Three-dimensional true wind measurement data analysis method and system based on cloud computing

Abstract

The present application relates to the field of data analysis technology, and in particular provides a three-dimensional true wind measurement data analysis method and system based on cloud computing, and proposes a technical idea that combines a mobile small-scale dynamic compensation three-dimensional true wind measurement technology with the data processing capabilities of cloud computing. By introducing advanced dynamic compensation true wind measurement technology, high-precision, real-time measurement of wind speed and direction is achieved. At the same time, with the help of the powerful data processing capabilities of the cloud computing platform, the collected three-dimensional true wind data is efficiently processed, stored and analyzed, thereby providing users with real-time wind speed and direction information. In addition, in response to the shortcomings of traditional technologies in data processing and analysis, it is further proposed to deeply explore the correlation between data by determining the correlation characteristics of wind condition changes and the correlation characteristics of monitoring elements. This not only improves the efficiency and accuracy of data processing, but also enables real-time monitoring and prediction of wind condition changes, thereby meeting the modern complex and changeable wind condition monitoring needs.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a three-dimensional true wind measurement data analysis method and system based on cloud computing. Background Art

[0002] Accurate measurement and real-time analysis of wind speed and direction are crucial in fields such as meteorology, aviation, and navigation. However, traditional true wind measurement technologies are often limited by equipment size, measurement accuracy, and data processing capabilities, making them incapable of meeting the complex and ever-changing needs of modern wind monitoring. This is especially true when processing large-scale, high-frequency data, making traditional technologies inadequate and unable to provide timely and accurate wind information. Summary of the Invention

[0003] In order to improve the above problems, the present application provides a three-dimensional true wind measurement data analysis method and system based on cloud computing.

[0004] The present application provides a cloud computing-based three-dimensional true wind measurement data analysis method, which is applied to a cloud computing analysis system. The method includes:

[0005] Acquire first dynamic compensation true wind monitoring data and second dynamic compensation true wind monitoring data to be analyzed;

[0006] determining, based on an attention tag for each piece of wind condition change information in the first dynamically compensated true wind monitoring data and an attention tag for each piece of wind condition change information in the second dynamically compensated true wind monitoring data, a wind condition change correlation feature between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the wind condition change correlation feature being used to reflect a state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data at a wind condition change level;

[0007] determining, based on a first monitoring element embedding vector corresponding to the first dynamically compensated true wind monitoring data and a second monitoring element embedding vector corresponding to the second dynamically compensated true wind monitoring data, a monitoring element association feature between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the monitoring element association feature is used to reflect a state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data at a monitoring element level;

[0008] The global state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data is determined based on the wind condition change correlation characteristics and the monitoring element correlation characteristics.

[0009] In some design ideas, determining the wind condition change correlation feature between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on the attention tag of each wind condition change information in the first dynamically compensated true wind monitoring data and the attention tag of each wind condition change information in the second dynamically compensated true wind monitoring data includes:

[0010] For each piece of wind condition change information in the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, determining a target attention tag corresponding to the wind condition change information based on each wind direction attention tag of the wind condition change information;

[0011] Based on the target attention tags corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data and the target attention tags corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data, at least one of the wind direction jump trend vector and the meteorological environment impact characteristic value is determined as the wind condition change association feature.

[0012] In some design ideas, the wind direction jump trend vector is determined based on the following steps:

[0013] determining a wind direction change probability distribution based on target attention labels corresponding to respective pieces of wind condition change information in the first dynamic compensation true wind monitoring data and target attention labels corresponding to respective pieces of wind condition change information in the second dynamic compensation true wind monitoring data;

[0014] The wind direction jump trend vector is determined according to the quantitative characteristics of each wind condition change information and the wind direction change probability distribution.

[0015] In some design concepts, the meteorological environment impact characteristic value is determined based on the following steps:

[0016] determining a first environmental disturbance feature based on the attention weights of the target attention tags corresponding to the respective wind condition change information in the first dynamic compensation true wind monitoring data and the attention weights of the target attention tags corresponding to the respective wind condition change information in the second dynamic compensation true wind monitoring data;

[0017] determining a second environmental disturbance feature based on a confidence coefficient of a target attention label corresponding to each piece of wind condition change information in the first dynamic compensation true wind monitoring data and a confidence coefficient of a target attention label corresponding to each piece of wind condition change information in the second dynamic compensation true wind monitoring data;

[0018] Determining a consistency analysis feature between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on target attention tags corresponding to respective pieces of wind condition change information in the first dynamically compensated true wind monitoring data and target attention tags corresponding to respective pieces of wind condition change information in the second dynamically compensated true wind monitoring data; determining an environmental involvement vector based on the consistency analysis feature, a confidence coefficient of the target attention tag corresponding to respective pieces of wind condition change information in the first dynamically compensated true wind monitoring data, and a confidence coefficient of the target attention tag corresponding to respective pieces of wind condition change information in the second dynamically compensated true wind monitoring data;

[0019] The meteorological environment impact characteristic value is determined according to the first environmental disturbance characteristic, the second environmental disturbance characteristic, and the environmental involvement vector.

[0020] In some design concepts, determining the monitoring element association feature of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on the first monitoring element embedding vector corresponding to the first dynamically compensated true wind monitoring data and the second monitoring element embedding vector corresponding to the second dynamically compensated true wind monitoring data includes at least one of the following:

[0021] performing compensation monitoring analysis on the first dynamic compensated true wind monitoring data and the second dynamic compensated true wind monitoring data through a compensation monitoring processing network to obtain each first compensation monitoring data stream in the first dynamic compensated true wind monitoring data and its corresponding compensation monitoring linear variable, as well as each second compensation monitoring data stream in the second dynamic compensated true wind monitoring data and its corresponding compensation monitoring linear variable; determining a compensation monitoring association characteristic value based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable and the second compensation monitoring data stream and its corresponding compensation monitoring linear variable as the monitoring element association feature;

[0022] Through the monitoring data mining model, monitoring vector mining is performed on the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data to obtain the first true wind measurement data spatiotemporal linkage vector corresponding to the first dynamically compensated true wind monitoring data, and the second true wind measurement data spatiotemporal linkage vector corresponding to the second dynamically compensated true wind monitoring data; based on the first true wind measurement data spatiotemporal linkage vector and the second true wind measurement data spatiotemporal linkage vector, the spatiotemporal monitoring decision characteristic value is determined as the monitoring element association feature.

[0023] In some design ideas, determining the compensation monitoring associated characteristic value based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable and the second compensation monitoring data stream and its corresponding compensation monitoring linear variable includes:

[0024] Using each of the first compensation monitoring data streams and their corresponding compensation monitoring linear variables, and each of the second compensation monitoring data streams and their corresponding compensation monitoring linear variables, a plurality of initial compensation monitoring data stream tuples are generated; the initial compensation monitoring data stream tuple includes one of the first compensation monitoring data stream and one of the second compensation monitoring data streams, and the compensation monitoring linear variables corresponding to the first compensation monitoring data stream and the second compensation monitoring data stream belonging to the same initial compensation monitoring data stream tuple meet a first true wind monitoring scenario matching threshold;

[0025] Determining a target compensation monitoring data stream binary from the multiple initial compensation monitoring data stream binary groups; the compensation monitoring linear variables corresponding to the first compensation monitoring data stream and the second compensation monitoring data stream in the target compensation monitoring data stream binary group meet a second true wind monitoring scenario matching threshold;

[0026] The compensation monitoring association characteristic value is determined by using the number of the first compensation monitoring data stream and the second compensation monitoring data stream included in each target compensation monitoring data stream tuple, as well as the total number of the first compensation monitoring data stream in the first dynamic compensation true wind monitoring data and the total number of the second compensation monitoring data stream in the second dynamic compensation true wind monitoring data.

[0027] In some design ideas, the first true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data, and the second true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data; the determining of the spatiotemporal monitoring decision characteristic value based on the first true wind measurement data spatiotemporal linkage vector and the second true wind measurement data spatiotemporal linkage vector includes:

[0028] For each wind condition change information binary group, based on the multidimensional wind field environment vectors corresponding to the two wind condition change information pieces in the wind condition change information binary group, a wind condition change disturbance error corresponding to the wind condition change information binary group is determined; the wind condition change information binary group includes one piece of wind condition change information belonging to the first dynamically compensated true wind monitoring data and one piece of wind condition change information belonging to the second dynamically compensated true wind monitoring data, and the two pieces of wind condition change information belonging to the same wind condition change information binary group correspond to the same distribution label in the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data;

[0029] The spatiotemporal monitoring decision feature value is determined by utilizing the wind condition change disturbance error corresponding to each of the wind condition change information binary groups.

[0030] In some design ideas, when the wind condition change association characteristics include a wind direction jump trend vector and a meteorological environment impact characteristic value, and the monitoring element association characteristics include a compensation monitoring association characteristic value and a spatiotemporal monitoring decision characteristic value, determining the global state prediction correlation between the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data based on the wind condition change association characteristics and the monitoring element association characteristics includes:

[0031] When the wind direction jump trend vector belongs to the first quantized trend range, determining that the global state prediction correlation is non-correlation;

[0032] When the spatiotemporal monitoring decision characteristic value is less than the set spatiotemporal monitoring characteristic value, determining that the global state prediction correlation is correlated;

[0033] When the wind direction jump trend vector belongs to the second quantitative trend range, if the meteorological environment impact characteristic value is not greater than the set impact characteristic value, or the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then it is determined that the global state prediction correlation is not associated; otherwise, it is determined that the global state prediction correlation is associated; the wind direction jump amplitude included in the second quantitative trend range is greater than the wind direction jump amplitude included in the first quantitative trend range;

[0034] When the wind direction jump trend vector belongs to the third quantized trend range, if the meteorological environment impact characteristic value is not greater than the set impact characteristic value, and the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then it is determined that the global state prediction correlation is not associated; otherwise, it is determined that the global state prediction correlation is associated; the wind direction jump amplitude included in the third quantized trend range is greater than the wind direction jump amplitude included in the second quantized trend range;

[0035] When the wind direction jump trend vector belongs to a fourth quantized trend range, the global state prediction correlation is determined to be correlated; and the wind direction jump amplitude included in the fourth quantized trend range is greater than the wind direction jump amplitude included in the third quantized trend range.

[0036] In some design ideas, obtaining the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data to be analyzed includes:

[0037] Acquire first true wind monitoring data to be processed and second true wind monitoring data to be processed;

[0038] Determining one of the first true wind monitoring data to be processed and the second true wind monitoring data to be processed as dynamic compensation true wind monitoring reference data, and determining the other true wind monitoring data to be processed as dynamic compensation true wind monitoring data to be optimized;

[0039] Optimizing the dynamic compensation type true wind monitoring data to be optimized based on the dynamic compensation type true wind monitoring reference data to obtain dynamic compensation type true wind monitoring optimized data;

[0040] The dynamically compensated true wind monitoring reference data and the dynamically compensated true wind monitoring optimization data are used as the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data.

