Meteorological element monitoring method and system based on multi-source data fusion

Abstract

The application provides a meteorological element monitoring method and system based on multi-source data fusion. First, a multi-source meteorological data set containing meteorological element observation data units with spatial and temporal markers collected by different types of monitoring equipment is obtained. Then, feature extraction is performed based on the multi-source meteorological data set to generate a fusion meteorological feature set containing multi-element correlation features. Next, meteorological element anomaly recognition processing is performed using the fusion meteorological feature set to determine the meteorological element anomaly distribution area and anomaly type in the monitoring area. Finally, a meteorological monitoring analysis report containing spatial and temporal positioning information is generated based on the anomaly distribution area and anomaly type. Thus, by fusing multi-source data, the accuracy and comprehensiveness of meteorological element monitoring are improved.

Description

Technical Field

[0001] This application relates to the field of meteorological monitoring technology, and more specifically, to a method and system for monitoring meteorological elements based on multi-source data fusion. Background Technology

[0002] Accurate monitoring of meteorological elements is crucial for numerous fields, including weather forecasting, disaster warning, agricultural production, and transportation. Currently, meteorological element monitoring mainly relies on single types of monitoring equipment, such as ground-based weather stations, meteorological satellites, and radar. These single devices have significant limitations in data acquisition. Ground-based weather stations are geographically limited, only able to acquire meteorological data for local areas, and have limited monitoring capabilities for meteorological elements in upper atmospheres; while meteorological satellites can acquire meteorological information over a wide area, their data resolution is relatively low, making it difficult to accurately capture meteorological details in local areas; radar is ineffective in monitoring non-precipitating cloud systems when monitoring meteorological elements such as precipitation. Furthermore, the data formats, accuracy, and spatiotemporal resolutions collected by different monitoring devices vary, and the lack of effective data fusion methods makes it difficult to comprehensively and accurately reflect the true state of meteorological elements, failing to meet the growing demand for high-precision, comprehensive meteorological monitoring. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a meteorological element monitoring method and system based on multi-source data fusion.

[0004] In conjunction with the first aspect of this application, a meteorological element monitoring method based on multi-source data fusion is provided, applied to a meteorological element monitoring system based on multi-source data fusion, the method comprising:

[0005] Acquire a multi-source meteorological data set, which includes observation data units of meteorological elements with spatiotemporal markers collected by different types of monitoring equipment;

[0006] Feature extraction is performed based on the multi-source meteorological data set to generate a fused meteorological feature set containing multi-element correlation features;

[0007] The fused meteorological feature set is used to identify meteorological element anomalies, thereby determining the distribution area and type of anomalies of meteorological elements within the monitoring area.

[0008] A meteorological monitoring and analysis report containing spatiotemporal positioning information is generated based on the abnormal distribution area and abnormal type.

[0009] In conjunction with the second aspect of this application, a meteorological element monitoring system based on multi-source data fusion is provided. The meteorological element monitoring system based on multi-source data fusion includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the meteorological element monitoring system based on multi-source data fusion implements the aforementioned meteorological element monitoring method based on multi-source data fusion.

[0010] In conjunction with a third aspect of this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, and when the computer-executable instructions are executed, the aforementioned meteorological element monitoring method based on multi-source data fusion is implemented.

[0011] By combining any of the above aspects, and by acquiring a multi-source meteorological data set containing spatiotemporally labeled meteorological element observation data units collected by different types of monitoring equipment, the advantages of various monitoring devices can be fully utilized. Based on this multi-source meteorological data set, feature extraction is performed to generate a fused meteorological feature set containing multi-element correlation features. This achieves effective fusion of data from different sources and of different types, uncovers potential correlations between meteorological elements, and uses the fused meteorological feature set for meteorological element anomaly identification processing. This can accurately determine the distribution area and anomaly type of meteorological elements within the monitoring area, greatly improving the accuracy and comprehensiveness of meteorological element anomaly monitoring. Finally, a meteorological monitoring and analysis report containing spatiotemporal positioning information is generated based on the anomaly distribution area and anomaly type. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained in conjunction with these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating the meteorological element monitoring method based on multi-source data fusion provided in this application embodiment. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0016] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0017] Figure 1 This document illustrates a flowchart of a meteorological element monitoring method based on multi-source data fusion provided in an embodiment of this application. It should be understood that in other embodiments, the order of some steps in the meteorological element monitoring method based on multi-source data fusion may be shared based on actual needs, or some steps may be omitted or maintained. The detailed components of this meteorological element monitoring method based on multi-source data fusion are as follows:

