direction, there are also set target scale coupling evolution vectors at different scales, such as 0.6 at the small scale and 0.4 at the medium scale. For example, the evolution vectors of wind speed and wind direction at different scales and their weights are combined to consider the evolution trend of the meteorological state of the entire wind farm area.

[0038] If the wind speed shows a rapid increase trend at a small scale (such as the evolution vector of the wind speed at a small scale mentioned above), and the wind direction has a slight clockwise rotation trend at a small scale (for example, in 10 time intervals, the clockwise rotation angles of the wind direction are [2 degrees, 3 degrees, 4 degrees, 3 degrees, 5 degrees, 4 degrees, 6 degrees, 5 degrees, 7 degrees, 6 degrees]), the weights of wind speed and wind direction and their evolution at their respective scales are comprehensively considered to obtain a global trend evolution parameter that can reflect the comprehensive evolution trend of wind speed and wind direction within the entire wind farm. This parameter can be understood as a comprehensive description, which takes into account different meteorological elements, different scales and their respective importance, and uses an overall value or feature to represent the trend of the meteorological state of the entire wind farm developing in a certain direction. For example, this global trend evolution parameter may indicate that the overall meteorological state of the wind farm is developing in the direction of increasing wind speed and clockwise rotation of wind direction.

[0039] 4. Multi-source coupling evolution discrete index

[0040] The multi-source coupling evolution discrete index measures the discrete degree of the coupling evolution vector of the set target scale between different data sources from the perspective of multi-source data.

[0041] Take the observation of wind speed from two data sources, meteorological station and satellite remote sensing, as an example. For example, the wind speed values ​​at a set target scale (such as a small scale near the ground) observed by the meteorological station at 10 observation times are [8 m / s, 8.5 m / s, 9 m / s, 9.2 m / s, 9.5 m / s, 9.3 m / s, 9.8 m / s, 10 m / s, 10.2 m / s, 10.5 m / s], while the wind speed values ​​at the same scale obtained by satellite remote sensing at the same 10 observation times are [9 m / s, 9.2 m / s, 9.5 m / s, 9.8 m / s, 10 m / s, 10.2 m / s, 10.5 m / s, 10.8 m / s, 11 m / s, 11.2 m / s].

[0042] It can be seen that there is a certain difference between the two. The embodiment of this difference in the set target scale coupling evolution vector is calculated to obtain the multi-source coupling evolution discrete index. For example, the variance statistic of the difference between the observation values ​​of the two data sources can be calculated. If this difference is relatively large, it will lead to a higher multi-source coupling evolution discrete index of the set target scale coupling evolution vector, such as 0.6. If the wind speed values ​​observed by the weather station are [9 m / s, 9.1 m / s, 9.2 m / s, 9.3 m / s, 9.4 m / s, 9.5 m / s, 9.6 m / s, 9.7 m / s, 9.8 m / s, 9.9 m / s], and the values ​​obtained by satellite remote sensing are [9.2 m / s, 9.3 m / s, 9.4 m / s, 9.5 m / s, 9.6 m / s, 9.7 m / s, 9.8 m / s, 9.9 m / s, 10 m / s, 10.1 m / s], the difference between the two is small, and the multi-source coupling evolution discrete index may be 0.2. This index reflects the consistency of different data sources for the same set target scale coupling evolution vector, which is of great significance for evaluating the accuracy and reliability of meteorological state forecasts.

[0043] In summary, the embodiment of the present application can fully mine the information in the data and obtain detailed descriptive distribution labels by acquiring multi-source wind farm observation data and performing state trend description event identification of target meteorological observation data, laying the foundation for subsequent analysis. Determining the scale coupling evolution weights based on these labels helps to accurately grasp the degree of influence of meteorological elements on the overall meteorological state at different scales. Obtaining global trend evolution parameters and multi-source coupling evolution discrete indicators can comprehensively consider the overall characteristics of various meteorological elements and multi-source data, and improve the comprehensiveness of meteorological state analysis. Based on these parameters, the state prediction viewpoint and confidence characteristics are determined, so that the prediction results have both a clear direction and a credibility assessment. The final meteorological state prediction results are more accurate and can provide reliable support for wind farm operation management, such as optimizing wind turbine layout, improving power generation efficiency, and ensuring safe operation of wind turbines.

[0044] In some preferred embodiments, the method of obtaining the global trend evolution parameters and multi-source coupling evolution discrete indicators of each set target scale coupling evolution vector includes: performing physical process quantitative mapping on the target meteorological observation data in the multi-source wind farm observation data to obtain the target physical process quantitative characteristics corresponding to the multi-source wind farm observation data; and obtaining the global trend evolution parameters and multi-source coupling evolution discrete indicators corresponding to each set target scale coupling evolution vector through the target physical process quantitative characteristics.

[0045] On the basis of this preferred embodiment, the method also includes: obtaining a number of multi-source historical observation data, and performing state trend description event identification on the historical meteorological observation data included in each multi-source historical observation data to obtain historical description distribution labels of a number of historical state trend description events in each historical meteorological observation data; determining the historical scale coupling evolution weight of at least one set target scale coupling evolution vector in each historical meteorological observation data based on the historical description distribution labels of the several historical state trend description events in each historical meteorological observation data; performing physical process quantitative mapping on each historical meteorological observation data to obtain the historical physical process quantitative characteristics of each multi-source historical observation data; determining the state evolution thermodynamic relationship network corresponding to each set target scale coupling evolution vector in the same historical physical process quantitative characteristics through the historical scale coupling evolution weights of each set target scale coupling evolution vector in the historical meteorological observation data with the same historical physical process quantitative characteristics.

[0046] In the process of forecasting the meteorological state of a wind farm, this embodiment first focuses on obtaining the global trend evolution parameters and multi-source coupled evolution discrete indicators of each set target scale coupled evolution vector. The first key step is to perform physical process quantitative mapping of the target meteorological observation data in the multi-source wind farm observation data. For example, if the target meteorological observation data includes wind speed and wind direction data, for wind speed, its physical process quantitative mapping may involve converting wind speed values ​​at different times and different spatial positions into a quantifiable feature according to certain physical rules. For example, at different observation points in a certain wind farm, the wind speed has different values ​​at different times. By considering the relationship between wind speed and physical factors such as air density and the influence of topography on airflow, the wind speed observation value is converted into a quantitative feature that can reflect the essence of its physical process. For wind direction, it is similar. It is necessary to consider the influence of factors such as the rotation of the earth and the pressure gradient force on the wind direction, so as to obtain the quantitative characteristics of the target physical process corresponding to the multi-source wind farm observation data. This target physical process quantitative feature is an important basis for the subsequent acquisition of global trend evolution parameters and multi-source coupled evolution discrete indicators.

[0047] When obtaining the global trend evolution parameters and multi-source coupled evolution discrete indicators corresponding to each set target scale coupled evolution vector through the quantitative characteristics of the target physical process, taking wind speed as an example, the wind speed evolution vectors at different scales are comprehensively analyzed by integrating multi-source data under the framework of the quantitative characteristics of the target physical process. For example, the quantitative characteristics of the near-ground small-scale wind speed in a specific time period are obtained from the meteorological station data, and the quantitative characteristics of the large-scale wind speed in the same time period are obtained from the satellite remote sensing data. The global trend evolution parameters are determined by comprehensively considering the wind speed quantitative characteristics of these different scales and different data sources. This parameter reflects the comprehensive evolution trend of wind speed at different scales in the entire wind farm area. For the multi-source coupled evolution discrete indicators, the degree of dispersion of wind speed in the set target scale coupled evolution vector is analyzed from different data sources on the basis of the quantitative characteristics of the target physical process. For example, the quantitative characteristics of wind speed at a certain scale observed by the meteorological station show a relatively stable but slightly fluctuating state for a period of time, while the quantitative characteristics of wind speed at the same scale obtained by satellite remote sensing have a large fluctuation range and are somewhat different from the meteorological station data. This difference will be reflected in the multi-source coupled evolution discrete indicators.

[0048] On the basis of this preferred embodiment, the technical solution also includes a series of related operations. Acquiring several multi-source historical observation data is an important step for further in-depth analysis. These multi-source historical observation data contain rich historical meteorological observation data. The state trend description event identification is performed on the historical meteorological observation data included in each multi-source historical observation data, and this process is similar to the operation on the current target meteorological observation data. For example, for historical wind speed data, the continuous increase in wind speed (such as in multiple consecutive historical observation periods, the wind speed increases more significantly each time), the stable wind speed (over a long period of time, the wind speed fluctuates very little), the wind speed decrease, etc. are also identified as different events; for historical wind direction data, the clockwise rotation, counterclockwise rotation, stability, etc. of the wind direction are identified as different events, thereby obtaining the historical description distribution labels of several historical state trend description events in each historical meteorological observation data.