[0041] Under some independent design ideas, the dynamic compensation type true wind monitoring data to be optimized is optimized based on the dynamic compensation type true wind monitoring reference data to obtain the dynamic compensation type true wind monitoring optimization data, including:

[0042] When the feature fine granularity of the dynamically compensated true wind monitoring data to be optimized is different from that of the dynamically compensated true wind monitoring reference data, adjusting the feature fine granularity of the dynamically compensated true wind monitoring data to be optimized, and updating the feature fine granularity of the dynamically compensated true wind monitoring data to be the feature fine granularity of the dynamically compensated true wind monitoring reference data;

[0043] When the measurement standards of the dynamically compensated true wind monitoring data to be optimized are different from those of the dynamically compensated true wind monitoring reference data, adjusting the measurement standards of the dynamically compensated true wind monitoring data to be optimized by using a measurement standard adjustment rule, and updating the measurement standards of the dynamically compensated true wind monitoring data to be optimized to the measurement standards of the dynamically compensated true wind monitoring reference data;

[0044] Based on the dynamically compensated true wind monitoring data to be optimized and the dynamically compensated true wind monitoring reference data having the same feature granularity and the same measurement standard, a data optimization indication feature is determined; the wind condition change information of the dynamically compensated true wind monitoring data to be optimized is optimized using the data optimization indication feature to obtain the dynamically compensated true wind monitoring optimized data.

[0045] Under some independent design ideas, determining the data optimization indication feature based on the dynamically compensated true wind monitoring data to be optimized and the dynamically compensated true wind monitoring reference data with the same feature granularity and the same measurement standard includes:

[0046] performing compensation monitoring analysis on the dynamic compensated true wind monitoring reference data and the dynamic compensated true wind monitoring data to be optimized through a compensation monitoring processing network to obtain each third compensation monitoring data stream in the dynamic compensated true wind monitoring reference data and its corresponding compensation monitoring linear variable, as well as each fourth compensation monitoring data stream in the dynamic compensated true wind monitoring data to be optimized and its corresponding compensation monitoring linear variable;

[0047] Determining a target compensation monitoring data stream tuple using each of the third compensation monitoring data streams and their corresponding compensation monitoring linear variables, and each of the fourth compensation monitoring data streams and their corresponding compensation monitoring linear variables; the target compensation monitoring data stream tuple includes one of the third compensation monitoring data streams and one of the fourth compensation monitoring data streams, and the compensation monitoring linear variables corresponding to the third compensation monitoring data streams and the fourth compensation monitoring data streams belonging to the same target compensation monitoring data stream tuple satisfy a set association relationship;

[0048] The data optimization indication feature is determined based on the third compensation monitoring data stream and the fourth compensation monitoring data stream included in the target compensation monitoring data stream tuple.

[0049] An embodiment of the present application provides a cloud computing analysis 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 performs the above-mentioned method.

[0050] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above method when executed.

[0051] In the embodiment of the present application, a technical idea is proposed that combines a mobile small-scale dynamic compensation three-dimensional true wind measurement technology with the data processing capabilities of cloud computing. By introducing advanced dynamic compensation true wind measurement technology, high-precision, real-time measurement of wind speed and direction is achieved. At the same time, with the help of the powerful data processing capabilities of the cloud computing platform, the collected three-dimensional true wind data is efficiently processed, stored and analyzed, thereby providing users with real-time wind speed and direction information. In addition, in response to the shortcomings of traditional technologies in data processing and analysis, it is further proposed to deeply explore the correlation between data by determining the correlation characteristics of wind condition changes and the correlation characteristics of monitoring elements. This not only improves the efficiency and accuracy of data processing, but also can monitor and predict wind condition changes in real time, provide important decision-making support for related industries, and thus meet the modern complex and changeable wind condition monitoring needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flowchart of a three-dimensional true wind measurement data analysis method based on cloud computing provided in an embodiment of the present application.

[0053] Figure 2 A schematic diagram of the structure of a cloud computing analysis system 200 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to better understand the above technical solution, the technical solution of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0055] Figure 1 A three-dimensional true wind measurement data analysis method based on cloud computing is shown, which is applied to a cloud computing analysis system. The method includes the following steps 110 to 140.

[0056] Before describing the following steps 110 to 140 in detail, a complete application scenario will be used for illustration.

[0057] In one application scenario, the cloud computing analysis system has been connected to multiple dynamic compensation true wind monitoring stations across the country, which monitor wind conditions in real time and provide valuable meteorological data for various industries.

[0058] The cloud computing analysis system first obtains first and second dynamically compensated true wind data from two different dynamically compensated true wind monitoring stations via a network connection. The first and second dynamically compensated true wind data include wind speed, wind direction, and various other meteorological parameters. All data is updated in real time, ensuring the timeliness and accuracy of the analysis.

[0059] The cloud computing analysis system then conducted an in-depth analysis of the two sets of dynamically compensated true wind monitoring data (the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data). The cloud computing analysis system first focused on the wind condition change information in each set of data. The wind condition change information was assigned an attention label so that the system could quickly identify the key change points of the wind conditions. The cloud computing analysis system compared these labels in the two sets of dynamically compensated true wind monitoring data and determined the correlation characteristics of the two sets of dynamically compensated true wind monitoring data at the wind condition change level through the corresponding algorithm. The correlation characteristics at the wind condition change level revealed the similarities and differences between the two sets of dynamically compensated true wind monitoring data in terms of wind speed and wind direction changes, providing an important basis for subsequent predictions.

[0060] Next, the cloud computing analysis system further processed the two sets of dynamically compensated true wind monitoring data and extracted the respective monitoring element embedding vectors. These monitoring element embedding vectors not only include basic meteorological parameters such as wind speed and direction, but also incorporate various information such as time and geographic location, forming a multi-dimensional data representation. Using these embedding vectors, the cloud computing analysis system determined the correlation characteristics of the two sets of dynamically compensated true wind monitoring data at the monitoring element level. This correlation characteristic at the monitoring element level reflects the inherent connection between the different monitoring elements of the two sets of dynamically compensated true wind monitoring data, providing rich data support for global state prediction.

[0061] After obtaining the correlation characteristics of wind condition changes and monitoring element correlation characteristics, the cloud computing analysis system begins a comprehensive assessment. Using advanced machine learning algorithms, it integrates these correlation characteristics and determines the global state prediction correlation between the two sets of dynamically compensated true wind monitoring data. This global state prediction correlation not only considers real-time changes in wind conditions but also the interactions between various monitoring elements, enabling a more comprehensive prediction of future wind conditions.

[0062] Through this precise and efficient data processing process, the cloud computing analysis system provides valuable information support for fields such as meteorology, aviation, and navigation. Whether it's short-term wind forecasts or long-term climate change trend analysis, this system provides accurate and timely data support, helping various industries make more informed decisions.

[0063] In combination with the above application scenarios, steps 110 to 140 are respectively introduced and explained below.

[0064] Step 110 : Acquire first dynamic compensation true wind monitoring data and second dynamic compensation true wind monitoring data to be analyzed.

[0065] In an embodiment of the present application, the first dynamically compensated true wind monitoring data refers to the data obtained from the first dynamically compensated true wind monitoring station connected to the cloud computing analysis system. This data is collected by a mobile small-scale dynamically compensated three-dimensional true wind measurement technology, which can measure wind speed and wind direction with high precision and in real time. In addition to basic wind speed and wind direction information, the first dynamically compensated true wind monitoring data can also include other meteorological parameters, such as temperature, humidity, etc. These data are updated in real time to ensure their timeliness and accuracy. Such data are of great significance to many fields such as weather forecasting, aviation safety, and marine navigation. The second dynamically compensated true wind monitoring data refers to the data obtained from the second dynamically compensated true wind monitoring station connected to the cloud computing analysis system. Similar to the first dynamically compensated true wind monitoring data, it is also obtained by dynamically compensated true wind measurement technology, which can capture subtle changes in wind conditions in real time and provide high-precision wind speed and wind direction information. These data also contain multiple meteorological parameters and are transmitted to the cloud computing analysis system in real time via the network. The second dynamic compensation true wind monitoring data cooperates with the first dynamic compensation true wind monitoring data to provide the cloud computing analysis system with more comprehensive wind condition information, thereby improving the accuracy and reliability of wind condition prediction.

[0066] In step 110, the cloud computing analysis system actively obtains data from two different dynamic compensation true wind monitoring sites through a network connection. The two sites can be located in different geographical locations, so the data they provide can reflect the changes in wind conditions in different areas. The system first confirms a stable connection with the two sites, and then sends a data request. After receiving the request, the monitoring site will transmit the latest true wind monitoring data it has collected to the cloud computing analysis system. These data include the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, which are updated in real time to ensure the timeliness and accuracy of the analysis. Once the data is successfully transmitted to the cloud computing analysis system, the system will perform preliminary data verification and collation to prepare for subsequent in-depth analysis. The successful completion of this step is the basis of the entire data processing process. It ensures that the cloud computing analysis system can obtain the latest and most accurate wind data, thereby providing a solid data foundation for subsequent wind forecasts and decision support.

[0067] Step 120: Based on the attention tags of each wind condition change information in the first dynamically compensated true wind monitoring data and the attention tags of each wind condition change information in the second dynamically compensated true wind monitoring data, determine the wind condition change correlation characteristics between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the wind condition change correlation characteristics are used to reflect the state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data at the wind condition change level.

[0068] In the embodiments of this application, attention tags are a data tagging method used to highlight or mark key information in the data. In cloud computing analysis systems, attention tags specifically refer to the tags assigned to wind change information. These tags help the system quickly identify and focus on changes in key meteorological parameters such as wind speed and direction. By adding attention tags to data, the system can more effectively extract and analyze real-time changes in wind conditions, thereby improving the accuracy of forecasts.

[0069] Wind condition correlation features are extracted by comparing and analyzing wind condition variation information from two or more sets of real wind monitoring data, reflecting the similarities and differences between these data in terms of wind speed, wind direction, and other wind condition variations. These features are derived through in-depth analysis of attention-tagged data using a cloud computing analysis system, providing a critical basis for subsequent wind condition forecasting.

[0070] State prediction correlation refers to the mutual correlation between two or more sets of real wind monitoring data when predicting future wind conditions. This correlation takes into account real-time changes in wind conditions and the mutual influence of various monitoring factors, and serves as a critical reference for cloud computing analysis systems in making global state predictions. By comprehensively assessing the state prediction correlations between different data sets, the system can more comprehensively predict future wind conditions, providing support for decision-making across various industries.

[0071] In step 120, the cloud computing analysis system begins to deeply analyze wind change information within the first and second dynamically compensated true wind monitoring data. To efficiently identify and analyze key change points within this data, the system utilizes attention tags. These tags are pre-applied to the data to highlight significant wind changes, such as sudden increases or decreases in wind speed or sharp shifts in wind direction. The system first examines the attention tags within each data set; these tags indicate the data segments that require special attention. Next, the cloud computing analysis system leverages its powerful computing power to compare and analyze the wind change information within the two data sets that bear the attention tags. This process examines multiple dimensions, including the trend, magnitude, and frequency of wind speed and direction changes. Through complex algorithmic processing, the system identifies correlation features between the two data sets at the wind condition change level. These features reveal not only similarities in wind speed and direction changes between the two data sets, such as the presence of synchronized increases or decreases, but also differences, such as more dramatic wind speed changes at a particular monitoring point during a specific time period. These wind change correlation features are crucial for subsequent wind forecasting. These data provide a comprehensive perspective for the cloud computing analysis system, enabling insights into the inherent connections and patterns in wind conditions across different monitoring sites. This in-depth analysis enables the system to more accurately predict future wind conditions, providing more reliable information support for fields such as meteorology, aviation, and navigation.