[0018] Step S110: Obtain a multi-source meteorological data set, which includes observation data units of meteorological elements with spatiotemporal markers collected by different types of monitoring equipment.

[0019] This embodiment establishes a vast meteorological monitoring area, similar to a large, comprehensive region, encompassing diverse terrains and landforms. Various meteorological monitoring devices are distributed throughout this area.

[0020] First, there are ground-based meteorological monitoring stations, evenly distributed or distributed according to the terrain throughout the region. These stations are equipped with various instruments to collect meteorological data, such as instruments for measuring temperature, humidity, and wind speed. These instruments operate continuously, recording the corresponding meteorological values. Each recorded data point is accompanied by a precise timestamp, accurate to the minute or even the second, to ensure the data's temporal accuracy. Simultaneously, the data also includes spatial markers, represented by latitude and longitude coordinates, accurate to a certain decimal place, to clearly identify the geographical location of the data collection.

[0021] In addition to ground-based meteorological monitoring stations, there are also upper-air meteorological sounding devices, such as weather balloons. Weather balloons carry various meteorological sensors and rise from the ground, continuously collecting meteorological data at different altitudes during their ascent. Similarly, these data are also marked with the time of collection and spatial location information obtained through a positioning system.

[0022] In addition, the region can be observed using meteorological satellites. These satellites survey the area from space, acquiring large-scale meteorological information through remote sensing technology, and then converting this information into meteorological data with time and spatial markers, which is then transmitted back to ground receiving stations.

[0023] Through data transmission networks, data from ground-based meteorological monitoring stations, upper-air meteorological balloons, and meteorological satellites are all aggregated into a single data processing center. At this center, these meteorological element observation data units, each with temporal and spatial markers from different monitoring devices, are integrated to form a multi-source meteorological data set.

[0024] Step S120: Perform feature extraction based on the multi-source meteorological data set to generate a fused meteorological feature set containing multi-element correlation features.

[0025] Starting with the acquired multi-source meteorological data set, feature extraction is performed to generate a fused meteorological feature set containing multi-element correlation features.

[0026] Step S121: Extract different types of meteorological element data from the multi-source meteorological data set to generate temperature element data subsets, humidity element data subsets, and wind speed element data subsets.

[0027] In the data processing center, a specially written data filtering program is used to process the multi-source meteorological dataset. This program categorizes the data based on the types of meteorological elements they represent. For each data unit in the multi-source meteorological dataset, assuming the data unit's storage format is [time stamp, spatial stamp, temperature value, humidity value, wind speed value, other meteorological element values, etc.], the filtering program iterates through these data units one by one according to predefined rules. When the program scans a data unit, it can identify the temperature value and extract the corresponding time stamp and spatial stamp. These extracted temperature values, time stamps, and spatial stamps are combined into a new data record. The aggregation of numerous such data records forms a subset of the temperature element data.

[0028] Similarly, for generating a subset of humidity element data, the program extracts the humidity value and the corresponding time and space markers from each data unit in the same way, and then combines these data into a subset of humidity element data.

[0029] The generation of wind speed data subsets follows the same process. The filtering procedure extracts wind speed values ​​and related time and spatial markers from each data unit, thereby forming a wind speed data subset. In this way, the data that was originally mixed in the multi-source meteorological dataset is initially classified according to the three important meteorological element types of temperature, humidity, and wind speed, and stored in their respective independent data subsets.