[0049] According to the historical description distribution labels of several historical state trend description events in each historical meteorological observation data, the historical scale coupling evolution weight of at least one set target scale coupling evolution vector in each historical meteorological observation data is determined. Taking the historical wind speed data as an example, if in a certain historical period, a state trend description event of wind speed at a small scale near the ground (such as a continuous increase in wind speed) occurs frequently, and the data such as the wind farm operation records at that time show that the change of wind speed at such a small scale has an important impact on the meteorological state of the entire wind farm, then the historical scale coupling evolution weight of the set target scale coupling evolution vector related to the small scale near the ground in this historical period will be higher. This weight determination process fully considers the occurrence frequency of various state trend description events in the historical data and the degree of influence on the overall meteorological state.

[0050] The physical process quantitative mapping of each historical meteorological observation data is similar to the physical process quantitative mapping of the target meteorological observation data in the current multi-source wind farm observation data, but it is for historical data. By considering the influence of various physical factors in the historical period, such as the topography and atmospheric circulation characteristics at that time on meteorological elements, the historical meteorological observation data are converted into the quantitative characteristics of historical physical processes. For example, in a certain climate period in the past, the area where a certain wind farm is located was affected by a special atmospheric circulation. When performing the physical process quantitative mapping of wind speed, it is necessary to consider the influence of this special circulation on wind speed, so as to obtain accurate quantitative characteristics of historical physical processes.

[0051] Finally, the state evolution thermodynamic relationship network corresponding to the same historical physical process quantitative characteristics of each set target scale coupling evolution vector is determined by the historical scale coupling evolution weights of each set target scale coupling evolution vector in the historical meteorological observation data of the same historical physical process quantitative characteristics. Taking wind speed and wind direction as an example, for example, under a certain historical physical process quantitative characteristics (such as a specific climate mode), the historical scale coupling evolution weight of wind speed at the near-surface small scale is higher, and the historical scale coupling evolution weight of wind direction at the mesoscale is higher. Then, when constructing the state evolution thermodynamic relationship network, the wind speed and wind direction evolution vectors at these two scales will occupy an important position in the relationship network. This state evolution thermodynamic relationship network reflects the mutual relationship between different set target scale coupling evolution vectors under the specific historical physical process quantitative characteristics. This relationship is determined based on the scale coupling evolution weights in the historical data, which can provide an important reference for the current meteorological state prediction. For example, if it is found in historical data that under a certain state evolution thermal relationship network, the subsequent meteorological conditions tend to develop in the direction of increasing wind speed and clockwise wind direction, then in the current meteorological state forecast, if it is found that the current state evolution thermal relationship network is similar to this historical situation, it can be used as an important reference factor to improve the accuracy of the forecast.

[0052] It can be seen that by quantitatively mapping the physical process of the target meteorological observation data of multi-source wind farm observation data, obtaining global trend evolution parameters and multi-source coupled evolution discrete indicators, we can more deeply explore the physical information in the data and improve the accuracy of meteorological state analysis. Multi-source historical observation data are used for multi-faceted analysis, including state trend description event identification, determination of historical scale coupled evolution weights, construction of state evolution thermal relationship network, etc., making full use of historical data to provide more reference for current meteorological state prediction. These operations help to more comprehensively and accurately grasp the evolution relationship of meteorological elements at different scales, improve the accuracy of meteorological state prediction, and thus provide more reliable meteorological support for the operation and management of wind farms.

[0053] In an alternative embodiment, the state prediction viewpoints of each set target scale coupling evolution vector and the confidence features of each state prediction viewpoint are determined through the global trend evolution parameters, multi-source coupling evolution discrete indicators and scale coupling evolution weights of each set target scale coupling evolution vector, including: embedding the scale coupling evolution weights of each set target scale coupling evolution vector through the global trend evolution parameters and multi-source coupling evolution discrete indicators of each set target scale coupling evolution vector to obtain the scale coupling evolution embedding features of each set target scale coupling evolution vector; obtaining the vortex mode characterization vector corresponding to each set target scale coupling evolution vector; and using each set target scale coupling evolution vector to characterize the vortex mode. The state prediction viewpoint of each set target scale coupling evolution vector is determined by the vortex mode characterization vector corresponding to the target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector; the confidence feature of each state prediction viewpoint is determined by the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector; or, the confidence feature of each state prediction viewpoint is determined according to the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, the global trend evolution parameter of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight.

[0054] In the next step, when the target meteorological observation data is wind field environment sensor monitoring data, the vortex mode characterization vector corresponding to the scale coupling evolution weight of each set target scale coupling evolution vector and the scale coupling evolution embedded feature of each set target scale coupling evolution vector are used to determine the state prediction viewpoint of each set target scale coupling evolution vector, including: when the set target scale coupling evolution vector is a ground scale environment vector or a cumulus scale boundary vector, based on the first vortex mode characterization vector and the scale coupling evolution embedded feature of the set target scale coupling evolution vector, the state prediction viewpoint of the scale coupling evolution embedded feature of the set target scale coupling evolution vector is determined, wherein the first vortex mode characterization vector is used to indicate that the state prediction viewpoint has a first quantitative relationship with the scale coupling evolution embedded feature; when the scale coupling evolution weight of the set target scale coupling evolution vector is a cloud microphysical environment vector, based on the second vortex mode characterization vector and the set The scale-coupled evolution embedded feature of the target scale-coupled evolution vector is used to determine the state prediction viewpoint of the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector, wherein the second vortex mode characterization vector is used to indicate that the state prediction viewpoint has a second quantitative relationship with the scale-coupled evolution embedded feature; when the scale-coupled evolution weight of the set target scale-coupled evolution vector is one of the road surface mode weight, the gravity wave crushing mode weight, the ratio of the first radiation transmission mode weight and the second radiation transmission mode weight, the third radiation transmission mode weight and the heat diffusion mode weight, based on the third vortex mode characterization vector and the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector, the state prediction viewpoint of the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector is determined, wherein the third vortex mode characterization vector is used to indicate the correlation between the state prediction viewpoint and the prediction offset, and the prediction offset is the comparison result of the scale-coupled evolution embedded feature and the state probability statistical graph.

[0055] In this replaceable embodiment, first, the key part of the state prediction viewpoints of each set target scale coupling evolution vector and the confidence characteristics of each state prediction viewpoint are determined through the global trend evolution parameters of each set target scale coupling evolution vector, the multi-source coupling evolution discrete indicators and the scale coupling evolution weights.

[0056] Through the global trend evolution parameters and multi-source coupling evolution discrete indicators of each set target scale coupling evolution vector, the scale coupling evolution weight of each set target scale coupling evolution vector is feature embedded, so as to obtain the scale coupling evolution embedding features of each set target scale coupling evolution vector. For example, taking the meteorological element of wind speed as an example, in the meteorological observation of a certain wind farm, there are two set target scale coupling evolution vectors of wind speed near the ground, small scale and medium scale. The global trend evolution parameter shows that the wind speed has a certain change trend as a whole (for example, the wind speed is generally on the rise over a period of time, and the increase has a certain quantitative value, such as an average increase of 2 meters per second per hour), the multi-source coupling evolution discrete indicator reflects the discrete degree of wind speed observation between different data sources (such as meteorological stations and satellite remote sensing data) (for example, the difference between the wind speed values ​​observed by meteorological stations and the wind speed values ​​obtained by satellite remote sensing in certain periods is between 1-3 meters per second), and the scale coupling evolution weight indicates that the small scale wind speed near the ground has a greater impact on the overall meteorological state (for example, the weight is 0.7), and the medium scale wind speed has a relatively small impact (the weight is 0.3). Based on these numerical conditions, feature embedding operations are performed to obtain the respective scale coupling evolution embedding features. This embedding feature is a more in-depth feature representation that integrates the global trend, multi-source discreteness and its own weight.

[0057] Obtain the vortex mode characterization vector corresponding to each set target scale coupling evolution vector. For different set target scale coupling evolution vectors, the vortex mode characterization vector has different meanings and numerical characteristics. For example, for certain scale coupling evolution vectors that reflect the rotation characteristics of airflow in a wind farm, the vortex mode characterization vector may include quantitative characteristics such as the speed and direction of airflow rotation and the interaction with the surrounding environment. For example, in a specific area of ​​a wind farm, the vortex mode characterization vector shows that the rotation speed of the airflow at a certain height is a certain value (such as 3 rotations per minute), the rotation direction is clockwise, and there is a specific interaction relationship with the surrounding environment (such as terrain, interference from other airflows, etc.).

[0058] The state prediction viewpoint of each set target scale coupling evolution vector is determined by using the corresponding vortex mode characterization vector of each set target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector. When the target meteorological observation data is wind field environment sensor monitoring data, the situation is more specific. When the set target scale coupling evolution vector is a ground scale environment vector or a cumulus scale boundary vector, the state prediction viewpoint of the scale coupling evolution embedding feature of the set target scale coupling evolution vector is determined based on the first vortex mode characterization vector and the scale coupling evolution embedding feature of the set target scale coupling evolution vector. Taking the ground scale environment vector as an example, for example, the ground scale environment vector includes the influence of factors such as ground roughness and surface temperature on the wind field. The first vortex pattern characterization vector indicates that the state prediction viewpoint has a first quantitative relationship with the scale-coupled evolution embedded feature. For example, the first vortex pattern characterization vector shows a specific quantitative relationship related to ground roughness and surface temperature (such as a certain quantitative correlation between the wind speed change trend on the ground scale and the vortex pattern for every 5 degrees increase in surface temperature). Combined with the scale-coupled evolution embedded feature (including the previously mentioned global trend, multi-source discreteness, and weights, etc.), the state prediction viewpoint is determined. For example, the state prediction viewpoint may be that in the future, due to the increase in ground temperature and the unchanged ground roughness, the wind speed on the ground scale will increase by a certain value (such as an increase of 3 meters / second).