[0072] Step 130: Based on the first monitoring element embedding vector corresponding to the first dynamically compensated true wind monitoring data and the second monitoring element embedding vector corresponding to the second dynamically compensated true wind monitoring data, determine the monitoring element association characteristics of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the monitoring element association characteristics are used to reflect the state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data at the monitoring element level.

[0073] In an embodiment of the present application, the first monitoring element embedding vector refers to a high-dimensional data representation extracted from the first dynamic compensation true wind monitoring data. The embedding vector is a fusion of multiple elements (such as wind speed, wind direction, temperature, humidity, etc.) and other related information (such as timestamp, geographic location, etc.) in the original monitoring data into a vector through a specific data conversion technology. This embedding vector not only contains all the key information of the original data, but also makes the similarity and correlation between the data easier to calculate and compare in the form of a vector. The first monitoring element embedding vector specifically refers to this embedding vector extracted from the first group of true wind monitoring data. Similar to the first monitoring element embedding vector, the second monitoring element embedding vector is a high-dimensional data representation extracted from the second dynamic compensation true wind monitoring data. It also fuses all the key monitoring elements and other related information of the group of data to form a comprehensive data description. Through the second monitoring element embedding vector, the correlation and similarity between the second group of true wind monitoring data and other data can be easily analyzed and compared. The monitoring element association feature refers to the feature extracted by analyzing the relationship between the first monitoring element embedding vector and the second monitoring element embedding vector. These features reflect the inherent connections and similarities between the two sets of real wind monitoring data at the monitoring element level (such as wind speed, wind direction, temperature, and humidity). Monitoring element correlation features not only consider the correlation between individual monitoring elements but also integrate the mutual influence of multiple elements, thereby more comprehensively describing the state prediction correlation between the two sets of data. These features are of great significance for subsequent wind forecasting and decision support.

[0074] In step 130, the cloud computing analysis system aims to deeply explore the correlation between the first and second dynamically compensated true wind monitoring data at the monitoring element level. To achieve this goal, the system first utilizes advanced data conversion techniques to extract the first and second monitoring element embedding vectors from the two data sets. These embedding vectors are comprehensive, high-dimensional representations of the original monitoring data, incorporating information on multiple key monitoring elements such as wind speed, wind direction, temperature, and humidity, as well as other important data such as timestamps and geographic location. This form of data representation enables the system to more easily analyze and compare the inherent connections between the two data sets. Next, the cloud computing analysis system utilizes complex algorithms to compare and analyze the relationship between the first and second monitoring element embedding vectors. The system focuses on the similarities, differences, and changing trends between the two data sets at the monitoring element level, thereby extracting monitoring element correlation features. These correlation features reveal the inherent connections and mutual influences between the two data sets regarding various monitoring elements such as wind speed and wind direction. For example, the system can discover that two sets of data exhibit similar patterns in wind speed and direction changes under specific meteorological conditions, or that temperature changes at one monitoring point significantly influence humidity changes at another. Through this in-depth analysis, the cloud computing analysis system can more comprehensively understand the correlation between the two sets of real wind monitoring data at the monitoring element level. This provides strong data support for subsequent wind forecasts and decision support, enabling the system to more accurately predict future weather conditions and help various industries make more informed decisions.

[0075] Step 140: Determine the global state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on the wind condition change correlation characteristics and the monitoring element correlation characteristics.

[0076] In an embodiment of the present application, the global state prediction correlation is a comprehensive evaluation index, which considers the relationship and influence of two or more groups of dynamic compensation true wind monitoring data on the overall wind state and monitoring elements. This correlation is obtained by comprehensively analyzing the correlation characteristics of wind condition changes and the correlation characteristics of monitoring elements, and aims to comprehensively and accurately reflect the global changes and trends of wind conditions between different monitoring sites. The determination of the global state prediction correlation is of great significance for improving the accuracy of wind condition forecasts, realizing cross-regional meteorological monitoring and early warning, and optimizing resource allocation. It not only takes into account real-time wind condition changes, but also integrates the inherent connections between multiple monitoring elements, so that it can more scientifically predict future meteorological conditions and provide decision makers with comprehensive and reliable data support.

[0077] In step 140, the cloud computing analysis system conducts a comprehensive analysis of the previously extracted wind condition correlation features and monitoring element correlation features to determine the global state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data. First, the system reviews and integrates the wind condition correlation features, which reveal similarities and differences between the two data sets in terms of wind speed, wind direction, and other wind condition variations. By comparing these features, the system can identify consistency or differences in wind condition variation trends between the two data sets, providing important clues for global state prediction. Next, the system combines these features with the monitoring element correlation features. These features reflect the inherent connections between the two data sets regarding monitoring elements (such as temperature, humidity, and air pressure). By comprehensively considering the mutual influence of these elements, the system can more comprehensively assess the correlation between the two data sets. After integrating the wind condition correlation features and monitoring element correlation features, the cloud computing analysis system uses advanced algorithmic models to deeply integrate and analyze these features. Through this process, the system can determine the global state predictive correlation between the two data sets, namely, the comprehensive similarity and influence between the overall wind conditions and monitoring elements. Finally, the system presents this global state prediction correlation in a visual or numerical form for reference by decision makers. The determination of this correlation not only helps to improve the accuracy of wind forecasts, but also provides valuable information support for decision makers in meteorology, aviation, navigation and other fields, helping them to make more informed and timely decisions. It is worth mentioning that the above-mentioned wind condition change correlation characteristics, the first monitoring element embedding vector, the second monitoring element embedding vector and the monitoring element correlation characteristics involve the core inventive ideas of the embodiments of this application. Based on this, these four features / vectors are further introduced and explained below.

[0078] 1) Wind change correlation features are extracted by comparing and analyzing real wind data from different monitoring stations, reflecting the similarities and differences in wind changes. These features help predict global wind change trends.

[0079] For example, consider data from two monitoring stations, Station A and Station B. By analyzing their wind speed and direction data, we find that as the wind speed at Station A increases, the wind speed at Station B also increases, and the wind direction changes remain consistent between the two. This correlation can be extracted as a wind change correlation feature. In meteorological forecasting, wind change correlation features can help predict how wind conditions in one area affect another, thereby providing more accurate weather forecasts and disaster warnings.

[0080] 2) The first monitoring factor embedding vector is a high-dimensional data representation extracted from the first set of real wind monitoring data. It integrates information from multiple monitoring factors such as wind speed, wind direction, temperature, and humidity. This embedding vector makes it easier to calculate similarities and associations between data.

[0081] For example, the embedding vector of the first monitoring factor is [0.8, 0.3, -0.2, 0.5]. This four-dimensional vector may represent the fused information of monitoring factors such as wind speed, wind direction, temperature, and humidity. Each dimension corresponds to the characteristics of a specific monitoring factor. In a meteorological monitoring network, the embedding vector of the first monitoring factor can be used to quickly compare data from different monitoring sites, thereby promptly detecting abnormal weather conditions or making wind forecasts.

[0082] 3) Similar to the first monitoring factor embedding vector, the second monitoring factor embedding vector is a high-dimensional data representation extracted from the second set of real wind monitoring data. It also integrates information from multiple monitoring factors for a comprehensive description and comparison of the data.

[0083] If the second monitoring factor embedding vector is [0.7, 0.4, -0.1, 0.6], it similarly represents the feature fusion of another set of monitoring data. In meteorological data analysis, by comparing the first and second monitoring factor embedding vectors, we can analyze the similarities and differences between meteorological data from different regions or time points, thereby revealing the patterns and trends of wind changes.

[0084] 4) Monitoring element correlation features are extracted by analyzing the relationship between the embedding vectors of the first and second monitoring elements. They reflect the intrinsic connection between the two sets of real wind monitoring data at the monitoring element level. These features help to fully understand the complexity of wind changes and the mutual influence of multiple factors.

[0085] When analyzing the two sets of monitoring data, it was found that when the temperature in the first set increased, the humidity in the second set also showed an upward trend. This correlation between temperature and humidity can be extracted as a monitoring factor correlation feature. In meteorological forecasting and climate change research, monitoring factor correlation features can help scientists more accurately understand the interactions between various meteorological factors, thereby improving the accuracy and reliability of prediction models. These features can also help develop more effective strategies to respond to climate change and extreme weather events.

[0086] As can be seen, the embodiments of this application significantly improve the accuracy and real-time performance of wind speed and direction measurements by integrating small, dynamic, and compensated 3D true wind measurement technology with cloud computing data processing capabilities. This not only enables efficient processing and storage of large amounts of 3D true wind data, but also enables in-depth analysis of this data via a cloud computing platform, providing users with real-time wind speed and direction information. This innovative combination not only optimizes the data processing process and efficiency, but also ensures the accuracy of data analysis.

[0087] It's worth noting that this embodiment further explores the deep-seated correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data by determining the correlation characteristics of wind condition changes and monitoring element correlation characteristics. This correlation analysis not only facilitates real-time monitoring of wind condition changes but also provides strong support for future wind condition forecasting.

[0088] In summary, the embodiments of the present application not only increase the degree of dependence of various industries on wind condition information, but also improve the utilization rate of wind condition information through accurate data analysis, providing important decision-making basis for meteorology, aviation, navigation and other fields.

[0089] In some optional embodiments, the wind condition change association characteristics of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data are determined based on the attention tags of each wind condition change information in the first dynamically compensated true wind monitoring data and the attention tags of each wind condition change information in the second dynamically compensated true wind monitoring data, including: for each wind condition change information in the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data, based on each wind direction attention tag of the wind condition change information, determining the target attention tag corresponding to the wind condition change information; based on the target attention tag corresponding to each wind condition change information in the first dynamically compensated true wind monitoring data and the target attention tag corresponding to each wind condition change information in the second dynamically compensated true wind monitoring data, determining at least one of a wind direction jump trend vector and a meteorological environment impact characteristic value as the wind condition change association characteristic.

[0090] In this embodiment, the cloud computing analysis system, as the core execution entity, will deeply analyze how to determine the wind condition change correlation characteristics between the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data based on the attention tags of the wind condition change information.

[0091] First, the cloud computing analysis system processes each wind change in the first and second dynamically compensated true wind monitoring data. This information may include key data points such as sudden changes in wind speed and wind direction. Each data point is accompanied by a corresponding attention tag, which indicates the data's importance and change characteristics.

[0092] During processing, the cloud computing analysis system pays special attention to the wind direction attention tags associated with each wind change. These tags are derived from a comprehensive analysis of historical and real-time data and accurately reflect the changing trends of wind direction. The system uses specific algorithms, such as weighted averaging or pattern recognition, to determine the target attention tag associated with each wind change. These target attention tags serve as the basis for further analysis.

[0093] Next, the cloud computing analysis system determines at least one of a wind direction jump trend vector and a meteorological environmental impact characteristic value based on the target attention tags corresponding to each wind change information in the first and second dynamically compensated true wind monitoring data. The wind direction jump trend vector reflects the similarities and differences in wind direction changes between the two sets of data and is an important indicator for assessing the correlation of wind condition changes. The meteorological environmental impact characteristic value, on the other hand, considers the combined effects of more external factors, such as temperature and humidity, on wind condition changes.