[0030] Step S122: Perform spatiotemporal feature extraction processing on each subset of the element data to generate a single-element feature vector containing spatiotemporal continuous change characteristics.

[0031] For a subset of temperature data, spatiotemporal feature extraction is performed to generate a single-feature temperature feature vector that contains continuous spatiotemporal variation characteristics.

[0032] Step S1221: Divide the subset of element data into multiple time windows according to time order. Each time window contains meteorological element data corresponding to multiple consecutive standardized timestamps.

[0033] Taking a subset of temperature data as an example, the data processing system divides it chronologically. First, a suitable time window length is determined, which is set based on the actual situation and data analysis needs. Let's assume a time window length of T, where T represents a specific time span. Then, starting from the start time of the temperature data subset, multiple time windows are sequentially divided at intervals of T. For example, the first time window contains temperature data corresponding to all standardized timestamps within the time span from the start time to the start time plus T. Each standardized timestamp corresponds to temperature data collected at a specific moment, and these timestamps are evenly distributed at certain time intervals to ensure the temporal continuity and representativeness of the data within each time window.

[0034] Step S1222: For meteorological element data within each time window, calculate its mean, standard deviation, and extreme values ​​in the spatial dimension as spatial statistical characteristic parameters.

[0035] Within each defined time window, spatial analysis is performed on the temperature data. Taking a specific time window as an example, this window contains temperature data collected from multiple different spatial locations, determined by the spatial markers (latitude and longitude coordinates) of the data. To obtain the spatial characteristics of the temperature data within this time window, the average value of these temperature data points is calculated. The sum of all temperature data values ​​within the time window is then divided by the number of data points; the result is the spatial average value of the temperature data within that time window.

[0036] Next, the standard deviation is calculated. The standard deviation measures the spatial dispersion of these temperature data. The calculation process involves first finding the difference between each temperature data point and the mean, then squared these differences, summing them, dividing by the number of data points, and finally taking the square root of the result. The resulting value is the standard deviation.

[0037] Simultaneously, the maximum and minimum values ​​of the temperature data within this time window are identified; these two values ​​are the extreme values. By calculating the mean, standard deviation, and extreme values, the statistical characteristic parameters of the temperature data in the spatial dimension within this time window are obtained. These parameters reflect the spatial distribution of temperature.

[0038] Step S1223: Analyze the changing trends of meteorological element data within adjacent time windows, and calculate the rate and direction of element change within the time interval as time series characteristic parameters.

[0039] Observe two adjacent time windows, taking the first and second time windows as examples. First, determine the time interval between these two time windows, let's say Δt. Compare the average temperature data within the two time windows, and calculate the change in temperature within that time interval Δt. Subtract the average temperature data of the first time window from the average temperature data of the second time window to obtain the change in temperature. Then divide this change by the time interval Δt to obtain the rate of temperature change within that time interval.

[0040] The direction of change is determined by comparing the average temperature data from two time windows. If the average value of the second time window is greater than that of the first time window, the temperature is increasing; conversely, if the average value of the second time window is less than that of the first time window, the temperature is decreasing. In this way, the rate and direction of temperature change within adjacent time windows are obtained, serving as characteristic parameters of the temperature data over time. These parameters reflect the trend of temperature change over time.

[0041] Step S1224: Normalize the spatial statistical feature parameters and time series feature parameters, and then concatenate the normalized spatial statistical feature parameters and time series feature parameters in a preset order to generate a single-element feature vector containing spatiotemporal continuous change characteristics.

[0042] The calculated spatial statistical characteristic parameters (mean, standard deviation, and extreme values) and time series characteristic parameters (rate of change and direction of change) are normalized. The purpose of normalization is to unify these parameters within a defined numerical range for effective subsequent analysis and fusion. For each characteristic parameter, a predefined normalization method is used, such as subtracting the minimum value of each parameter across all time windows, and then dividing by the difference between the maximum and minimum values ​​across all time windows. This maps the value of each parameter to the range of 0 to 1.