[0059] When the scale-coupled evolution weight of the target scale-coupled evolution vector is set as the cloud microphysical environment vector, the state prediction viewpoint of the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector is determined based on the second vortex mode characterization vector and the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector. The cloud microphysical environment vector may involve the impact of cloud type, cloud droplet size distribution, etc. on the wind field. For example, the second vortex mode characterization vector reflects the quantitative relationship between the cloud droplet size distribution and the vertical movement of the airflow in the wind field (such as the quantitative correlation between the change in vertical airflow velocity and the vortex mode for every increase of 1 micron in the average radius of cloud droplets). Combined with the scale-coupled evolution embedded feature (which contains comprehensive information on the cloud microphysical environment vector in terms of global trends, multi-source discreteness and weights), the state prediction viewpoint is determined. For example, the state prediction viewpoint may be that due to the change in cloud droplet size, the vertical airflow will be enhanced, thereby affecting the overall stability of the wind field.

[0060] When the scale coupling evolution weight of the target scale coupling evolution vector is set to be one of the road surface mode weight, the gravity wave breaking mode weight, the ratio of the first radiation transmission mode weight and the second radiation transmission mode weight, the third radiation transmission mode weight and the heat diffusion mode weight, the state prediction viewpoint of the scale coupling evolution embedded feature of the set target scale coupling evolution vector is determined based on the third vortex mode characterization vector and the scale coupling evolution embedded feature of the set target scale coupling evolution vector. Taking the road surface mode weight as an example, factors such as the roughness and temperature of the road surface will affect the wind field. The third vortex mode characterization vector is used to indicate the correlation between the state prediction viewpoint and the prediction offset, and the prediction offset is the comparison result of the scale coupling evolution embedded feature and the state probability statistical map. For example, the scale coupling evolution embedded feature shows that the road surface temperature is high and the roughness is large, and the state probability statistical map shows that the wind speed has a greater probability of decreasing in this case. The third vortex mode characterization vector represents an association relationship (such as the quantitative association between the probability of wind speed reduction and the vortex mode), thereby determining the state prediction viewpoint, such as predicting that the wind speed will decrease by a certain value (such as a decrease of 2 meters / second).

[0061] In terms of determining the confidence features of each state prediction viewpoint, it is also based on the above-mentioned various vectors and features. The confidence features of each state prediction viewpoint are determined by the corresponding vortex mode characterization vector of each set target scale coupling evolution vector and the scale coupling evolution embedded feature of each set target scale coupling evolution vector, or the confidence features of each state prediction viewpoint are determined according to the corresponding vortex mode characterization vector of each set target scale coupling evolution vector, the global trend evolution parameter of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight. For example, for the state prediction viewpoint of wind speed at the ground scale mentioned above (the wind speed will increase by 3 meters / second), if the relationship between the airflow rotation and the ground environment shown by the vortex mode characterization vector is very stable (such as the rotation speed fluctuation is very small), the multi-source coupling evolution discrete index in the scale coupling evolution embedded feature is low (indicating good consistency between data sources), the global trend evolution parameter also shows that the wind speed increase trend is obvious, and the scale coupling evolution weight is reasonable (the ground scale environment vector weight is reasonable), then the confidence feature of this state prediction viewpoint will be very high (such as the confidence is 0.9). On the contrary, if there are contradictions or uncertainties between these factors (such as the vortex mode characterization vector shows that the airflow rotation is unstable, the multi-source coupling evolution discrete index is high, and the global trend evolution parameters fluctuate greatly, etc.), the confidence feature will be low (such as a confidence level of 0.3).

[0062] In this way, the state prediction viewpoint and confidence features are determined through complex and orderly operations. The feature embedding operation integrates multiple key parameters to obtain a deeper feature representation. The introduction of the vortex mode characterization vector provides a basis for determining the state prediction viewpoint that is closely related to the meteorological physical process, especially the special treatment of the coupled evolution vectors of different types of set target scales, which makes the state prediction viewpoint more in line with the actual meteorological conditions. When determining the confidence features, multiple factors are considered, so that the prediction results not only have clear views, but also accurate credibility assessments, which greatly improves the accuracy and reliability of meteorological state predictions as a whole, and provides strong support for meteorological analysis and decision-making related to wind farms.

[0063] Under one technical idea, the confidence features of each state prediction viewpoint are determined through the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedded feature of each set target scale coupling evolution vector, including: obtaining a state probability statistical graph vector; performing full connection processing on the state probability statistical graph vector through the scale coupling evolution embedded feature of each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, so as to obtain the fully connected feature value corresponding to each set target scale coupling evolution vector; determining the confidence features corresponding to each set target scale coupling evolution vector through the fully connected feature value corresponding to each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector.

[0064] Under another technical idea, the confidence features of each state prediction viewpoint are determined based on the corresponding vortex mode characterization vector of each set target scale coupling evolution vector, the global trend evolution parameter of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight, including: determining the state heat map corresponding to each set target scale coupling evolution vector through the global trend evolution parameter of each set target scale coupling evolution vector and the multi-source coupling evolution discrete index; performing full connection processing on the state heat map according to the scale coupling evolution weight of each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector to obtain the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector; determining the confidence features corresponding to each set target scale coupling evolution vector through the corresponding fully connected eigenvalues ​​of each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector.

[0065] In this technical solution, there are two technical ideas to determine the confidence features corresponding to each set target scale coupling evolution vector.

[0066] First, the first technical idea is described. Under this idea, the confidence features of each state prediction viewpoint are determined by the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector.

[0067] This process first requires obtaining the state probability statistical graph vector. The state probability statistical graph vector is a quantitative representation of the distribution of various possibilities of meteorological states, which covers the probability information of meteorological elements in different value ranges or states. For example, for the meteorological element of wind speed, the state probability statistical graph vector may contain the probability of wind speed occurring in different speed ranges (such as 0-5 m / s, 5-10 m / s, etc.). In the meteorological observation of a certain wind farm, after comprehensive analysis of historical data and current multi-source observation data, it may be concluded that the probability of wind speed in the range of 0-5 m / s is 0.3, the probability in the range of 5-10 m / s is 0.5, the probability in the range above 10 m / s is 0.2, and similar situations. These probability information constitute the state probability statistical graph vector of wind speed.

[0068] Then, the state probability statistical graph vector is fully connected through the scale coupling evolution embedded features of each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, and the fully connected eigenvalue corresponding to each set target scale coupling evolution vector is obtained. Full connection processing is a calculation process that comprehensively considers multiple factors. For example, taking the set target scale coupling evolution vector corresponding to wind speed as an example, the scale coupling evolution embedded features may contain information such as the correlation between wind speed and other meteorological elements (such as wind direction, temperature, etc.) at different spatial scales (such as near-ground small scale and mesoscale). The vortex mode characterization vector may reflect information such as the airflow rotation mode related to wind speed. When the state probability statistical graph vector of wind speed is fully connected, the scale correlation information in the scale coupling evolution embedded features, the airflow rotation mode information in the vortex mode characterization vector, and the wind speed probability distribution information in the state probability statistical graph vector are comprehensively considered. For example, in some application scenarios, the fully connected eigenvalue of the set target scale coupling evolution vector corresponding to wind speed is 0.7.

[0069] Finally, the confidence features corresponding to the coupled evolution vectors of the set target scales are determined through the fully connected eigenvalues ​​corresponding to the coupled evolution vectors of the set target scales and the vortex mode characterization vectors corresponding to the coupled evolution vectors of the set target scales. Continuing with the wind speed as an example, if the vortex mode characterization vector shows that the airflow rotation pattern of the wind speed is relatively stable and the fully connected eigenvalue is 0.7, this may indicate that the prediction of the wind speed has a high degree of credibility, thereby determining that the confidence feature of the coupled evolution vector of the set target scale corresponding to the wind speed is a higher value (such as 0.8). This is because the fully connected eigenvalue integrates a variety of information related to the wind speed, and the stability of the vortex mode characterization vector further enhances this credibility.

[0070] Next, the second technical idea is explained. Under this idea, the confidence characteristics of each state prediction viewpoint are determined based on the corresponding vortex mode characterization vector of each set target scale coupling evolution vector, the global trend evolution parameter of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight.