[0094] Specifically, the cloud computing analysis system identifies wind direction jump trends by comparing the target attention tags corresponding to the same or similar wind change information in two sets of data. If the target attention tags corresponding to the same wind change information in the two sets of data show consistent or similar change trends, the system can determine that there is a correlation between the wind direction jumps between the two sets of data. At the same time, the system also combines other meteorological environmental data, such as temperature and humidity, to calculate the meteorological environmental impact characteristic value to more comprehensively evaluate the correlation of wind condition changes.

[0095] Through the above steps, the cloud computing analysis system can accurately determine the wind condition change correlation characteristics between the first and second dynamically compensated true wind monitoring data. These characteristics not only include the jump trend of wind direction but also take into account the comprehensive influence of the meteorological environment. Therefore, they can more comprehensively and accurately reflect the correlation between the two sets of data in terms of wind condition change state prediction.

[0096] Through the technical solution of this embodiment, the cloud computing analysis system enables high-precision, real-time measurement and data analysis of wind speed and direction, further enhancing the reliance and utilization of wind information across various industries. By identifying correlated features of wind changes, the system not only improves the efficiency and accuracy of data processing but also enables real-time monitoring and prediction of wind changes, providing important decision-making support for fields such as meteorology, aviation, and navigation. This solution, which combines advanced true wind measurement technology with the big data processing capabilities of cloud computing, significantly enhances the ability of various industries to cope with complex wind conditions.

[0097] In the next step, the wind direction jump trend vector is determined based on the following steps: based on the target attention label corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data and the target attention label corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data, the wind direction change probability distribution is determined; according to the quantitative characteristics of each wind condition change information and the wind direction change probability distribution, the wind direction jump trend vector is determined.

[0098] Based on this embodiment, determining the wind direction jump trend vector is a key step. This embodiment will elaborate on how the cloud computing analysis system determines the wind direction jump trend vector based on the target attention tags in the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data.

[0099] First, the cloud computing analysis system processes two sets of real wind monitoring data: the first dynamically compensated real wind monitoring data and the second dynamically compensated real wind monitoring data. Both sets contain multiple wind condition variation information, each of which is assigned a corresponding target attention label. These labels are indicators of data importance and are automatically assigned by the system based on the data's characteristics and variation patterns.

[0100] Next, the cloud computing analysis system analyzes the target attention labels corresponding to each wind change in the two data sets. The system compares these labels to identify similarities and differences in wind direction changes between the two data sets. By comparing the distribution and intensity of these labels, the system constructs a probability distribution model for wind direction changes. This model reflects the probability of wind direction changes under different wind conditions.

[0101] To more accurately characterize wind direction fluctuations, the cloud computing analysis system further considers the quantitative characteristics of each wind change. These characteristics may include the specific wind speed, the angle of wind direction change, and the duration of the wind change. The system combines these quantitative characteristics with the previously constructed wind direction change probability distribution model for comprehensive analysis.

[0102] Based on comprehensive analysis, the cloud computing analysis system uses specific algorithms, such as machine learning or statistical analysis methods, to determine the wind direction jump trend vector. This vector not only captures the overall trend of wind direction changes but also reflects the sensitivity and stability of wind direction changes under different wind conditions. In this way, the system can more comprehensively understand the wind direction jump relationship between the two sets of real wind monitoring data.

[0103] Finally, the cloud computing analysis system stores the generated wind direction jump trend vector in a database for subsequent analysis and forecasting. This vector is an important basis for the system to evaluate the correlation of wind changes and provides valuable information for applications such as weather forecasting and wind energy resource assessment.

[0104] Designed in this way, the cloud computing analysis system can accurately determine the wind direction jump trend vector, thereby gaining a deeper understanding of the wind direction change correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data. This not only improves the system's sensitivity to wind changes and its prediction accuracy, but also provides more reliable and accurate wind direction change information for fields such as meteorology, aviation, and navigation. By combining target attention tags, quantitative features, and the probability distribution of wind direction changes, the cloud computing analysis system achieves a comprehensive understanding of wind direction jump trends, further enhancing the efficiency of wind information utilization and decision-making support capabilities across various industries.

[0105] It is worth mentioning that the wind direction jump trend vector is determined based on the quantitative characteristics of each wind condition change information and the wind direction change probability distribution, including: constructing a multi-dimensional wind direction change space model, in which each dimension corresponds to a quantitative characteristic of wind direction change information; in the wind direction change space model, a corresponding weight value is assigned to each quantitative feature dimension according to the wind direction change probability distribution; the quantitative feature value of each wind condition change information is mapped to the corresponding dimension, and weighted summation is performed according to the weight value to obtain a comprehensive wind direction change index; based on the comprehensive wind direction change index, the wind direction jump trend vector is determined using a vector operation method, and the vector represents the wind direction jump direction and intensity in the multi-dimensional quantitative feature space.

[0106] Specifically, the cloud computing analysis system employs a multidimensional, comprehensive analysis approach to determine wind direction jump trend vectors. The core of this approach lies in constructing a multidimensional spatial model of wind direction variation to more comprehensively and accurately capture every detail of wind change information. First, the cloud computing analysis system identifies and extracts quantitative features for each wind change information. These quantitative features may include, but are not limited to, wind speed, wind direction angle, and wind stability, which together form a rich picture of wind changes. The system ensures that each feature is accurately measured and converted into comparable numerical values. Next, the system constructs a multidimensional spatial model of wind direction variation. In this model, each dimension corresponds to a quantitative feature of wind change information. For example, one dimension may represent wind speed, and another dimension represents wind direction angle. This ensures that each wind change information has a unique location in this multidimensional space. The cloud computing analysis system then assigns a weight to each quantitative feature dimension based on the probability distribution of wind direction changes. This step is crucial because it reflects the importance of different features in wind direction jumps. For example, in some cases, changes in wind direction angle may be more important than changes in wind speed and therefore be given a higher weight. Next, the system maps the quantitative feature values ​​of each wind change information element to the corresponding dimension and performs a weighted sum based on the previously assigned weights. This process effectively calculates a comprehensive wind direction change index, which integrates information from all relevant features and considers their respective contributions. Finally, based on this comprehensive wind direction change index, the cloud computing analysis system uses vector operations to determine the wind direction jump trend vector. This vector not only indicates the approximate direction of the wind direction jump but also reflects the intensity or magnitude of the jump. In the multidimensional quantitative feature space, this vector acts as a compass, guiding the system to gain a deeper understanding and prediction of wind changes. This enables the cloud computing analysis system to analyze and predict wind direction jump trends with unprecedented accuracy and comprehensiveness. The multidimensional wind direction change spatial model ensures that every detail of the wind change information is fully considered, while the weighted summation method allows the system to flexibly adjust the emphasis of each feature based on actual conditions. The resulting wind direction jump trend vector not only provides valuable decision support for fields such as meteorological forecasting and energy management, but also significantly enhances the cloud computing analysis system's ability and accuracy in processing complex wind data.

[0107] In other possible embodiments, the meteorological environment impact characteristic value is determined based on the following steps: determining the first environmental disturbance characteristic based on the attention weight of the target attention tag corresponding to each wind condition change information in the first dynamic compensation type true wind monitoring data, and the attention weight of the target attention tag corresponding to each wind condition change information in the second dynamic compensation type true wind monitoring data; determining the second environmental disturbance characteristic based on the confidence coefficient of the target attention tag corresponding to each wind condition change information in the first dynamic compensation type true wind monitoring data, and the confidence coefficient of the target attention tag corresponding to each wind condition change information in the second dynamic compensation type true wind monitoring data; determining the second environmental disturbance characteristic based on the confidence coefficient of the target attention tag corresponding to each wind condition change information in the first dynamic compensation type true wind monitoring data The target attention tags corresponding to the respective wind change information and the target attention tags corresponding to the respective wind condition change information in the second dynamically compensated true wind monitoring data are used to determine the consistency analysis features between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; based on the consistency analysis features, the confidence coefficients of the target attention tags corresponding to the respective wind condition change information in the first dynamically compensated true wind monitoring data, and the confidence coefficients of the target attention tags corresponding to the respective wind condition change information in the second dynamically compensated true wind monitoring data, the environmental involvement vector is determined; based on the first environmental disturbance feature, the second environmental disturbance feature and the environmental involvement vector, the meteorological environment impact characteristic value is determined.

[0108] In a cloud computing analysis system, determining the meteorological environmental impact characteristic value is a complex but critical process, involving multiple steps and comprehensive analysis of multiple data sources. This example will detail this process, specifically how to combine the target attention tags in the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data to derive the final meteorological environmental impact characteristic value.

[0109] First, the system identifies the first environmental disturbance signature based on the attention weights of the target attention tags corresponding to each wind change in the two sets of real wind monitoring data. These attention weights reflect the importance of different wind change information in the overall data analysis. By comparing and comprehensively analyzing the attention weights of the two data sets, the system can identify which wind change information receives the most attention in both sets and thus use this information as a key indicator of environmental disturbance.

[0110] Next, the system considers the confidence coefficient of the target attention tag to determine the second environmental disturbance signature. The confidence coefficient represents the system's confidence in the accuracy of each wind change judgment. By combining the confidence coefficients of the two sets of data, the system can further filter out wind changes that are not only important but also highly accurate, forming the basis for the second environmental disturbance signature.

[0111] To more comprehensively assess the relationship between the two sets of real wind monitoring data, the system also performs a consistency analysis. During this step, the system compares the target attention labels corresponding to each wind change in the two data sets to determine their degree of consistency. This consistency analysis helps identify common wind change information of interest in the two data sets, as well as potential differences and inconsistencies.

[0112] Based on the results of the consistency analysis and the confidence coefficients of each wind change in the two data sets, the system further determines the environmental impact vector. This vector reflects the common characteristics and differences in the environmental disturbances between the two data sets and is an important basis for the subsequent determination of the characteristic value of the meteorological environmental impact.

[0113] Finally, the system determines the meteorological environmental impact characteristic value based on the first and second environmental disturbance characteristics, as well as the environmental involvement vector. This characteristic value is a comprehensive indicator that incorporates multiple aspects of information, including the significance of the wind change, the accuracy of the judgment, and the consistency between the two sets of data. Using this characteristic value, the system can more comprehensively assess the impact of the meteorological environment on the current wind change, providing more accurate data support for subsequent wind forecasts and decision-making.

[0114] In this process, the introduction of numerical eigenvectors to represent and calculate these eigenvalues ​​and vectors is crucial. Numerical eigenvectors allow the system to quantify various wind condition information and related characteristics into specific numerical values, enabling precise mathematical operations and analysis. This not only improves the accuracy and efficiency of analysis but also provides strong support for the automation and intelligentization of the system.

[0115] For example, there are two sets of dynamic compensation real wind monitoring data, called data set A and data set B. Each set of data contains multiple wind condition change information, and each wind condition change information corresponds to a target attention label, which includes an attention weight and a confidence coefficient.

[0116] For each wind condition change information in data sets A and B, there are the following attention weights: the attention weight of wind condition 1 in data set A is 0.8, and the attention weight of wind condition 2 is 0.6; the attention weight of wind condition 1 in data set B is 0.7, and the attention weight of wind condition 2 is 0.5. By comparing the two sets of data, the system can identify that wind condition 1 receives greater attention in both sets of data (higher attention weight), and therefore uses wind condition 1 as one of the key indicators of environmental disturbance. The first environmental disturbance feature can be a weighted average or other comprehensive indicator, such as (0.8+0.7) / 2=0.75.