[0043] After normalization, the normalized feature parameters are concatenated in a preset order. Assume the preset order is to concatenate spatial statistical feature parameters first, followed by time series feature parameters. First, the three spatial statistical feature parameters—mean, standard deviation, and extreme values—are concatenated. Then, the two time series feature parameters—rate of change and direction of change—are concatenated, thus generating a single-factor temperature feature vector containing multiple values.

[0044] In the same manner, steps S1221 to S1224 are performed on the subsets of humidity and wind speed data to generate single-element feature vectors of humidity and wind speed that contain spatiotemporal continuous variation characteristics, respectively.

[0045] Step S123: Establish a multi-factor correlation analysis model and calculate the correlation coefficients between the feature vectors of temperature, humidity and wind speed.

[0046] After obtaining the single-factor feature vectors of temperature, humidity, and wind speed, a multi-factor correlation analysis model is established to calculate the correlation coefficients between them.

[0047] Step S1231: Align the temperature feature vector, humidity feature vector, and wind speed feature vector according to the timestamp to form a multi-element time series vector group with equal time intervals.

[0048] The generated single-factor feature vectors for temperature, humidity, and wind speed are organized according to their corresponding timestamps. Since each single-factor feature vector is generated based on a different time window, and each time window has a corresponding timestamp, the feature vectors of the three elements are arranged according to the timestamps, ensuring that feature vectors corresponding to the same time window are in the same group. This method forms a multi-factor time series vector group with equal time intervals. For example, for each time window, there is a corresponding single-factor feature vector for temperature, humidity, and wind speed. These vectors are arranged together according to the order of the time windows to form a multi-factor time series vector group with equal time intervals.

[0049] Step S1232: For each timestamp corresponding to a multi-factor time series vector, calculate the point-state correlation index of each factor feature vector in the spatial dimension.

[0050] For each vector group in the multi-factor time series vector set (i.e., the single-factor feature vectors of temperature, humidity, and wind speed corresponding to each timestamp), the correlation between them is analyzed in the spatial dimension. Taking a certain set of vectors as an example, the numerical distribution of the single-factor feature vectors of temperature, humidity, and wind speed in the spatial dimension is observed. By using a defined analysis method, such as comparing the corresponding values ​​of the three-factor feature vectors at each spatial location (determined by spatial markers), the degree of correlation between them is determined. For each spatial location, an index value reflecting the correlation of the three-factor feature vectors at that location is calculated; this index value is the point-state correlation index. By performing the above calculations on all spatial locations, a set of point-state correlation indices reflecting the correlation of the three factors at each spatial location is obtained. These point-state correlation indices describe the degree of association between different factors in the spatial dimension.

[0051] Step S1233: Perform sliding window integration on the point-state correlation index of multiple consecutive timestamps to generate a dynamic correlation coefficient matrix that characterizes the spatiotemporal coupling degree of multiple elements.

[0052] After obtaining the point-state correlation index corresponding to each timestamp, a sliding window integration method is used to further analyze the spatiotemporal coupling relationship of multiple elements. A sliding window of length M is set, where M represents the number of consecutive timestamps. Starting from the first timestamp, the point-state correlation index of M consecutive timestamps is taken. These indices are calculated according to certain integration rules, such as summing the point-state correlation indices at each spatial location within these M timestamps. Through the above integration operation, a numerical value reflecting the comprehensive spatial correlation of multiple elements within these M timestamps is obtained. This calculation is performed for each spatial location, resulting in a set of values. These values ​​form a matrix, which is the dynamic correlation coefficient matrix characterizing the degree of spatiotemporal coupling of multiple elements. As the sliding window moves sequentially along the time series, new dynamic correlation coefficient matrices are continuously generated, allowing observation of the changes in the degree of spatiotemporal coupling of multiple elements over time.