[0071] First, the state heat map corresponding to each set target scale coupling evolution vector is determined by the global trend evolution parameter and multi-source coupling evolution discrete index of each set target scale coupling evolution vector. The global trend evolution parameter reflects the overall evolution trend of the set target scale coupling evolution vector, and the multi-source coupling evolution discrete index reflects the degree of discreteness between multi-source data. Taking wind direction as an example, the global trend evolution parameter may indicate that the wind direction has a clockwise rotation trend over a period of time, and the rotation amplitude has a certain quantitative value (such as an average of 5 degrees clockwise rotation per hour). The multi-source coupling evolution discrete index may show the difference in wind direction observations from different data sources (such as meteorological stations and satellite remote sensing data). For example, the average deviation angle between the wind direction observed by the meteorological station and the wind direction obtained by satellite remote sensing in certain periods is 10 degrees. This degree of difference is reflected in the multi-source coupling evolution discrete index. Based on this information, a state heat map is constructed. The state heat map is a graphical way to intuitively represent the distribution of wind direction states and their trends. Different areas in the map may represent different wind direction ranges. Color depth or other marks can represent information such as the probability or trend strength of the wind direction in the area.

[0072] Then, according to the scale coupling evolution weights of each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, the state heat map is fully connected to obtain the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector. Taking the set target scale coupling evolution vector corresponding to wind direction as an example, the scale coupling evolution weight may indicate the degree of influence of wind direction on the overall meteorological state at different spatial scales. For example, the scale coupling evolution weight of wind direction at a small scale near the ground is 0.6, which means that the wind direction at a small scale near the ground has a relatively large influence on the overall meteorological state. The vortex mode characterization vector may contain information about the relationship between wind direction and factors such as airflow rotation and terrain. When the state heat map of wind direction is fully connected, the scale influence information in the scale coupling evolution weight, the wind direction correlation information in the vortex mode characterization vector, and the wind direction state distribution and trend information in the state heat map are comprehensively considered. For example, after calculation, the fully connected eigenvalue of the set target scale coupling evolution vector corresponding to wind direction is 0.6.

[0073] Finally, the confidence features corresponding to each set target scale coupled evolution vector are determined through the corresponding fully connected eigenvalues ​​of each set target scale coupled evolution vector and the vortex mode characterization vectors corresponding to each set target scale coupled evolution vector. For example, if the vortex mode characterization vector shows that the relationship between wind direction and surrounding environmental factors is relatively complex, and the fully connected eigenvalue is 0.6, this may indicate that the prediction credibility of wind direction is at a medium level, thereby determining that the confidence feature of the set target scale coupled evolution vector corresponding to the wind direction is about 0.6. This is because the fully connected eigenvalue integrates a variety of information related to wind direction, and the complex relationship reflected by the vortex mode characterization vector affects the credibility of the prediction to a certain extent.

[0074] In this way, the confidence features are determined through two different technical approaches, providing a more comprehensive and flexible way to evaluate the credibility of meteorological state forecasts. Whether it is the first approach based on factors such as scale coupling evolution embedding features, or the second approach combined with global trend evolution parameters, both fully consider the characteristics of meteorological elements in different aspects through multi-step complex calculations and information fusion. This helps to improve the accuracy and reliability of meteorological state forecast results and provide a more scientific decision-making basis for the operation and management of wind farms.

[0075] Based on any one of the above two technical ideas, the confidence features of each set target scale coupling evolution vector are determined through the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector and the vortex mode characterization vectors corresponding to each set target scale coupling evolution vector, including: when the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the first vortex mode characterization vector, the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector; when the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the second vortex mode characterization vector, the difference between the first set value and the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector; when the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the third vortex mode characterization vector, the weighted result of the second set value and the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector.

[0076] In the case of any of the above two technical ideas, the core part of determining the confidence characteristics of each set target scale coupling evolution vector through the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector and the vortex mode characterization vector corresponding to each set target scale coupling evolution vector needs to be elaborated in detail from many aspects.

[0077] First, let's look at the situation when the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the first vortex mode characterization vector. In the meteorological state prediction system, under this setting, the fully connected eigenvalue is directly determined as the confidence feature corresponding to the set target scale coupling evolution vector. In order to understand this process more clearly, the set target scale coupling evolution vector related to wind speed is used as an example for detailed explanation.

[0078] For example, in the meteorological observation data of a wind farm, after a series of preliminary processing and analysis, the fully connected eigenvalue related to wind speed was obtained. The fully connected eigenvalue is a value that comprehensively reflects the characteristics of wind speed in multiple aspects. Its formation involves the fusion of multiple information in many previous steps. For example, the evolution of wind speed at different spatial scales, such as the specific values ​​of wind speed at small scales near the ground in different time periods. For example, in 10 consecutive observation periods, the wind speeds were [3 m / s, 3.2 m / s, 3.5 m / s, 3.3 m / s, 3.6 m / s, 3.8 m / s, 4 m / s, 4.2 m / s, 4.5 m / s, 4.8 m / s], which reflect the fluctuations of small-scale wind speeds near the ground. There are also similar observation values ​​of wind speed at the mesoscale, such as [2.5 m / s, 2.8 m / s, 3 m / s, 3.2 m / s, 3.5 m / s, 3.3 m / s, 3.6 m / s, 3.8 m / s, 4 m / s, 4.2 m / s]. In addition to wind speed values ​​at different scales, the fully connected eigenvalue also incorporates information from multiple sources, such as wind speed data directly observed by weather stations and wind speed data obtained by satellite remote sensing. Weather station data may provide more local but more accurate wind speed information due to its high precision and high spatiotemporal resolution, while satellite remote sensing data covers a larger range, and there are certain differences and connections between the two. After comprehensive analysis of these data, a comprehensive fully connected eigenvalue is obtained, for example, this value is 0.8.

[0079] When the vortex mode characterization vector of the set target scale coupled evolution vector corresponding to the wind speed is the first vortex mode characterization vector, the confidence feature corresponding to this wind speed vector is directly determined to be 0.8. Behind this result is a profound meteorological significance and data logic relationship. Under this specific vortex mode characterization, the fully connected eigenvalue can fully represent the confidence feature, which means that under the comprehensive consideration of various factors affecting wind speed, including the wind speed evolution trend at different scales, the consistency of multi-source data, etc., these factors are highly consistent with the comprehensive situation represented by the fully connected eigenvalue. From a meteorological point of view, this may mean that under the current meteorological environment, the change trend of wind speed is relatively stable and is subject to fewer interference factors, so its prediction has a higher credibility. For example, if the geographical location of the wind farm is in a relatively flat and open area, and there are no obvious terrain undulations or large obstacles around to interfere with the airflow, then the change of wind speed is relatively easier to predict, and the comprehensive information represented by the fully connected eigenvalue can well reflect this relatively simple wind speed change pattern, so that the confidence feature is 0.8, indicating that the prediction related to wind speed has a high credibility.

[0080] Next, when the vortex mode characterization vector corresponding to the target scale coupling evolution vector is set to the second vortex mode characterization vector, the situation becomes more complicated. At this time, the difference between the first set value and the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector. In order to deeply understand this mechanism, the target scale coupling evolution vector is set by the wind direction in the wind field to illustrate.

[0081] For example, in the analysis of wind direction, the first set value is set to 1. This first set value is not set arbitrarily. It may be a standard reference value related to wind direction obtained based on long-term meteorological research, theoretical models, or statistical analysis of a large amount of historical data. This standard reference value represents the comprehensive value of wind direction-related factors under a certain ideal or standard state. For example, under an ideal atmospheric circulation model, when there are no interfering factors such as local terrain and thermal differences, the wind direction may change in a relatively stable pattern. This 1 may be a quantitative representation of the comprehensive wind direction-related factors under this ideal state.

[0082] For example, the fully connected eigenvalue related to wind direction is 0.6. The process of obtaining the fully connected eigenvalue also involves the observation data of wind direction at different scales and the comprehensive analysis of multi-source data. For example, in the observation of wind direction at different altitudes, from near the ground to the higher atmosphere, the wind direction may have different change patterns at different altitudes. At the same time, the observation data of wind direction from different data sources (such as weather station wind vanes, Doppler radars, etc.) will also be comprehensively considered. In this case, the confidence feature is 1-0.6=0.4.

[0083] This calculation method shows that the prediction credibility of wind direction is determined by the difference between the first set value and the fully connected eigenvalue. The greater the difference between the two, the lower the confidence characteristic. This is reasonable in meteorology, because when the fully connected eigenvalue differs greatly from the first set value representing the ideal state, it means that the prediction of wind direction deviates greatly from the ideal or standard situation, which may be affected by some complex local factors, such as terrain undulations, thermal unevenness, etc. For example, if the wind farm is located in a mountainous area, the obstruction of the mountains and the airflow channel effect of the valley will make the wind direction complex and changeable, which is very different from the wind direction pattern under the ideal state, so the gap between the fully connected eigenvalue and the first set value will be large, resulting in a lower confidence characteristic, indicating that the prediction credibility of wind direction is low.

[0084] Finally, when the vortex mode characterization vector corresponding to the target scale coupling evolution vector is set as the third vortex mode characterization vector, the weighted result of the second set value and the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector. The temperature setting target scale coupling evolution vector in the wind field is taken as an example for detailed description.

[0085] For example, the second set value is 0.5, which is also set based on specific meteorological meanings and data. It may be related to certain baseline characteristics or expected characteristics of temperature under specific circumstances. For example, in a specific season, geographical location, and climate background, there is an expected baseline value related to temperature, which takes into account the basic influence of macro factors such as solar radiation and atmospheric circulation on temperature.