[0117] Next, consider the confidence coefficient: the confidence coefficient of wind condition 1 in data set A is 0.9, and the confidence coefficient of wind condition 2 is 0.8; the confidence coefficient of wind condition 1 in data set B is 0.85, and the confidence coefficient of wind condition 2 is 0.75. The system combines the confidence coefficients of the two sets of data and further filters out wind condition 1 because it is not only important (based on the attention weight) but also has high judgment accuracy (based on the confidence coefficient). The second environmental disturbance feature can be a weighted average based on the confidence coefficient, such as (0.9 + 0.85) / 2 = 0.875.

[0118] The system compares the target attention labels corresponding to each wind condition change in the two data sets to determine the degree of consistency. In this example, wind condition 1 has a higher attention weight and confidence coefficient in both data sets, indicating that both data sets have consistent attention to wind condition 1.

[0119] Based on the consistency analysis results and the confidence coefficient, the system determines an environmental impact vector. This vector can be multidimensional, encompassing characteristics such as the importance, consistency, and confidence of each wind condition change. In this example, the environmental impact vector may emphasize the importance of wind condition 1.

[0120] Finally, the system comprehensively determines a meteorological environmental impact characteristic value based on the first environmental disturbance characteristic (0.75), the second environmental disturbance characteristic (0.875), and the environmental involvement vector. This value may be a weighted average or a more complex comprehensive calculation result, such as a weighted sum or product. For example, the meteorological environmental impact characteristic value calculated by a certain algorithm is 0.8. This characteristic value (0.8) can now be used as data support for subsequent wind forecasts and decision-making, indicating the overall impact of the current meteorological environment on wind changes. Through this specific numerical example, we can see how the cloud computing analysis system combines multiple data sources and multiple features to determine a comprehensive meteorological environmental impact characteristic value.

[0121] As can be seen, the cloud computing analysis system comprehensively considers multiple factors, including the importance of wind change information, the accuracy of judgments, and the consistency of data, to derive a comprehensive and accurate characteristic value for the impact of the meteorological environment. This characteristic value not only reflects the actual impact of the current meteorological environment on wind changes but also provides strong data support for subsequent wind forecasts and decision-making. By introducing numerical characteristic vectors for precise calculation and analysis, the system's accuracy and efficiency have been significantly improved, providing strong technical support for research and application in related fields.

[0122] In some other optional embodiments, the monitoring element association characteristics of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data are determined based on the first monitoring element embedding vector corresponding to the first dynamically compensated true wind monitoring data and the second monitoring element embedding vector corresponding to the second dynamically compensated true wind monitoring data, including at least one of the following: (1) performing compensation monitoring analysis on the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data through a compensation monitoring processing network to obtain each first compensated monitoring data stream in the first dynamically compensated true wind monitoring data and its corresponding compensated monitoring linear variable, as well as each second compensated monitoring data stream in the second dynamically compensated true wind monitoring data and its corresponding compensated monitoring linear variable. variables; based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variables and the second compensation monitoring data stream and its corresponding compensation monitoring linear variables, determine the compensation monitoring association characteristic value as the monitoring element association characteristic; (2) through the monitoring data mining model, perform monitoring vector mining on the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data to obtain the first true wind measurement data time-space linkage vector corresponding to the first dynamic compensation true wind monitoring data, and the second true wind measurement data time-space linkage vector corresponding to the second dynamic compensation true wind monitoring data; based on the first true wind measurement data time-space linkage vector and the second true wind measurement data time-space linkage vector, determine the time-space monitoring decision characteristic value as the monitoring element association characteristic.

[0123] Specifically, when processing the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, the cloud computing analysis system will use at least one of the following two methods for in-depth analysis in order to determine the correlation characteristics of the monitoring elements of the two sets of data.

[0124] The first method involves analysis through a compensation monitoring processing network. The cloud computing analysis system first uses this network to perform compensation monitoring analysis on the two sets of true wind monitoring data. This step aims to identify and extract the compensation monitoring data streams and their corresponding compensation monitoring linear variables within the data. For the first set of dynamically compensated true wind monitoring data, the system extracts each first compensation monitoring data stream and its corresponding compensation monitoring linear variable. Similarly, for the second set of dynamically compensated true wind monitoring data, the system extracts each second compensation monitoring data stream and its corresponding compensation monitoring linear variable. These compensation monitoring data streams and linear variables reflect different aspects and characteristics of changing wind conditions. After extracting this data, the cloud computing analysis system further determines the compensation monitoring correlation feature value. This feature value is calculated based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable, as well as the second compensation monitoring data stream and its corresponding compensation monitoring linear variable. It reveals the inherent connections and common characteristics of the two sets of true wind monitoring data in terms of compensation monitoring, and thus serves as part of the monitoring element correlation feature.

[0125] The second method is to mine monitoring vectors using a monitoring data mining model. The cloud computing analysis system uses this model to conduct an in-depth analysis of the two sets of true wind monitoring data, mining the temporal and spatial linkage vectors of the true wind measurement data hidden within the data. For the first set of dynamically compensated true wind monitoring data, the system mines the temporal and spatial linkage vectors of the first true wind measurement data; for the second set of dynamically compensated true wind monitoring data, it mines the temporal and spatial linkage vectors of the second true wind measurement data. These temporal and spatial linkage vectors reflect the temporal and spatial correlation and dynamic characteristics of wind changes. Based on the mined temporal and spatial linkage vectors of the first and second true wind measurement data, the cloud computing analysis system further determines the temporal and spatial monitoring decision eigenvalues. This eigenvalue reveals the common characteristics and decision-making basis of the temporal and spatial linkage between the two sets of true wind monitoring data, and is therefore considered another important part of the correlation characteristics of monitoring elements.

[0126] The following is a specific numerical example to illustrate in detail how the cloud computing analysis system processes the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data and determines the correlation characteristics of the monitoring elements.

[0127] First approach: Processing network analysis through compensation monitoring

[0128] Data extraction: For the first dynamic compensated true wind monitoring data (referred to as Data1), the system extracted three first compensated monitoring data streams (Stream1_1, Stream1_2, Stream1_3), and their corresponding compensated monitoring linear variables (Variable1_1, Variable1_2, Variable1_3); for the second dynamic compensated true wind monitoring data (referred to as Data2), the system also extracted three second compensated monitoring data streams (Stream2_1, Stream2_2, Stream2_3), and their corresponding compensated monitoring linear variables (Variable2_1, Variable2_2, Variable2_3).

[0129] Compensation monitoring data stream and linear variables of Data1:

[0130] Stream1_1: [1.2, 1.5, 1.8, 2.1], Variable1_1: 0.5;

[0131] Stream1_2: [0.8, 0.7, 0.9, 1.0], Variable1_2: -0.2;

[0132] Stream1_3: [3.0, 2.8, 3.2, 3.1], Variable1_3: 0.1.

[0133] Data2's compensation monitoring data stream and linear variables:

[0134] Stream2_1: [1.3, 1.6, 1.9, 2.2], Variable2_1: 0.4;

[0135] Stream2_2: [0.9, 0.8, 1.0, 1.1], Variable2_2: -0.1;

[0136] Stream2_3: [3.1, 2.9, 3.3, 3.2], Variable2_3: 0.2.

[0137] The cloud computing analysis system uses complex algorithms (such as weighted averaging and correlation analysis) to process these data streams and linear variables to calculate the compensation monitoring correlation characteristic value. For example, a value of 0.85 indicates that Data 1 and Data 2 have a high correlation in compensation monitoring.

[0138] The second method: through monitoring data mining model analysis

[0139] Data mining: For Data1, the system mines the time-space linkage vector (Vector1) of the first true wind measurement data; for Data2, the system mines the time-space linkage vector (Vector2) of the second true wind measurement data.

[0140] Data Example

[0141] Vector1: [(1.0, 2.0, 3.0), (0.5, 1.5, 2.5), (2.0, 3.0, 4.0)];

[0142] Vector2: [(1.1, 2.1, 3.1), (0.6, 1.6, 2.6), (2.1, 3.1, 4.1)].

[0143] Each element in these vectors represents a measurement of the wind conditions at a specific point in time and space.

[0144] The system calculates the spatiotemporal monitoring decision eigenvalue by comparing and analyzing the similarities, differences, and dynamic changes between Vector1 and Vector2 to determine a spatiotemporal monitoring decision eigenvalue. This value may be based on complex calculations such as the distance, angle, and correlation between the vectors. For example, a value of 0.9 indicates a high degree of similarity between the two sets of data in terms of spatiotemporal linkage.

[0145] Through a comprehensive analysis of the two aforementioned methods, the cloud computing analysis system was able to derive correlation eigenvalues ​​for the two sets of dynamically compensated true wind monitoring data, both in terms of compensation monitoring and spatiotemporal linkage. These eigenvalues ​​are crucial for subsequent wind forecasting, decision support, and the optimal utilization of wind energy resources. In this example, the correlation eigenvalue for compensation monitoring was 0.85, and the eigenvalue for spatiotemporal monitoring decision-making was 0.9, both demonstrating high correlation.

[0146] In this way, the cloud computing analysis system can comprehensively and deeply analyze the correlation characteristics of monitoring elements in two sets of dynamically compensated true wind monitoring data. This not only improves the accuracy and reliability of wind monitoring, but also provides stronger data support for subsequent wind forecasting and decision-making. Furthermore, by mining the temporal and spatial linkage characteristics and compensated monitoring correlation characteristics hidden in the data, the system can better understand the complexity and dynamic nature of wind conditions, enabling more scientific and accurate decision-making. This has important implications for wind power generation, weather forecasting, environmental protection, and other fields.

[0147] In the next step, the compensation monitoring associated characteristic value is determined based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable and the second compensation monitoring data stream and its corresponding compensation monitoring linear variable, including: using each of the first compensation monitoring data streams and its corresponding compensation monitoring linear variable, and each of the second compensation monitoring data streams and its corresponding compensation monitoring linear variable, to generate a plurality of initial compensation monitoring data stream tuples; the initial compensation monitoring data stream tuple includes one first compensation monitoring data stream and one second compensation monitoring data stream, and the first compensation monitoring data stream and the second compensation monitoring data stream belonging to the same initial compensation monitoring data stream tuple each corresponding to the compensation monitoring linear variable. The measured linear variables meet the first true wind monitoring scenario matching threshold; a target compensation monitoring data stream tuple is determined among the multiple initial compensation monitoring data stream tuples; the compensation monitoring linear variables corresponding to the first compensation monitoring data stream and the second compensation monitoring data stream in the target compensation monitoring data stream tuple meet the second true wind monitoring scenario matching threshold; the compensation monitoring association characteristic value is determined by using the number of the first compensation monitoring data stream and the second compensation monitoring data stream included in each of the target compensation monitoring data stream tuples, the total number of the first compensation monitoring data stream in the first dynamic compensation true wind monitoring data, and the total number of the second compensation monitoring data stream in the second dynamic compensation true wind monitoring data.