[0053] Step S1234: Determine the time-sensitive parameters of the multi-factor association analysis model based on the fluctuation range of the correlation coefficients between the elements in the dynamic correlation coefficient matrix.

[0054] Observe the fluctuations in correlation coefficients between different elements in the dynamic correlation coefficient matrix. Taking the correlation coefficient between temperature and humidity as an example, in the dynamic correlation coefficient matrix, this correlation coefficient will fluctuate within a certain range over time. Analyze the magnitude and characteristics of this fluctuation range, such as calculating the difference between the maximum and minimum values ​​of the correlation coefficient, or observing the frequency of correlation coefficient fluctuations. Based on these analytical results, determine a time-sensitive parameter that can reflect the sensitivity of the correlation between multiple elements to changes over time. By determining the time-sensitive parameter, the dynamic changes of the correlation between multiple elements over time can be better considered in subsequent calculations.

[0055] Step S1235: Use the time-sensitive parameter to perform weighted correction on the correlation coefficients of different timestamps to obtain the target correlation coefficient that reflects the spatiotemporal dynamic correlation characteristics of multiple elements.

[0056] For the correlation coefficient corresponding to each timestamp in the dynamic correlation coefficient matrix, a weighted correction is performed using the previously determined time-sensitive parameter. Taking the correlation coefficient between temperature and humidity at a certain timestamp as an example, the correlation coefficient is adjusted according to the magnitude of the time-sensitive parameter. If the time-sensitive parameter indicates that the correlation between multiple factors is highly sensitive to time changes within that time period, then a larger weight is given to the correlation coefficient of that timestamp; conversely, if the time-sensitive parameter indicates that the correlation is not very sensitive to time changes, then a smaller weight is given. Through the above weighted correction operation, the correlation coefficient of each timestamp is adjusted, ultimately resulting in a set of target correlation coefficients that more accurately reflect the spatiotemporal dynamic correlation characteristics of multiple factors. These target correlation coefficients comprehensively consider the interrelationships of multiple factors in time and space.

[0057] Step S124: Determine the weight parameters of each single-element feature vector in the fusion process based on the correlation coefficient, wherein the weight parameters are positively correlated with the correlation coefficient.

[0058] After obtaining the target correlation coefficients reflecting the spatiotemporal dynamic correlation characteristics of multiple elements, the weight parameters for the feature vectors of individual elements such as temperature, humidity, and wind speed are determined based on these coefficients during the fusion process. Since the weight parameters are positively correlated with the correlation coefficients, taking the target correlation coefficient between temperature and humidity as an example, a higher target correlation coefficient indicates a stronger spatiotemporal correlation between temperature and humidity elements, thus assigning relatively higher weights to the feature vectors of individual elements such as temperature and humidity during the fusion process; conversely, a lower target correlation coefficient indicates a weaker correlation, thus assigning relatively lower weights. Through this method, based on the target correlation coefficients between each element and other elements, corresponding weight parameters are determined for the feature vectors of individual elements such as temperature, humidity, and wind speed, ensuring that elements with strong correlations play a greater role in the fusion result.

[0059] Step S125: Perform weighted fusion processing on the single-element feature vector according to the weight parameters to generate a fused meteorological feature vector containing the spatiotemporal correlation characteristics of multiple elements, and combine all the fused meteorological feature vectors to generate the fused meteorological feature set.

[0060] Based on the determined weight parameters, the single-element feature vectors of temperature, humidity, and wind speed are weighted and fused. Taking a set of single-element feature vectors as an example, each value of the temperature single-element feature vector is multiplied by its corresponding weight parameter, each value of the humidity single-element feature vector is multiplied by its corresponding weight parameter, and each value of the wind speed single-element feature vector is multiplied by its corresponding weight parameter. Then, these weighted vectors are concatenated in a certain way. For example, the weighted temperature single-element feature vector is placed first, followed by the weighted humidity single-element feature vector, and finally the weighted wind speed single-element feature vector is concatenated to generate a fused meteorological feature vector that contains the spatiotemporal correlation characteristics of multiple elements.