[0086] For example, the temperature-related fully connected eigenvalue is 0.7, and the fully connected eigenvalue is also derived based on various temperature observation data and analysis. For example, temperature observations on different surface types (such as grass, sand, water surface, etc.), and temperature changes at different times (day, night), etc. These data come from multiple data sources, including thermometers at ground weather stations, thermal infrared data from satellite remote sensing, etc. According to the weighted calculation (for example, the weight is 1:1), the confidence feature is (0.5+0.7) / 2=0.6.

[0087] This weighted result reflects the comprehensive consideration of the specific factors represented by the second set value and the actual comprehensive situation represented by the fully connected eigenvalue. This confidence feature of 0.6 indicates that the credibility of the temperature forecast is at an intermediate level. This means that in this vortex mode, the temperature forecast is jointly affected by multiple factors (reflected by the fully connected eigenvalue) and specific baseline factors (reflected by the second set value). For example, in a wind farm close to a large body of water, the water body has a regulating effect on the surrounding temperature, but it is also affected by atmospheric circulation and local meteorological conditions, making the temperature forecast neither completely in line with the ideal baseline situation (represented by the second set value) nor completely dominated by actual observation data (represented by the fully connected eigenvalue), but a comprehensive result of the two, so the confidence feature is 0.6, indicating that the credibility of the temperature forecast is at an intermediate level.

[0088] This method of determining the confidence feature based on the characterization vector of different vortex modes is a more detailed quantification of the credibility of the meteorological state prediction results based on the previous technical solutions. The previous technical solutions have involved many aspects of processing meteorological observation data, such as identifying state trend description events of target meteorological observation data in multi-source wind farm observation data, determining scale coupling evolution weights, obtaining global trend evolution parameters and multi-source coupling evolution discrete indicators, etc.

[0089] In the process of identifying state trend description events, various meteorological elements (such as wind speed, wind direction, temperature, etc.) are divided into different events according to their specific performance in the observation data. For example, for wind speed, continuous increase in wind speed (such as wind speed in each period increased by more than 1.5 m / s compared with the previous period in 5 consecutive observation periods), stable wind speed (wind speed fluctuation range within ±1 m / s in 10 observation periods), and decrease in wind speed (wind speed in each period decreased by more than 1 m / s compared with the previous period in 3 consecutive observation periods) are identified as different events; for wind direction, large clockwise rotation of wind direction (in 6 consecutive observation periods, each rotation angle exceeds 15 degrees), large counterclockwise rotation (in 8 consecutive observation periods, each rotation angle exceeds 15 degrees), etc. are identified as different events. Different events include wind direction fluctuation (in 12 observation periods, the wind direction fluctuation angle is within ±5 degrees), etc.; for temperature, rapid temperature rise (in 4 consecutive observation periods, the temperature rises by more than 3 degrees each time), slow temperature rise (in 8 consecutive observation periods, the temperature rises by 1-2 degrees each time), stable temperature (in 10 observation periods, the temperature fluctuation range is within ±1 degree), and temperature drop (in 3 consecutive observation periods, the temperature drops by more than 2 degrees each time). Through this detailed event identification, we can better grasp the changing trends and characteristics of meteorological elements.

[0090] In determining the scale coupling evolution weights, the degree of influence of meteorological elements on the overall meteorological state at different spatial scales is taken into account. For example, for wind speed, changes in wind speed at a small scale near the ground may have a greater impact on the meteorological state of the local area within the wind farm, while changes in mesoscale wind speed over a larger range have an important impact on the macro-meteorological environment of the entire wind farm. Based on the analysis of these degrees of influence, the scale coupling evolution weights of wind speed at different scales are determined. For example, the scale coupling evolution weight of small-scale wind speed near the ground is 0.7, and the scale coupling evolution weight of mesoscale wind speed is 0.3, indicating that the relative importance of small-scale wind speed near the ground in the evolution of the overall meteorological state is higher. Similarly, similar scale coupling evolution weights will be determined for meteorological elements such as wind direction and temperature.

[0091] Obtaining global trend evolution parameters and multi-source coupling evolution discrete indicators is also an important part of the previous technical solution. The global trend evolution parameter is a parameter obtained by comprehensively processing the scale coupling evolution weights of each set target scale coupling evolution vector of several different target meteorological observation data. It reflects the overall evolution trend of the set target scale coupling evolution vector in the entire wind farm area and considering multiple meteorological elements. For example, by comprehensively considering the scale coupling evolution weights of wind speed and wind direction and their evolution trends at different scales, a global trend evolution parameter that can reflect the comprehensive evolution trend of wind speed and wind direction within the entire wind farm is obtained. The multi-source coupling evolution discrete indicator measures the degree of discreteness of the set target scale coupling evolution vector between different data sources from the perspective of multi-source data. For example, there may be certain differences between the wind speed data observed by the meteorological station and the wind speed data obtained by satellite remote sensing. The multi-source coupling evolution discrete indicator can reflect the embodiment of this difference in the set target scale coupling evolution vector. If the average wind speed observed by the meteorological station at a set target scale is 8 m / s, and the average wind speed obtained by satellite remote sensing at the same scale is 10 m / s, this large difference will lead to a higher multi-source coupling evolution discrete index of the coupling evolution vector at the set target scale, such as 0.6; if the difference between the two is small, such as the average value observed by the meteorological station is 9 m / s and the average value obtained by satellite remote sensing is 9.5 m / s, the multi-source coupling evolution discrete index may be 0.2.

[0092] On this basis, the confidence feature is determined by the relationship between different vortex mode characterization vectors and fully connected eigenvalues, further improving the accuracy assessment of meteorological state prediction. Different vortex mode characterization vectors correspond to different meteorological elements or special relationships under meteorological conditions. Through this targeted calculation method, the credibility of meteorological forecasts under different conditions can be more accurately reflected, thereby providing a more comprehensive and accurate evaluation basis for the meteorological state prediction results of wind farms. For example, in the operation and management of wind farms, accurate meteorological forecast credibility assessment is crucial. If the confidence feature of wind speed prediction is high, the wind farm can more confidently adjust the operating parameters of the wind turbine according to the wind speed prediction results, such as the pitch angle of the blades, to improve the power generation efficiency; if the confidence feature of wind direction prediction is low, the wind farm needs to be more cautious about the wind direction prediction results, and may need to add additional observation methods or adopt conservative operation strategies to avoid damage to the wind turbine due to sudden changes in wind direction; for the confidence feature of temperature prediction, it can help wind farms make more reasonable decisions in equipment maintenance and operation safety, such as taking antifreeze measures for equipment in advance when the temperature is low.

[0093] With this design, the credibility assessment of meteorological state forecast is made more accurate by using different methods to determine the confidence features for different vortex mode characterization vectors. Directly using fully connected eigenvalues, calculating difference values ​​or weighted results, etc., which are respectively applicable to different meteorological elements and meteorological states, can better reflect the reliability of forecast results under various circumstances. This helps wind farm operators and other relevant personnel to more accurately judge the credibility of meteorological forecast results, so as to make more reasonable decisions and improve the efficiency and safety of related matters such as wind farm operation management. In addition, this accurate credibility assessment is based on the previous in-depth processing of meteorological observation data in many aspects. The entire technical solution forms a complete system, from the comprehensive use of multi-source data to the detailed analysis of meteorological elements, and then to the final credibility assessment, which comprehensively improves the accuracy and practicality of meteorological state forecasts.

[0094] In some replaceable embodiments, the state trend description event identification of the target meteorological observation data included in the multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data includes: data extraction of the multi-source wind farm observation data to obtain initial target meteorological observation data in the multi-source wind farm observation data; when it is determined that the initial target meteorological observation data meets the optimization conditions, determining the optimization indication characteristics of the initial target meteorological observation data; optimizing the initial target meteorological observation data by using the optimization indication characteristics to obtain optimized target meteorological observation data; and state trend description event identification of the optimized target meteorological observation data to obtain description distribution labels of several target state trend description events.

[0095] In this alternative embodiment, an exemplary introduction to the process of identifying state trend description events for target meteorological observation data included in multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data is as follows.

[0096] The first step is to extract data from multi-source wind farm observation data to obtain the initial target meteorological observation data in the multi-source wind farm observation data. Multi-source wind farm observation data comes from a wide range of sources, including meteorological stations, wind profiler radars, satellite remote sensing and other sources. From these rich data sources, the initial target meteorological observation data extracted covers a variety of meteorological elements, such as wind speed, wind direction, temperature, humidity, etc. Taking wind speed data as an example, hourly wind speed observations may be obtained from meteorological stations, and a larger range of wind speed distribution may be analyzed from satellite remote sensing data. These data from different sources together constitute the wind speed part of the initial target meteorological observation data.