[0148] Specifically, when processing the first dynamic compensated true wind monitoring data and the second dynamic compensated true wind monitoring data, the cloud computing analysis system performs the following steps to determine the compensated monitoring associated characteristic value:

[0149] First, the cloud computing analysis system uses the first compensated monitoring data stream and its corresponding compensated monitoring linear variable, as well as the second compensated monitoring data stream and its corresponding compensated monitoring linear variable, extracted from two sets of real wind monitoring data to generate multiple initial compensated monitoring data stream tuples. These tuples consist of a first compensated monitoring data stream and a second compensated monitoring data stream.

[0150] Crucially, the data streams that make up these binary pairs must meet a single condition: their corresponding compensation monitoring linear variables must meet the first true wind monitoring scenario matching threshold. This threshold is a pre-set criterion used by the system to determine whether two compensation monitoring data streams have similarities or correlations under certain specific monitoring scenarios or conditions.

[0151] Next, the cloud computing analysis system selects target compensated monitoring data stream tuples from these initial compensated monitoring data stream tuples. These target tuples are characterized by the fact that the compensated monitoring linear variables corresponding to the first and second compensated monitoring data streams not only meet a first matching threshold for the real wind monitoring scenario, but also further meet a second matching threshold for the real wind monitoring scenario. This second threshold may be a more stringent or specific matching condition, ensuring that the selected data stream tuples have higher relevance and importance in a specific real wind monitoring scenario.

[0152] Finally, the cloud computing analysis system uses the number of filtered target compensation monitoring data stream tuples and the total number of compensation monitoring data streams in the original first and second dynamic compensation true wind monitoring data to determine the compensation monitoring correlation characteristic value. This characteristic value is actually a comprehensive indicator that reflects the inherent connection and similarity between the two sets of true wind monitoring data in terms of compensation monitoring.

[0153] In some examples, the cloud computing analysis system extracts three first compensated monitoring data streams (Stream1_1, Stream1_2, Stream1_3) and their corresponding compensated monitoring linear variables (Variable1_1, Variable1_2, Variable1_3) from the first dynamically compensated true wind monitoring data (Data1). Similarly, the cloud computing analysis system extracts three second compensated monitoring data streams (Stream2_1, Stream2_2, Stream2_3) and their corresponding compensated monitoring linear variables (Variable2_1, Variable2_2, Variable2_3) from the second dynamically compensated true wind monitoring data (Data2).

[0154] Data1:

[0155] Stream1_1: [1.2, 1.5, 1.8], Variable1_1: 0.5;

[0156] Stream1_2: [0.8, 0.7, 1.0], Variable1_2: -0.2;

[0157] Stream1_3: [3.0, 3.2, 3.1], Variable1_3: 0.1.

[0158] Data2:

[0159] Stream2_1: [1.3, 1.6, 1.9], Variable2_1: 0.4;

[0160] Stream2_2: [0.9, 0.8, 1.1], Variable2_2: -0.1;

[0161] Stream2_3: [3.1, 3.3, 3.2], Variable2_3: 0.2.

[0162] The system combines the qualified data streams into an initial compensation monitoring data stream binary according to the first true wind monitoring scene matching threshold (e.g., 0.3). For example, (Stream1_1, Stream2_1), (Stream1_2, Stream2_2), and (Stream1_3, Stream2_3) may all meet the conditions.

[0163] Next, the system selects the target compensated monitoring data stream pair from these initial pairs based on a second true wind monitoring scenario matching threshold (e.g., 0.1). In this example, only the pair (Stream1_2, Stream2_2) meets the condition because the absolute value of the difference between their linear variables (-0.2 - (-0.1) = -0.1) is less than the second threshold.

[0164] Finally, the cloud computing analysis system uses the number of target compensation monitoring data stream dyads and the total number of data streams to determine the compensation monitoring correlation feature value. In this example, there is one target dyad and the total number of data streams is six (three from Data1 and three from Data2). Therefore, the compensation monitoring correlation feature value can be calculated using a specific algorithm, such as using the ratio of the number of target dyads to the total number of data streams, or other more complex statistical methods.

[0165] For example, if the algorithm used by the system is a simple proportional calculation, the compensation monitoring correlation characteristic value will be 1 / 6 ≈ 0.167. This value means that in the two sets of real wind monitoring data, only about 16.7% of the data streams show strong correlation under the condition that the matching threshold of the specific monitoring scenario is met.

[0166] In this example, we can see how the cloud computing analysis system gradually processes and analyzes two sets of dynamically compensated true wind monitoring data to determine the associated eigenvalues ​​of the compensated monitoring. This process involves extracting, combining, filtering, and calculating eigenvalues ​​from data streams, ultimately yielding a comprehensive indicator reflecting the inherent connections between the data.

[0167] As can be seen, the cloud computing analysis system can accurately determine the correlation characteristic values ​​of the two sets of dynamically compensated true wind monitoring data in terms of compensation monitoring. This not only improves the accuracy and reliability of wind monitoring, but also provides strong data support for subsequent wind forecasting and decision-making. By setting different matching thresholds for true wind monitoring scenarios, the system can flexibly adapt to different monitoring needs and environmental conditions, thereby making more scientific and accurate decisions. This has important implications for a variety of fields, including wind power generation, weather forecasting, and environmental protection. Furthermore, this data stream-based correlation characteristic analysis method also provides new ideas and methodologies for other similar data processing scenarios.

[0168] In some other possible embodiments, the first true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data, and the second true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data; the spatiotemporal monitoring decision feature value is determined based on the first true wind measurement data spatiotemporal linkage vector and the second true wind measurement data spatiotemporal linkage vector, including: for each wind condition change information binary group, based on the two wind condition change information in the wind condition change information binary group, the multidimensional wind field environment vector corresponding to each wind condition change information The wind condition change disturbance error corresponding to the wind condition change information binary group is determined according to the corresponding multi-dimensional wind field environment vector; the wind condition change information binary group includes a wind condition change information belonging to the first dynamically compensated true wind monitoring data and a wind condition change information belonging to the second dynamically compensated true wind monitoring data, and the two wind condition change information belonging to the same wind condition change information binary group correspond to the same distribution label in the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the spatiotemporal monitoring decision feature value is determined by using the wind condition change disturbance error corresponding to each of the wind condition change information binary groups.

[0169] Based on this embodiment, when processing the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, the cloud computing analysis system adopts a method based on a multi-dimensional wind field environment vector in order to determine the spatiotemporal monitoring decision feature value.

[0170] First, the first and second true wind measurement data spatiotemporal linkage vectors each contain a multidimensional wind field environment vector corresponding to each wind condition change in their respective dynamically compensated true wind monitoring data. These multidimensional wind field environment vectors may include various wind field environment-related parameters such as wind direction, wind speed, and wind pressure. Together, they form a multidimensional data representation that comprehensively describes the environmental state during wind changes.

[0171] Next, the cloud computing analysis system operates on each pair of wind condition change information. These pairs consist of one wind condition change information from the first dynamically compensated true wind monitoring data and one wind condition change information from the second dynamically compensated true wind monitoring data. Importantly, these two wind condition change information correspond to the same distribution label in their respective monitoring data, meaning they were recorded under similar or identical environmental conditions.

[0172] For each wind condition change binary, the cloud computing analysis system determines the corresponding wind condition change disturbance error based on the multidimensional wind field environment vectors corresponding to each of the two wind condition change information. This disturbance error actually reflects the consistency of wind condition change information under the same environmental conditions in two different real wind monitoring data sets. If the disturbance error is small, it indicates that the wind condition change information of the two monitoring data sets under the same environmental conditions is relatively consistent; otherwise, it indicates a significant difference.

[0173] Finally, the cloud computing analysis system uses the wind disturbance error corresponding to each wind condition information binary to determine the spatiotemporal monitoring decision-making characteristic value. This characteristic value is a comprehensive indicator that takes into account the disturbance error of all wind condition information binary tuples, thereby fully reflecting the similarities and differences between the two sets of real wind monitoring data in terms of spatiotemporal monitoring. This characteristic value provides an important reference for subsequent wind condition forecasting and decision-making.

[0174] Through the above method, the cloud computing analysis system can comprehensively consider multi-dimensional wind farm environmental parameters and more comprehensively evaluate the consistency of wind condition change information between two sets of real wind monitoring data under the same environmental conditions. This not only improves the accuracy and reliability of wind condition monitoring, but also provides more scientific data support for subsequent wind condition forecasting and decision-making. At the same time, this method can also effectively identify the differences in wind condition change information existing in different monitoring data, providing a useful reference for further optimizing real wind monitoring technology. Overall, this spatiotemporal monitoring and decision-making method based on multi-dimensional wind farm environmental vectors is of great significance for improving the monitoring level and technological progress in fields such as wind power generation and meteorological forecasting.

[0175] In an optional embodiment, when the wind condition change association characteristics include a wind direction jump trend vector and a meteorological environment impact characteristic value, and the monitoring element association characteristics include a compensation monitoring association characteristic value and a spatiotemporal monitoring decision characteristic value, the global state prediction correlation between the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data is determined based on the wind condition change association characteristics and the monitoring element association characteristics, including: when the wind direction jump trend vector belongs to a first quantitative trend range, determining that the global state prediction correlation is not correlated; when the spatiotemporal monitoring decision characteristic value is less than a set spatiotemporal monitoring characteristic value, determining that the global state prediction correlation is correlated; when the wind direction jump trend vector belongs to a second quantitative trend range, if the meteorological environment impact characteristic value is not greater than the set influence characteristic value, or the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then determining the global state prediction correlation. The measured correlation is that there is no correlation; otherwise, it is determined that the global state prediction correlation is correlated; the wind direction jump amplitude included in the second quantitative trend range is greater than the wind direction jump amplitude included in the first quantitative trend range; when the wind direction jump trend vector belongs to the third quantitative trend range, if the meteorological environment impact characteristic value is not greater than the set impact characteristic value, and the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then it is determined that there is no correlation in the global state prediction correlation; otherwise, it is determined that there is correlation in the global state prediction correlation; the wind direction jump amplitude included in the third quantitative trend range is greater than the wind direction jump amplitude included in the second quantitative trend range; when the wind direction jump trend vector belongs to the fourth quantitative trend range, it is determined that there is correlation in the global state prediction correlation; the wind direction jump amplitude included in the fourth quantitative trend range is greater than the wind direction jump amplitude included in the third quantitative trend range.

[0176] In cloud computing analysis systems, determining the global state prediction correlation is a complex but critical process that involves comprehensive consideration of multiple eigenvalues. In this optional embodiment, we will further explore how to determine the global state prediction correlation between the first and second dynamically compensated true wind monitoring data based on the wind direction jump trend vector, meteorological environment impact eigenvalue, compensation monitoring association eigenvalue, and spatiotemporal monitoring decision eigenvalue.

[0177] First, the cloud computing analysis system analyzes the wind direction jump trend vector. This vector reflects the degree and direction of wind direction change over a short period of time. Based on the magnitude of the wind direction jump, the system divides it into four quantitative trend ranges: the first through fourth quantitative trend ranges, each representing a gradually increasing magnitude of wind direction jump.

[0178] First, when the wind direction jump trend vector falls within the first quantitative trend range, it means that the wind direction change is relatively small, and the system determines that there is no global state prediction correlation between the two sets of monitoring data. This is because small wind direction changes usually do not have a significant impact on the global state.

[0179] Next, the system examines the spatiotemporal monitoring decision eigenvalue. If this eigenvalue is less than a preset spatiotemporal monitoring eigenvalue, the system deems the two sets of data to be related. This is because a smaller spatiotemporal monitoring decision eigenvalue may indicate similarities in the spatiotemporal distribution of the two sets of data, thus increasing their correlation.