[0061] The aforementioned weighted fusion process is applied to the single-element feature vectors corresponding to all time windows, resulting in multiple fused meteorological feature vectors. These fused meteorological feature vectors are then combined in chronological order to form a fused meteorological feature set. This fused meteorological feature set integrates the spatiotemporal correlation characteristics of multiple elements such as temperature, humidity, and wind speed.

[0062] Step S130: Use the fused meteorological feature set to perform meteorological element anomaly identification processing to determine the distribution area and anomaly type of meteorological elements within the monitoring area.

[0063] Based on the generated fused meteorological feature set, meteorological element anomaly identification processing is initiated to determine the distribution area and anomaly type of meteorological elements within the monitoring area.

[0064] Step S131: Establish a historical fusion meteorological feature database for a preset time period. The historical fusion meteorological feature database contains the fusion meteorological feature vectors of each grid node during the historical period.

[0065] First, a set of historical multi-source meteorological data for a preset time period is collected. This preset time period is set according to actual needs, such as the past year or the past month. These historical multi-source meteorological data sets have the same spatiotemporal coordinate system as the currently monitored multi-source meteorological data sets, ensuring data consistency and comparability.

[0066] Then, the same feature extraction operation as the current monitoring is performed on the collected historical multi-source meteorological data set. That is, following the methods in steps S120 to S125, different types of meteorological element data are extracted from the historical multi-source meteorological data set to generate subsets of temperature, humidity, and wind speed element data. Spatiotemporal feature extraction processing is performed on these subsets to generate single-element feature vectors. A multi-element correlation analysis model is established to calculate correlation coefficients, weight parameters are determined, and weighted fusion processing is performed to generate a historical fused meteorological feature set.

[0067] Next, the historical fused meteorological feature set is categorized and stored according to grid node coordinates and timestamps. The monitoring area is divided into multiple grid nodes, each with specific coordinates. For each historical fused meteorological feature vector, it is stored in the corresponding location based on its corresponding timestamp and the grid node coordinates. This establishes a historical fused meteorological feature database indexed by grid node coordinates.

[0068] To ensure the timeliness of the database, the historical fused meteorological feature database is updated regularly. For example, a time limit is set; when a fused meteorological feature vector exceeds this time limit, it is deleted from the database and replaced with the latest generated fused meteorological feature vector, ensuring that the database always contains the latest and most relevant historical data.

[0069] Step S132: For each fused meteorological feature vector in the fused meteorological feature set at the current moment, calculate its difference degree with the historical fused meteorological feature vector of the corresponding grid node.

[0070] At the current moment, a fused meteorological feature vector is extracted from the fused meteorological feature set. Based on the coordinates of the grid node corresponding to this vector, all fused meteorological feature vectors corresponding to this grid node in the historical period are found in the historical fused meteorological feature database.

[0071] For each fused meteorological feature vector in the current fusion meteorological feature set and the historical database, the degree of difference between them is calculated using a defined analysis method. For example, the values ​​at corresponding positions in the two vectors are compared, and the differences between these values ​​are calculated. Then, a comprehensive analysis is performed on the differences at all positions, such as calculating the sum of squares of the differences or the sum of the absolute values ​​of the differences. This method yields a degree of difference that reflects the degree of difference between the two vectors. By performing the above calculation on each vector in the current fused meteorological feature set, a set of degree of difference values ​​reflecting the degree of difference between the current vector and historical vectors is obtained.

[0072] Step S133: Mark grid nodes with a difference exceeding the anomaly determination threshold as anomaly candidate nodes.