[0097] After obtaining the initial target meteorological observation data, it is necessary to determine the situation when the initial target meteorological observation data meets the optimization condition. This optimization condition is set based on the characteristics of the meteorological data and the subsequent processing requirements. For example, for wind speed data, the optimization condition may be related to the accuracy, completeness and correlation with other meteorological elements of the data. If there are a large number of missing values ​​in the wind speed data within a certain period of time, or its range of variation exceeds the normal meteorological fluctuation range (for example, in a stable meteorological environment, the wind speed suddenly appears to be extremely large or extremely small and the difference with the data of the surrounding time period is too large), then the optimization condition may not be met. If the time series of the wind speed data is complete, its value is within a reasonable meteorological fluctuation range, and there is a reasonable correlation with meteorological elements such as wind direction and temperature in the same period (such as wind speed increases or decreases regularly with the increase of temperature), then it can be determined that the optimization condition is met.

[0098] When it is determined that the initial target meteorological observation data meets the optimization conditions, the optimization indication characteristics of the initial target meteorological observation data are determined. For wind speed data, the optimization indication characteristics may include statistical characteristics such as the mean, standard deviation, and extreme value of wind speed, as well as the correlation coefficient between wind speed and other meteorological elements. For example, in a certain period of time, the mean of wind speed is 5 m / s, the standard deviation is 1 m / s, and the correlation coefficient with temperature is 0.6 (indicating that there is a certain positive correlation between wind speed and temperature). These values ​​constitute the optimization indication characteristics of wind speed data in this period of time. For wind direction data, the optimization indication characteristics may include the dominant direction of wind direction, the frequency of wind direction changes, etc. For example, in a certain area, the dominant direction of wind direction is southeast wind, and the frequency of wind direction changes per hour during the observation period is low. These are the optimization indication characteristics of wind direction data.

[0099] The initial target meteorological observation data is optimized by optimizing the indicator features to obtain the optimized target meteorological observation data. Taking wind speed data as an example, if the standard deviation in the optimized indicator features is too large, it means that the wind speed data fluctuates violently, and there may be some outliers or it may be affected by local interference factors. At this time, the data can be smoothed according to the mean and standard deviation, and those outliers that obviously deviate too much from the mean can be removed to obtain the optimized wind speed data. For wind direction data, if it is found that the frequency of wind direction changes is too high and does not match the normal meteorological pattern, it may be due to factors such as data collection errors or local complex terrain. Then the wind direction data can be corrected according to the dominant direction, so that the wind direction data is more in line with the actual meteorological laws, and the optimized wind direction data is obtained.

[0100] Finally, the optimized target meteorological observation data is subjected to state trend description event recognition to obtain description distribution labels of several target state trend description events. For wind speed data, the wind speed continuously increases, the wind speed is stable, and the wind speed decreases are identified as different events. For example, in the optimized wind speed data, if the wind speed increases by more than 1.5 m / s in each period compared with the previous period in 5 consecutive observation periods, it can be identified as the event of continuous increase in wind speed; if the wind speed fluctuates within ±1 m / s in 10 observation periods, it is identified as a stable wind speed event; if the wind speed decreases by more than 1 m / s in each period compared with the previous period in 3 consecutive observation periods, it is identified as a wind speed decrease event. For wind direction data, the wind direction rotates clockwise (for example, the rotation angle exceeds 15 degrees each time in 6 consecutive observation periods), rotates counterclockwise (the rotation angle exceeds 10 degrees each time in 8 consecutive observation periods), and the wind direction is stable (the wind direction fluctuation angle is within ±5 degrees in 12 observation periods). For temperature data, rapid temperature rise (temperature rises by more than 3 degrees each time in 4 consecutive observation periods), slow temperature rise (temperature rises by 1-2 degrees each time in 8 consecutive observation periods), stable temperature (temperature fluctuation range is within ±1 degree in 10 observation periods), and temperature drop (temperature drops by more than 2 degrees each time in 3 consecutive observation periods) are identified as different events. Through such an identification process, the distribution of each event in the optimized target meteorological observation data is obtained, and these distributions constitute the descriptive distribution labels of several target state trend description events. For example, in a certain period of time, the descriptive distribution label of the event of continuous increase in wind speed may show that the frequency of occurrence of this event is 0.2 (that is, the situation of continuous increase in wind speed occurs in 20% of the observation periods), the descriptive distribution label of the event of stable wind speed may show that its frequency is 0.6, and the descriptive distribution label of the event of reduced wind speed may show that its frequency is 0.2.

[0101] In this way, by optimizing the initial target meteorological observation data, the data quality is improved, making the subsequent state trend description event identification more accurate. The process of determining the optimization conditions, optimizing the indication features and performing the optimization process can effectively remove the outliers and parts that do not conform to the meteorological laws in the data, thereby more accurately identifying the state trend description events of the meteorological elements. This helps to provide a more reliable data foundation in the entire meteorological state prediction system, improve the accuracy of meteorological state prediction, and provide more accurate meteorological information support for the operation and management of wind farms.

[0102] In an extensible technical solution, when the target meteorological observation data is wind farm environment sensor monitoring data, the set target scale coupling evolution vector includes: road surface mode weight, gravity wave breaking mode weight, third radiation transmission mode weight, wind speed attention weight, heat diffusion mode weight, ground scale environment vector, cloud microphysical environment vector, the ratio of the first radiation transmission mode weight to the second radiation transmission mode weight, three-dimensional wind farm mode weight, wind farm noise weight, wind farm remote sensing mode weight and multi-mode wind farm monitoring change weight. At least one of the weights, correspondingly, the scale coupling evolution weight of at least one set target scale coupling evolution vector is determined by describing the distribution labels of the several target state trend description events, including at least one of the following:

[0103] (1) determining a scale-coupled evolution weight of at least one of a road surface mode weight and a gravity wave breaking mode weight based on a plurality of description distribution labels of the first state trend description events;

[0104] (2) determining a scale-coupled evolution weight of at least one of a third radiation transmission mode weight, a wind speed attention weight, a heat diffusion mode weight, and a ground scale environment vector based on the description distribution labels of the plurality of fourth state trend description events;

[0105] (3) determining a scale coupling evolution weight of at least one of a cloud microphysical environment vector and a ratio of a first radiation transfer mode weight to a second radiation transfer mode weight based on the description distribution labels of the plurality of second state trend description events, the description distribution labels of the plurality of third state trend description events, and the description distribution labels of the plurality of fourth state trend description events;

[0106] (4) Determine the scale-coupled evolution weight of the three-dimensional wind field pattern weight based on the description distribution labels of several third-state trend description events;

[0107] (5) Based on the description distribution label of the fifth state trend description event, determine the scale coupling evolution weight of at least one of the wind farm noise weight, the wind farm remote sensing mode weight and the multi-mode wind farm monitoring change weight.

[0108] In this scalable technical solution, when the target meteorological observation data is wind farm environmental sensor monitoring data, the target scale coupling evolution vector is set to include a variety of different weights and vector types, which reflects the various characteristics and influencing factors of the wind farm meteorological state.

[0109] For the road surface mode weight and gravity wave breaking mode weight in the set target scale coupling evolution vector, the scale coupling evolution weight is determined by the description distribution labels of several first state trend description events. Taking the road surface mode weight as an example, some specific state trend description events in the wind farm environmental sensor monitoring data have a key impact on its weight. For example, in the wind farm environmental sensor monitoring data, if the first state trend description event contains the relevant description of the impact of road surface roughness changes on the near-surface wind speed and wind direction of the wind farm. When the description distribution label shows that the road surface roughness increase event occurs frequently in a certain period of time, and the wind speed near the ground associated with it changes greatly (for example, in 10 consecutive observation periods, the near-surface wind speed decreases by an average of 1.5 m / s after each increase in road surface roughness), and the wind direction also changes significantly (such as the average wind direction deviation angle reaches 10 degrees), this indicates that the road surface condition has a greater impact on the wind farm, then the scale coupling evolution weight of the road surface mode weight will increase accordingly. The weight of the gravity wave breaking mode is similar. If the first state trend describes the impact of gravity wave breaking on the vertical structure of the wind field in the event, and the description distribution label shows that the frequency of gravity wave breaking events increases within a specific time period, and the wind speed and temperature gradient in the vertical direction of the wind field have significant changes (such as at each certain height in the vertical direction, the wind speed change and temperature change exceed specific values), this means that the importance of the gravity wave breaking mode in the evolution of the wind field meteorological state has increased, and its scale-coupled evolution weight will also increase.