[0180] Third, when the wind direction jump trend vector enters the second quantized trend range, the system begins to consider the meteorological environment impact characteristic value and the compensation monitoring association characteristic value. If the meteorological environment impact characteristic value is not greater than a set impact characteristic value, or the compensation monitoring association characteristic value is not greater than a set compensation monitoring characteristic value, the system determines that there is no correlation between the two sets of data. This is because large wind direction changes may be caused by specific meteorological environment or compensation monitoring factors, and these factors may not appear in both sets of data at the same time.

[0181] Fourth, when the wind direction jump trend vector increases further into the third quantized trend range, the system adopts stricter criteria for determining the meteorological environment impact characteristic value and the compensation monitoring correlation characteristic value. Only when both characteristic values ​​are simultaneously greater than their respective set values ​​does the system consider a correlation between the two sets of data. This is because within this wind direction jump range, the influence of external factors may be more significant, requiring stricter conditions to confirm the correlation.

[0182] Finally, when the wind direction jump trend vector reaches the fourth quantized trend range, meaning the wind direction change is very significant, the system directly determines that there is a correlation between the two sets of data. This is likely because extreme wind direction changes significantly affect both sets of data, thus indicating a correlation.

[0183] By comprehensively considering factors such as wind direction jump trends, meteorological environmental impacts, compensation monitoring relevance, and spatiotemporal monitoring decisions, the cloud computing analysis system can more accurately determine the global state prediction correlation between two sets of dynamically compensated true wind monitoring data. This approach not only improves the accuracy and reliability of correlation analysis but also provides strong data support for subsequent wind energy forecasting, meteorological analysis, and wind farm operation and maintenance decision-making. Through refined correlation judgment, wind farms can more effectively utilize wind energy resources, optimize operation and maintenance strategies, reduce operating costs, and improve wind power generation efficiency and reliability.

[0184] Under some preferred design ideas, the acquisition of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data to be analyzed includes: acquiring the first true wind monitoring data to be processed and the second true wind monitoring data to be processed; determining one true wind monitoring data to be processed from the first true wind monitoring data to be processed and the second true wind monitoring data to be processed as the dynamically compensated true wind monitoring reference data, and determining another true wind monitoring data to be processed as the dynamically compensated true wind monitoring data to be optimized; optimizing the dynamically compensated true wind monitoring data to be optimized based on the dynamically compensated true wind monitoring reference data to obtain dynamically compensated true wind monitoring optimization data; using the dynamically compensated true wind monitoring reference data and the dynamically compensated true wind monitoring optimization data as the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data.

[0185] In some preferred embodiments, the process of obtaining the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data to be analyzed is highly sophisticated and technical, and is intended to ensure that the analyzed data has the highest accuracy and reliability.

[0186] First, the cloud computing analysis system obtains the first and second sets of true wind monitoring data to be processed. These data may come from different meteorological monitoring stations or be collected from the same station over different time periods. They contain raw records of key meteorological information such as wind speed and direction.

[0187] Next, the system selects one of the first and second sets of true wind monitoring data to be processed as the reference data for dynamic compensation. This selection may be based on a combination of factors, including data quality, continuity, and historical accuracy. The selected data will serve as the baseline for subsequent analysis, allowing comparison and optimization of other data.

[0188] At the same time, another set of real wind monitoring data to be processed was identified as dynamically compensated real wind monitoring data to be optimized. This means that this data may contain certain errors or inconsistencies and needs to be improved through comparison and optimization with the reference data.

[0189] The cloud computing analysis system then optimizes the dynamically compensated true wind monitoring data based on the dynamically compensated true wind monitoring reference data. This optimization process may include multiple steps, such as data cleaning, outlier processing, and missing value filling, to ensure that the optimized data maintains a high degree of consistency with the reference data in terms of statistical characteristics and change trends.

[0190] Finally, the optimized dynamic compensation true wind monitoring data (i.e., dynamic compensation true wind monitoring optimization data) and the original dynamic compensation true wind monitoring reference data will be used as the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, respectively, for subsequent analysis.

[0191] In this way, the cloud computing analysis system ensures the high accuracy and consistency of the first and second dynamically compensated true wind monitoring data. This not only improves the reliability of data analysis but also provides a solid data foundation for subsequent decisions such as wind energy resource assessment and wind farm operation optimization. Furthermore, this data optimization method can be applied to other similar meteorological or environmental monitoring fields, possessing broad practicality and social value.

[0192] In some independent embodiments, the dynamically compensated true wind monitoring data to be optimized is optimized based on the dynamically compensated true wind monitoring reference data to obtain dynamically compensated true wind monitoring optimization data, including: when the feature granularity of the dynamically compensated true wind monitoring data to be optimized is different from that of the dynamically compensated true wind monitoring reference data, the feature granularity of the dynamically compensated true wind monitoring data to be optimized is adjusted, and the feature granularity of the dynamically compensated true wind monitoring data to be optimized is updated to the feature granularity of the dynamically compensated true wind monitoring reference data; when the feature granularity of the dynamically compensated true wind monitoring data to be optimized is different from that of the dynamically compensated true wind monitoring reference data, the feature granularity of the dynamically compensated true wind monitoring data to be optimized is adjusted to the feature granularity of the dynamically compensated true wind monitoring reference data; When the measurement standards of the monitoring reference data are different, the measurement standard adjustment rules are adopted to adjust the measurement standards of the dynamically compensated true wind monitoring data to be optimized, and the measurement standards of the dynamically compensated true wind monitoring data to be optimized are updated to the measurement standards of the dynamically compensated true wind monitoring reference data; based on the dynamically compensated true wind monitoring data to be optimized and the dynamically compensated true wind monitoring reference data with the same feature granularity and the same measurement standards, the data optimization indication feature is determined; and the wind condition change information of the dynamically compensated true wind monitoring data to be optimized is optimized using the data optimization indication feature to obtain the dynamically compensated true wind monitoring optimization data.

[0193] Based on this embodiment, the process of optimizing dynamically compensated true wind monitoring data is a complex and sophisticated process. This process ensures that even if the data to be optimized differs from the reference data in characteristics or measurement standards, a series of adjustment steps can be used to bring the two into line, thereby achieving effective data optimization.

[0194] First, the system checks whether the feature granularity of the dynamically compensated true wind monitoring data to be optimized is the same as that of the dynamically compensated true wind monitoring reference data. Feature granularity can be understood as the level of detail or resolution of the data. For example, in wind monitoring, feature granularity might refer to the frequency of data collection, the accuracy of wind speed and direction, and so on.

[0195] If a difference in feature granularity is detected, the cloud computing analysis system adjusts the feature granularity of the target data. This means that the system can increase or decrease the detail of the data to match the feature granularity of the reference data through interpolation, smoothing, or other mathematical methods. For example, if the reference data records wind speed every minute, while the target data records it every five minutes, the system can use interpolation to generate minute-by-minute wind speed records for the target data.

[0196] Next, the system checks whether the measurement standards of the two are consistent. Measurement standards refer to the specifications and guidelines followed during data collection and recording. If the measurement standards are different, direct comparison of the data will be meaningless.

[0197] When a difference in measurement standards is detected, the cloud computing analysis system applies pre-defined measurement standard adjustment rules to the data being optimized. This may involve unit conversion, application of calibration factors, or data rescaling to ensure that the measurement standard of the optimized data matches the reference data. For example, if the reference data uses meters per second (m / s) for wind speed, while the optimized data uses kilometers per hour (km / h), the system will perform a unit conversion.

[0198] After adjusting the feature granularity and measurement standards, the cloud computing analysis system will identify features that indicate data optimization based on the two now-aligned data sets. These features may include sudden changes in wind speed and the stability of wind direction, providing guidance for subsequent data optimization.

[0199] Finally, using these data to optimize the indicator features, the system will optimize the wind condition information of the dynamically compensated true wind monitoring data to be optimized. This may include removing noise, filling missing values, and correcting outliers to obtain more accurate and reliable dynamically compensated true wind monitoring optimized data.

[0200] In other independent embodiments, the data optimization indication feature is determined based on the dynamically compensated true wind monitoring data to be optimized and the dynamically compensated true wind monitoring reference data with the same feature granularity and the same measurement standard, including: performing compensation monitoring analysis on the dynamically compensated true wind monitoring reference data and the dynamically compensated true wind monitoring data to be optimized through a compensation monitoring processing network, to obtain each third compensation monitoring data stream in the dynamically compensated true wind monitoring reference data and its corresponding compensation monitoring linear variables, as well as each fourth compensation monitoring data stream in the dynamically compensated true wind monitoring data to be optimized and its corresponding compensation monitoring linear variables; using each third compensation The target compensation monitoring data stream tuple is determined by using the monitoring data streams and their corresponding compensation monitoring linear variables, as well as each of the fourth compensation monitoring data streams and their corresponding compensation monitoring linear variables; the target compensation monitoring data stream tuple includes a third compensation monitoring data stream and a fourth compensation monitoring data stream, and the compensation monitoring linear variables corresponding to the third compensation monitoring data stream and the fourth compensation monitoring data stream belonging to the same target compensation monitoring data stream tuple meet the set association relationship; the data optimization indication feature is determined based on the third compensation monitoring data stream and the fourth compensation monitoring data stream included in the target compensation monitoring data stream tuple.

[0201] In this embodiment, the process of determining data optimization indicators involves in-depth analysis of dynamically compensated real wind monitoring data using a compensated monitoring processing network. This process aims to identify potential correlations between the data to be optimized and the reference data, thereby more accurately guiding data optimization efforts.

[0202] First, the cloud computing analysis system uses a compensation monitoring processing network to perform compensation monitoring analysis on the dynamic compensation true wind monitoring reference data and the dynamic compensation true wind monitoring data to be optimized. This network may be composed of a series of complex algorithms and models, which can deeply explore the hidden information in the data.

[0203] Through compensation monitoring analysis, the system obtains the third compensation monitoring data streams and their corresponding compensation monitoring linear variables in the dynamically compensated true wind monitoring reference data, as well as the fourth compensation monitoring data streams and their corresponding compensation monitoring linear variables in the dynamically compensated true wind monitoring data to be optimized. These data streams and linear variables are intermediate products of the analysis process, reflecting the details and characteristics of wind condition changes.

[0204] Next, the system uses these third and fourth compensation monitoring data streams and their corresponding compensation monitoring linear variables to determine the target compensation monitoring data stream tuple. This process compares and matches the linear variables between different data streams to find data stream pairs that meet predefined relationships. These predefined relationships may be defined based on historical data, expert knowledge, or statistical models, and they help identify key features and patterns in the data.

[0205] A target compensation monitoring data stream tuple includes a third compensation monitoring data stream and a fourth compensation monitoring data stream, and the compensation monitoring linear variables corresponding to the two data streams have a predetermined correlation relationship. This correlation relationship may be manifested as linear correlation, time delay similarity, or other forms of statistical dependence.

[0206] Finally, based on the third and fourth compensation monitoring data streams included in these target compensation monitoring data stream tuples, the cloud computing analysis system determines data optimization indicators. These indicators may include correlation coefficients between data streams, time delay parameters, or other indicators that reflect the correlation between data. These indicators will provide important reference and guidance for subsequent data optimization work.