[0073] An anomaly detection threshold is set, determined based on long-term meteorological data research and an understanding of the meteorological characteristics of the monitored area. The calculated difference degree for each grid node is compared with the anomaly detection threshold. If the difference degree of a grid node exceeds the threshold, it indicates a significant change in the meteorological characteristics at that node compared to historical data, and the node is marked as a candidate for anomaly detection. For example, analysis of extensive historical data reveals that when the difference degree exceeds a certain set level, it is often accompanied by abnormal changes in meteorological elements; therefore, this set level is set as the anomaly detection threshold. In this way, by comparing the difference degree and the anomaly detection threshold, grid nodes that may exhibit meteorological anomalies can be preliminarily screened.

[0074] Step S134: Analyze the spatial distribution pattern of the abnormal candidate nodes and identify continuous areas of abnormal candidate nodes as areas of abnormal distribution of meteorological elements.

[0075] First, a spatial adjacency table of grid nodes in the monitoring area is constructed. In this monitoring area, each grid node is spatially adjacent to some surrounding grid nodes. By identifying which other grid nodes are directly adjacent to each grid node, this information is organized into a spatial adjacency table. For example, for a grid node A, its directly adjacent grid nodes are identified as B, C, D, etc., and this information is recorded in the spatial adjacency table. This process is repeated for all grid nodes.

[0076] For each marked anomalous candidate node, the number of anomalous candidate nodes among its direct neighbors is counted according to the spatial adjacency table. For example, for anomalous candidate node X, all its direct neighbors are found by querying the spatial adjacency table, and then the number of these neighbors that are also marked as anomalous candidate nodes is counted. This number is the spatial anomalous clustering index of the anomalous candidate node. In this way, a spatial anomalous clustering index is calculated for each anomalous candidate node, which reflects the degree of spatial clustering of anomalous candidate nodes.

[0077] A spatial clustering threshold is set, which is also derived from the analysis of historical data and patterns of meteorological anomaly distribution in the region. Anomaly candidate nodes whose spatial anomaly clustering index exceeds the spatial clustering threshold, along with their adjacent anomaly candidate nodes, are grouped into the same anomaly clustering unit. For example, if the spatial anomaly clustering index of anomaly candidate node Y is greater than the spatial clustering threshold, then Y and all its directly adjacent nodes marked as anomaly candidate nodes are grouped into one anomaly clustering unit. In this way, numerous anomaly candidate nodes are divided into different anomaly clustering units based on the spatial anomaly clustering index and the spatial clustering threshold.

[0078] Spatial connectivity testing is performed on all predefined anomalous clusters. This involves checking for spatial connectivity between different anomalous clusters, specifically whether adjacent grid nodes allow them to connect. If two or more anomalous clusters are found to be connected, they are merged into a larger region. For example, anomalous clusters M and N are found to have adjacent candidate anomalous nodes, indicating spatial connectivity; therefore, M and N are merged into a new region. Through this spatial connectivity testing and merging process, a complete meteorological element anomalous distribution area is formed.

[0079] Finally, the spatial continuity level of the anomalous distribution area is determined based on the density of anomalous candidate nodes in each grid node within the anomalous distribution area. The total number of anomalous candidate nodes within the anomalous distribution area is calculated, and then divided by the total number of grid nodes in that area to obtain the anomalous candidate node density. Based on this density, the spatial continuity level of the anomalous distribution area is divided into different levels. For example, if the anomalous candidate node density is high, it indicates that the anomalous distribution within the area is relatively continuous, and its spatial continuity level may be set to a higher level; conversely, if the density is low, it is set to a lower level. By determining the spatial continuity level, the characteristics of the anomalous distribution area of ​​meteorological elements can be described in more detail.

[0080] Step S135: Based on the element composition of the fused meteorological feature vector corresponding to the abnormal candidate node, determine the meteorological element anomaly type of the abnormal distribution area. The anomaly type includes one or more combinations of temperature anomaly, humidity anomaly, and wind speed anomaly.