[0110] For the third radiation transfer mode weight, wind speed attention weight, heat diffusion mode weight and ground scale environment vector, their scale coupling evolution weights are determined based on the description distribution labels of several fourth state trend description events. As for the third radiation transfer mode weight, the fourth state trend description event may involve the relevant situation of radiation transmission and wind farm energy exchange. For example, in the wind farm environmental sensor monitoring data, if the fourth state trend description event contains information about the transmission of solar radiation in the wind farm and the energy conversion with the wind farm. When the description distribution label indicates that the solar radiation enhancement event occurs in a certain period of time (such as the solar radiation intensity increases by 50 watts / square meter each time in 5 consecutive observation periods), the energy distribution in the wind farm has obvious changes (for example, in a specific area, the increase in air temperature has a certain proportional relationship with the increase in radiation intensity, such as the air temperature increases by 2 degrees on average for every 50 watts / square meter increase in radiation intensity), which shows that radiation transmission has a greater impact on the energy state of the wind farm, so the scale coupling evolution weight of the third radiation transfer mode weight will be adjusted accordingly. For the wind speed attention weight, if the fourth state trend description event includes the impact of wind speed changes on the energy and material diffusion of the wind field, when the description distribution label shows that in certain time periods, a sudden increase in wind speed (such as wind speed increases by more than 2 m / s each time in three consecutive observation periods) will cause the material diffusion speed in the wind field to accelerate (for example, the diffusion range of a specific pollutant in the wind field increases in a certain proportion with the increase in wind speed), this indicates that the importance of wind speed in the evolution of wind field states increases, and the scale-coupled evolution weight of the wind speed attention weight will also change accordingly. The same is true for the heat diffusion pattern weight. If the heat diffusion in the wind field in the fourth state trend description event shows that in a certain area, heat diffuses from a high temperature area to a low temperature area, when the description distribution label indicates that in a specific time period, this heat diffusion speed accelerates (such as in four consecutive observation periods, the heat diffusion range increases by 10% each time) and is significantly related to factors such as wind speed and wind direction (such as the greater the wind speed, the more consistent the heat diffusion direction is with the wind direction, and the faster the diffusion speed), then the scale-coupled evolution weight of the heat diffusion pattern weight will be determined based on this situation. For the ground-scale environmental vector, the fourth state trend description event includes the impact of factors such as ground temperature, humidity, and roughness on the wind farm. When the description distribution label shows that in a certain period of time, the ground temperature rises (such as in 6 consecutive observation periods, the ground temperature rises by 1 degree each time), the ground humidity decreases (such as the humidity decreases by 5% each time), and the ground roughness remains unchanged, the wind speed, wind direction, and energy distribution of the wind farm near the ground have corresponding changes (such as the wind speed near the ground decreases and the wind direction has a small deflection), which shows the degree of influence of the ground-scale environment on the meteorological state of the wind farm, thereby determining the scale coupling evolution weight of the ground-scale environmental vector.

[0111] For the cloud microphysical environment vector, the ratio of the first radiation transfer mode weight to the second radiation transfer mode weight is determined based on the description distribution labels of several second state trend description events, the description distribution labels of several third state trend description events, and the description distribution labels of several fourth state trend description events. The cloud microphysical environment vector is closely related to the formation, development, and impact of clouds on the wind field. For example, the second state trend description event may involve changes in cloud types. When the description distribution label shows that the frequency of cumulus clouds increases over a period of time (such as the frequency of cumulus clouds increases from 10% to 30% in 10 observation periods), and the height and thickness of cumulus clouds also change (such as the average height of cumulus clouds increases from 1000 meters to 1500 meters, and the thickness increases from 500 meters to 800 meters), this will affect the vertical structure and water vapor distribution of the wind field. The third state trend description event may be related to the microphysical process in the cloud (such as the formation, growth and falling of water droplets). When the description distribution label shows that the average radius of water droplets in the cloud increases in certain time periods (such as the average radius of water droplets increases from 10 microns to 15 microns in 5 consecutive observation periods), this will affect the optical properties and precipitation potential of the cloud, and then have a feedback effect on the wind field. The fourth state trend description event involves the relationship between radiation transfer and wind field as mentioned above. Changes in clouds will change the radiation transfer characteristics. For example, when the cloud amount increases, the absorption, scattering and reflection of solar radiation change, thereby affecting the energy balance of the wind field. Combining the description distribution labels of these different state trend description events, the scale coupling evolution weight of the cloud microphysical environment vector can be determined. The ratio of the weight of the first radiation transfer mode to the weight of the second radiation transfer mode is also based on the description distribution labels of these state trend description events. For example, the first radiation transfer mode may be related to shortwave radiation, and the second radiation transfer mode may be related to longwave radiation. When cloud changes (describe events through the second and third state trends) and the interaction between radiation and wind fields (describe events through the fourth state trend) affect the transmission and conversion of shortwave radiation and longwave radiation, the scale-coupled evolution weight of the ratio of the first radiation transfer mode weight to the second radiation transfer mode weight is determined based on the relevant information in the distribution description label, such as the change in the energy ratio of shortwave radiation and longwave radiation under different cloud conditions and wind field conditions.

[0112] For the weight of the three-dimensional wind field pattern, its scale-coupled evolution weight is determined based on the description distribution labels of several third-state trend description events. The weight of the three-dimensional wind field pattern reflects the structure and evolution characteristics of the wind field in three-dimensional space. The third-state trend description event may include the changes in wind speed and wind direction in the horizontal and vertical directions of the wind field. For example, when the description distribution label shows that the wind speed shear in the vertical direction of the wind field increases within a certain period of time (such as within the height range of 100-500 meters, the vertical shear of the wind speed increases from 2 meters / second / 100 meters to 3 meters / second / 100 meters), and the horizontal shear of the wind direction in the horizontal direction also changes (such as within a range of 10 kilometers, the horizontal shear angle of the wind direction increases from 10 degrees to 15 degrees), which indicates that the three-dimensional structure of the wind field has changed significantly, and the importance of the three-dimensional wind field pattern in the evolution of the meteorological state of the wind field is highlighted, and its scale-coupled evolution weight will be adjusted accordingly.

[0113] For the wind farm noise weight, wind farm remote sensing mode weight and multi-mode wind farm monitoring change weight, the scale coupling evolution weight is determined based on the description distribution label of the fifth state trend description event. The wind farm noise weight is related to various interference factors in wind farm observation. The fifth state trend description event may include the impact of wind farm sensor failure, external environmental interference, etc. on the accuracy of wind farm observation data. For example, when the description distribution label shows that the frequency of wind farm sensor failures increases within a certain period of time (such as the number of sensor failures increases from 1 to 3 times within 10 observation periods), this will cause the noise in the wind farm observation data to increase. At this time, the scale coupling evolution weight of the wind farm noise weight will increase accordingly to reflect the impact of this interference factor on the analysis of the wind farm meteorological state. For the wind farm remote sensing mode weight, the fifth state trend description event may involve the accuracy and coverage of remote sensing data in wind farm monitoring. When the descriptive distribution label shows that in a certain period of time, the resolution of remote sensing data is improved (such as the spatial resolution is increased from 1 km to 500 meters), the coverage is expanded (such as the coverage area is expanded from 100 square kilometers to 150 square kilometers), and the consistency with other observation data (such as ground observation data) is enhanced (such as the average value of the wind speed difference between the two is reduced from 2 meters / second to 1 meter / second), this shows that the role of wind farm remote sensing mode in wind farm meteorological state monitoring is more important, and the scale coupling evolution weight of wind farm remote sensing mode weight will be adjusted accordingly. For the multi-mode wind farm monitoring change weight, the fifth state trend description event includes the comprehensive situation of wind farm change monitoring by multiple monitoring modes (such as ground observation, remote sensing observation, aerial detection, etc.). When the descriptive distribution labels show that within a certain time period, the synergy of multiple monitoring modes is enhanced (such as the deviation of wind speed and wind direction observation results in the same area of ​​the wind farm by different monitoring modes is reduced), and the ability to capture wind field changes is improved (such as the ability to more promptly detect sudden changes in wind speed and wind direction in the wind farm), this shows that the importance of multi-mode wind farm monitoring in the evolution of the meteorological state of the wind farm has increased, and the scale-coupled evolution weight of the multi-mode wind farm monitoring change weight will be determined based on this situation.

[0114] This design determines the scale coupling evolution weight by associating different set target scale coupling evolution vectors with the descriptive distribution labels of specific state trend description events, which can more accurately reflect the complex characteristics of the wind farm meteorological state. This targeted association takes into account the influence of various physical processes, environmental factors and observation patterns in the wind farm, so that each weight is determined based on solid data and events. This helps to improve the accuracy of wind farm meteorological state prediction and more comprehensively grasp the dynamic changes of wind farms, thereby providing a more reliable basis for wind farm operation management, meteorological research, etc.

[0115] Further, Figure 2 Schematic diagram of a meteorological state prediction system 200 provided in an embodiment of the present application. Figure 2 The meteorological state prediction system 200 shown includes a processor 210, and the processor 210 can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0116] Alternatively, if Figure 2 As shown, the meteorological state prediction system 200 may further include a memory 230. The processor 210 may call and run a computer program from the memory 230 to implement the method in the embodiment of the present application.

[0117] The memory 230 may be a separate device independent of the processor 210 , or may be integrated into the processor 210 .

[0118] Alternatively, if Figure 2 As shown, the weather status prediction system 200 may further include a transceiver 220, and the processor 210 may control the transceiver 220 to interact with other devices, specifically, may send information or data to other devices, or receive information or data sent by other devices.

[0119] Optionally, the meteorological condition prediction system 200 can implement the corresponding processes corresponding to the storage engine or components in the storage engine (such as a processing module) or a device deployed with a storage engine in each method of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0120] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities.

[0121] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. It should be noted that the memory of the systems and methods described herein is intended to include but is not limited to suitable types of memory.

[0122] Based on the above, a readable storage medium is provided, on which a program or instruction is stored, and the program or instruction implements the steps of the above method when executed by a processor.