[0207] Through these steps, the cloud computing analysis system can deeply explore the hidden information and correlations in dynamically compensated real wind monitoring data, more accurately identifying the similarities and differences between the data to be optimized and the reference data, helping to improve the accuracy and efficiency of data optimization. The data optimization indicators determined by this method are more targeted and practical, significantly improving the analytical capabilities and adaptability of the cloud computing analysis system in complex and changing wind conditions.

[0208] Further, Figure 2 This is a schematic diagram of the structure of a cloud computing analysis system 200 provided in an embodiment of the present application. Figure 2 The cloud computing analysis system 200 shown includes a processor 210, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0209] Alternatively, as Figure 2 As shown, the cloud computing analysis 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.

[0210] The memory 230 may be a separate device independent of the processor 210 , or may be integrated into the processor 210 .

[0211] Alternatively, as Figure 2As shown, the cloud computing analysis system 200 may further include a transceiver 220 , and the processor 210 may control the transceiver 220 to interact with other devices. Specifically, the transceiver 220 may send information or data to other devices, or receive information or data sent by other devices.

[0212] Optionally, the cloud computing analysis system 200 can implement the corresponding processes corresponding to the storage engine or components in the storage engine (such as processing modules) or devices deployed with the storage engine in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0213] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly implemented as a hardware decoding processor, or can be implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0214] 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 memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DRRAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0215] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0216] Based on the above, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program implements the above method when running.

[0217] The above are merely examples of the present application and are not intended to limit the present application. For those skilled in the art, the present application may be subject to various modifications and variations.

Claims

1. A three-dimensional true wind measurement data analysis method based on cloud computing, characterized in that: Applied to a cloud computing analysis system, the method includes: Acquire first dynamic compensation true wind monitoring data and second dynamic compensation true wind monitoring data to be analyzed; determining, based on an attention tag for each piece of wind condition change information in the first dynamically compensated true wind monitoring data and an attention tag for each piece of wind condition change information in the second dynamically compensated true wind monitoring data, a wind condition change correlation feature between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; the wind condition change correlation feature being used to reflect a state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data at a wind condition change level; Determining, based on a first monitoring element embedding vector corresponding to the first dynamically compensated true wind monitoring data and a second monitoring element embedding vector corresponding to the second dynamically compensated true wind monitoring data, a monitoring element association feature of the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data, including at least one of the following: performing compensation monitoring analysis on the first dynamic compensated true wind monitoring data and the second dynamic compensated true wind monitoring data through a compensation monitoring processing network to obtain each first compensation monitoring data stream in the first dynamic compensated true wind monitoring data and its corresponding compensation monitoring linear variable, as well as each second compensation monitoring data stream in the second dynamic compensated true wind monitoring data and its corresponding compensation monitoring linear variable; determining a compensation monitoring association characteristic value based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable and the second compensation monitoring data stream and its corresponding compensation monitoring linear variable as the monitoring element association feature; Performing monitoring vector mining on the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data through a monitoring data mining model to obtain a first true wind measurement data spatiotemporal linkage vector corresponding to the first dynamically compensated true wind monitoring data, and a second true wind measurement data spatiotemporal linkage vector corresponding to the second dynamically compensated true wind monitoring data; determining a spatiotemporal monitoring decision feature value based on the first true wind measurement data spatiotemporal linkage vector and the second true wind measurement data spatiotemporal linkage vector as the monitoring element association feature; The monitoring element association feature is used to reflect the state prediction correlation between the first dynamic compensation type true wind monitoring data and the second dynamic compensation type true wind monitoring data at the monitoring element level; Determining a global state prediction correlation between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on the wind condition change correlation characteristics and the monitoring element correlation characteristics; The determining, based on the attention tags of each wind condition change information in the first dynamic compensation true wind monitoring data and the attention tags of each wind condition change information in the second dynamic compensation true wind monitoring data, a wind condition change correlation feature between the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, includes: For each piece of wind condition change information in the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data, determining a target attention tag corresponding to the wind condition change information based on each wind direction attention tag of the wind condition change information; Based on the target attention tags corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data and the target attention tags corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data, at least one of the wind direction jump trend vector and the meteorological environment impact characteristic value is determined as the wind condition change correlation feature; wherein, the wind direction jump trend vector reflects the similarity and difference between the two sets of data in wind direction changes, and is an indicator for evaluating the correlation of wind condition changes; the meteorological environment impact characteristic value takes into account the comprehensive influence of external factors on wind condition changes, and the external factors include temperature and humidity.

2. The method according to claim 1, characterized in that The wind direction jump trend vector is determined based on the following steps: determining a wind direction change probability distribution based on target attention labels corresponding to respective pieces of wind condition change information in the first dynamic compensation true wind monitoring data and target attention labels corresponding to respective pieces of wind condition change information in the second dynamic compensation true wind monitoring data; The wind direction jump trend vector is determined according to the quantitative characteristics of each wind condition change information and the wind direction change probability distribution.

3. The method according to claim 1, characterized in that The meteorological environment impact characteristic value is determined based on the following steps: determining a first environmental disturbance feature based on the attention weights of the target attention tags corresponding to the respective wind condition change information in the first dynamic compensation true wind monitoring data and the attention weights of the target attention tags corresponding to the respective wind condition change information in the second dynamic compensation true wind monitoring data; determining a second environmental disturbance feature based on a confidence coefficient of a target attention label corresponding to each piece of wind condition change information in the first dynamic compensation true wind monitoring data and a confidence coefficient of a target attention label corresponding to each piece of wind condition change information in the second dynamic compensation true wind monitoring data; Determining consistency analysis features between the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data based on target attention labels corresponding to respective pieces of wind condition change information in the first dynamically compensated true wind monitoring data and target attention labels corresponding to respective pieces of wind condition change information in the second dynamically compensated true wind monitoring data; determining an environmental involvement vector based on the consistency analysis feature, the confidence coefficient of the target attention label corresponding to each piece of wind condition change information in the first dynamic compensation true wind monitoring data, and the confidence coefficient of the target attention label corresponding to each piece of wind condition change information in the second dynamic compensation true wind monitoring data; The meteorological environment impact characteristic value is determined according to the first environmental disturbance characteristic, the second environmental disturbance characteristic, and the environmental involvement vector.

4. The method according to claim 1, wherein The determining of the compensation monitoring associated characteristic value based on the first compensation monitoring data stream and its corresponding compensation monitoring linear variable and the second compensation monitoring data stream and its corresponding compensation monitoring linear variable includes: Using each of the first compensation monitoring data streams and their corresponding compensation monitoring linear variables, and each of the second compensation monitoring data streams and their corresponding compensation monitoring linear variables, a plurality of initial compensation monitoring data stream tuples are generated; the initial compensation monitoring data stream tuple includes one of the first compensation monitoring data stream and one of the second compensation monitoring data streams, and the compensation monitoring linear variables corresponding to the first compensation monitoring data stream and the second compensation monitoring data stream belonging to the same initial compensation monitoring data stream tuple meet a first true wind monitoring scenario matching threshold; Determining a target compensation monitoring data stream binary from the multiple initial compensation monitoring data stream binary groups; the compensation monitoring linear variables corresponding to the first compensation monitoring data stream and the second compensation monitoring data stream in the target compensation monitoring data stream binary group meet a second true wind monitoring scenario matching threshold; The compensation monitoring association characteristic value is determined by using the number of the first compensation monitoring data stream and the second compensation monitoring data stream included in each target compensation monitoring data stream tuple, as well as the total number of the first compensation monitoring data stream in the first dynamic compensation true wind monitoring data and the total number of the second compensation monitoring data stream in the second dynamic compensation true wind monitoring data.

5. The method according to claim 1, wherein The first true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the first dynamic compensation true wind monitoring data, and the second true wind measurement data spatiotemporal linkage vector includes a multidimensional wind field environment vector corresponding to each wind condition change information in the second dynamic compensation true wind monitoring data; determining the spatiotemporal monitoring decision characteristic value based on the first true wind measurement data spatiotemporal linkage vector and the second true wind measurement data spatiotemporal linkage vector includes: For each wind condition change information binary group, based on the multidimensional wind field environment vectors corresponding to the two wind condition change information pieces in the wind condition change information binary group, a wind condition change disturbance error corresponding to the wind condition change information binary group is determined; the wind condition change information binary group includes one piece of wind condition change information belonging to the first dynamically compensated true wind monitoring data and one piece of wind condition change information belonging to the second dynamically compensated true wind monitoring data, and the two pieces of wind condition change information belonging to the same wind condition change information binary group correspond to the same distribution label in the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data; The spatiotemporal monitoring decision feature value is determined by utilizing the wind condition change disturbance error corresponding to each of the wind condition change information binary groups.

6. The method according to any one of claims 1 to 5, characterized in that When the wind condition change association feature includes a wind direction jump trend vector and a meteorological environment impact characteristic value, and the monitoring element association feature includes a compensation monitoring association characteristic value and a spatiotemporal monitoring decision characteristic value, determining the global state prediction correlation between the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data based on the wind condition change association feature and the monitoring element association feature includes: When the wind direction jump trend vector belongs to the first quantized trend range, determining that the global state prediction correlation is non-correlation; When the spatiotemporal monitoring decision characteristic value is less than the set spatiotemporal monitoring characteristic value, determining that the global state prediction correlation is correlated; When the wind direction jump trend vector belongs to the second quantitative trend range, if the meteorological environment impact characteristic value is not greater than the set impact characteristic value, or the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then it is determined that the global state prediction correlation is not associated; otherwise, it is determined that the global state prediction correlation is associated; the wind direction jump amplitude included in the second quantitative trend range is greater than the wind direction jump amplitude included in the first quantitative trend range; When the wind direction jump trend vector belongs to the third quantized trend range, if the meteorological environment impact characteristic value is not greater than the set impact characteristic value, and the compensation monitoring association characteristic value is not greater than the set compensation monitoring characteristic value, then it is determined that the global state prediction correlation is not associated; otherwise, it is determined that the global state prediction correlation is associated; the wind direction jump amplitude included in the third quantized trend range is greater than the wind direction jump amplitude included in the second quantized trend range; When the wind direction jump trend vector belongs to a fourth quantized trend range, the global state prediction correlation is determined to be correlated; and the wind direction jump amplitude included in the fourth quantized trend range is greater than the wind direction jump amplitude included in the third quantized trend range.

7. The method according to claim 1, characterized in that The obtaining of the first dynamic compensation true wind monitoring data and the second dynamic compensation true wind monitoring data to be analyzed includes: Acquire first true wind monitoring data to be processed and second true wind monitoring data to be processed; Determining one of the first true wind monitoring data to be processed and the second true wind monitoring data to be processed as dynamic compensation true wind monitoring reference data, and determining the other true wind monitoring data to be processed as dynamic compensation true wind monitoring data to be optimized; Optimizing the dynamic compensation type true wind monitoring data to be optimized based on the dynamic compensation type true wind monitoring reference data to obtain dynamic compensation type true wind monitoring optimized data; The dynamically compensated true wind monitoring reference data and the dynamically compensated true wind monitoring optimization data are used as the first dynamically compensated true wind monitoring data and the second dynamically compensated true wind monitoring data.

8. A cloud computing analysis 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 7.

Images