[0123] It should be noted that, in this article, the term "includes", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the method and device in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0124] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), including a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the embodiment of the present application.

[0125] The embodiments of the embodiments of the present application are described above in conjunction with the accompanying drawings, but the embodiments of the present application are not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the embodiments of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the embodiments of the present application and the scope of protection of the embodiments of the present application, all of which are within the protection of the embodiments of the present application.

Claims

1. A meteorological state prediction method based on big data, characterized in that: The method is implemented by a meteorological state prediction system, and the method comprises: Acquire multi-source wind farm observation data, and perform state trend description event recognition on target meteorological observation data included in the multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data; Determining a scale coupling evolution weight of at least one set target scale coupling evolution vector in the target meteorological observation data through description distribution labels of the plurality of target state trend description events; Obtaining a global trend evolution parameter and a multi-source coupling evolution discrete index of each set target scale coupling evolution vector, wherein the global trend evolution parameter and the multi-source coupling evolution discrete index are obtained by processing the scale coupling evolution weight of each set target scale coupling evolution vector of a plurality of different target meteorological observation data; Determine the state prediction viewpoints of each set target scale coupling evolution vector and the confidence characteristics of each state prediction viewpoint according to the global trend evolution parameters, multi-source coupling evolution discrete indicators and scale coupling evolution weights of each set target scale coupling evolution vector; The meteorological state prediction result of the multi-source wind farm observation data is determined through the state prediction viewpoint of each set target scale coupled evolution vector.

2. The method according to claim 1, characterized in that The step of obtaining the global trend evolution parameters and multi-source coupling evolution discrete indicators of each set target scale coupling evolution vector includes: Performing physical process quantitative mapping on target meteorological observation data in the multi-source wind farm observation data to obtain quantitative characteristics of target physical processes corresponding to the multi-source wind farm observation data; The global trend evolution parameters and multi-source coupling evolution discrete indicators corresponding to the respective set target scale coupling evolution vectors are obtained through the quantitative characteristics of the target physical process.

3. The method as claimed in claim 2, characterized in that The method further comprises: Acquire a number of multi-source historical observation data, and perform status trend description event recognition on the historical meteorological observation data included in each multi-source historical observation data, and obtain historical description distribution labels of a number of historical status trend description events in each historical meteorological observation data; Determining a historical scale coupling evolution weight of at least one set target scale coupling evolution vector in each historical meteorological observation data according to the historical description distribution labels of the plurality of historical state trend description events in each historical meteorological observation data; Performing physical process quantitative mapping on each historical meteorological observation data to obtain the historical physical process quantitative characteristics of each multi-source historical observation data; The state evolution thermodynamic relationship network corresponding to the quantitative characteristics of the same historical physical process of each set target scale coupling evolution vector is determined by the historical scale coupling evolution weights of each set target scale coupling evolution vector in the historical meteorological observation data of the same historical physical process quantitative characteristics.

4. The method according to claim 1, characterized in that The state prediction viewpoints and confidence features of each state prediction viewpoint of each set target scale coupling evolution vector are determined through the global trend evolution parameters, multi-source coupling evolution discrete indicators and scale coupling evolution weights of each set target scale coupling evolution vector, including: By using the global trend evolution parameters and multi-source coupling evolution discrete indicators of the set target scale coupling evolution vectors, feature embedding is performed on the scale coupling evolution weights of the set target scale coupling evolution vectors to obtain the scale coupling evolution embedding features of the set target scale coupling evolution vectors; Obtaining the vortex mode characterization vector corresponding to each set target scale coupling evolution vector; Determine the state prediction viewpoint of each set target scale coupling evolution vector by using the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector; The confidence characteristics of each state prediction viewpoint are determined by the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedded characteristics of each set target scale coupling evolution vector; or, the confidence characteristics of each state prediction viewpoint are determined based on the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, the global trend evolution parameters of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight.

5. The method as claimed in claim 4, characterized in that When the target meteorological observation data is wind field environment sensor monitoring data, the state prediction viewpoint of each set target scale coupling evolution vector is determined by using the vortex mode characterization vector corresponding to the scale coupling evolution weight of each set target scale coupling evolution vector and the scale coupling evolution embedding feature of each set target scale coupling evolution vector, including: When the target scale-coupled evolution vector is set to be a ground-scale environment vector or a cumulus-scale boundary vector, based on a first vortex mode characterization vector and a scale-coupled evolution embedded feature of the set target scale-coupled evolution vector, a state prediction viewpoint of the scale-coupled evolution embedded feature of the set target scale-coupled evolution vector is determined, wherein the first vortex mode characterization vector is used to indicate that the state prediction viewpoint has a first quantitative relationship with the scale-coupled evolution embedded feature; When the scale coupling evolution weight of the set target scale coupling evolution vector is set to a cloud microphysical environment vector, based on a second vortex mode characterization vector and a scale coupling evolution embedded feature of the set target scale coupling evolution vector, a state prediction viewpoint of the scale coupling evolution embedded feature of the set target scale coupling evolution vector is determined, wherein the second vortex mode characterization vector is used to indicate that the state prediction viewpoint has a second quantitative relationship with the scale coupling evolution embedded feature; When the scale coupling evolution weight of the target scale coupling evolution vector is set to be one of the pavement mode weight, the gravity wave breaking mode weight, the ratio of the first radiation transmission mode weight and the second radiation transmission mode weight, the third radiation transmission mode weight and the heat diffusion mode weight, the state prediction viewpoint of the scale coupling evolution embedded feature of the set target scale coupling evolution vector is determined based on the third vortex mode characterization vector and the scale coupling evolution embedded feature of the set target scale coupling evolution vector, wherein the third vortex mode characterization vector is used to indicate the correlation between the state prediction viewpoint and the prediction offset, and the prediction offset is the comparison result of the scale coupling evolution embedded feature and the state probability statistical graph.

6. The method according to claim 4, characterized in that The confidence features of each state prediction viewpoint are determined by using the vortex mode characterization vector corresponding to each set target scale coupling evolution vector and the scale coupling evolution embedding features of each set target scale coupling evolution vector, including: Get the state probability statistics graph vector; By embedding the vortex mode characterization vectors corresponding to the set target scale coupling evolution vectors into the scale coupling evolution features of the set target scale coupling evolution vectors, the state probability statistical graph vector is fully connected to obtain the fully connected eigenvalues ​​corresponding to the set target scale coupling evolution vectors; The confidence features corresponding to the respective set target scale coupling evolution vectors are determined through the fully connected eigenvalues ​​corresponding to the respective set target scale coupling evolution vectors and the vortex mode characterization vectors corresponding to the respective set target scale coupling evolution vectors.

7. The method as claimed in claim 4, characterized in that The confidence features of each state prediction viewpoint are determined according to the vortex mode characterization vector corresponding to each set target scale coupling evolution vector, the global trend evolution parameter of each set target scale coupling evolution vector, the multi-source coupling evolution discrete index and the scale coupling evolution weight, including: Determine the state heat map corresponding to each set target scale coupling evolution vector through the global trend evolution parameter of each set target scale coupling evolution vector and the multi-source coupling evolution discrete index; According to the scale coupling evolution weights of each set target scale coupling evolution vector and the vortex mode characterization vectors corresponding to each set target scale coupling evolution vector, the state heat map is fully connected to obtain the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector; The confidence features corresponding to the respective set target scale coupling evolution vectors are determined through the fully connected eigenvalues ​​corresponding to the respective set target scale coupling evolution vectors and the vortex mode characterization vectors corresponding to the respective set target scale coupling evolution vectors.

8. The method according to claim 6 or 7, characterized in that The step of determining the confidence features of each set target scale coupling evolution vector by using the fully connected eigenvalues ​​corresponding to each set target scale coupling evolution vector and the vortex mode characterization vectors corresponding to each set target scale coupling evolution vector comprises: When the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the first vortex mode characterization vector, the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector; When the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is a second vortex mode characterization vector, determining a difference between the first set value and the fully connected eigenvalue as a confidence feature corresponding to the set target scale coupling evolution vector; When the vortex mode characterization vector corresponding to the set target scale coupling evolution vector is the third vortex mode characterization vector, a weighted result of the second set value and the fully connected eigenvalue is determined as the confidence feature corresponding to the set target scale coupling evolution vector.

9. The method according to any one of claims 1 to 7, characterized in that: The step of performing state trend description event recognition on the target meteorological observation data included in the multi-source wind farm observation data to obtain description distribution labels of several target state trend description events in the target meteorological observation data includes: Extracting data from the multi-source wind farm observation data to obtain initial target meteorological observation data from the multi-source wind farm observation data; When it is determined that the initial target meteorological observation data meets the optimization condition, determining the optimization indication feature of the initial target meteorological observation data; Optimizing the initial target meteorological observation data by using the optimization indication feature to obtain optimized target meteorological observation data; The optimized target meteorological observation data is subjected to state trend description event recognition to obtain description distribution labels of a plurality of target state trend description events.

10. A meteorological state prediction system, characterized in that: The method comprises at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the method according to any one of claims 1 to 9.

Images