[0081] For each candidate node in an anomaly distribution area, analyze its corresponding fused meteorological feature vector. Since the fused meteorological feature vector is obtained by weighted fusion of the feature vectors of individual elements such as temperature, humidity, and wind speed, the situation of each meteorological element can be inferred from the fused meteorological feature vector.

[0082] For example, examine the temperature-related portion of the fused meteorological feature vector. If this portion shows a significant deviation from historical data—for instance, if certain parameters in the temperature single-element feature vector (such as average value, rate of change, etc.) exceed the historical normal range—it can be determined that there is a temperature anomaly in the abnormal distribution area. Similarly, examine the humidity-related portion. If the parameters in the humidity single-element feature vector are abnormal, such as an excessively high or low average humidity value, or an abnormal rate of humidity change, it indicates a humidity anomaly. The same applies to wind speed; observe the performance of the wind speed single-element feature vector in the fused meteorological feature vector to determine if the wind speed is abnormal.

[0083] Based on these analysis results, the types of meteorological element anomalies in the abnormal distribution areas are determined. If only the temperature-related components are abnormal, the anomaly type is temperature anomaly; if both humidity and wind speed-related components are abnormal, while the temperature is normal, the anomaly type is a combination of humidity and wind speed anomalies; if temperature, humidity, and wind speed-related components all show anomalies, the anomaly type is a combination of all three. Through this detailed analysis, the types of meteorological element anomalies in each abnormal distribution area are accurately determined.

[0084] Step S140: Generate a meteorological monitoring and analysis report containing spatiotemporal positioning information based on the abnormal distribution area and abnormal type.

[0085] Starting from the identified abnormal distribution areas and abnormal types, a meteorological monitoring and analysis report containing spatiotemporal positioning information is generated.

[0086] Step S141: Extract the set of grid node coordinates of the abnormal distribution area, and calculate the spatial coverage and geometric center coordinates of the set of grid node coordinates.

[0087] From each anomalous distribution region, extract the grid node coordinates of all candidate anomalous nodes, and summarize these coordinates to form an anomalous node coordinate set. For example, for a certain anomalous distribution region, record the coordinates of all grid nodes marked as candidate anomalous nodes to form a set.

[0088] Within the set of coordinates for this abnormal node, determine the maximum and minimum horizontal coordinates, as well as the maximum and minimum vertical coordinates. By comparing all horizontal coordinates in the set, find the maximum and minimum values; similarly, compare the vertical coordinates to find the maximum and minimum values.

[0089] The lateral and longitudinal spans of the anomalous distribution area are calculated using the maximum and minimum lateral, maximum and minimum longitudinal coordinates. The lateral span equals the maximum lateral coordinate minus the minimum lateral coordinate, and the longitudinal span equals the maximum longitudinal coordinate minus the minimum longitudinal coordinate. These two span values ​​determine the extent of the anomalous distribution area in the lateral and longitudinal directions, thus generating the rectangular boundary coordinates of the spatial coverage. For example, using the minimum lateral and minimum longitudinal coordinates as the coordinates of one vertex of a rectangle, and the maximum lateral and maximum longitudinal coordinates as the coordinates of the opposite vertex, a rectangular boundary is determined, which describes the approximate spatial coverage of the anomalous distribution area.

[0090] The arithmetic mean of all horizontal coordinates in the set of anomaly node coordinates is used to obtain the horizontal coordinate of the geometric center of the anomaly distribution area. Specifically, this is calculated by adding all horizontal coordinate values ​​and then dividing by the number of coordinates. Similarly, the arithmetic mean of all vertical coordinates is used to obtain the vertical coordinate of the geometric center. Combining these two coordinates yields the geometric center coordinate of the anomaly distribution area.

[0091] Finally, the coordinates of the rectangular boundary and the geometric center are combined to generate a spatial location description that includes the boundary extent and center position. This spatial location description clearly indicates the specific location of the anomalous distribution area within the monitoring area.

[0092] Step S142: Determine the corresponding meteorological element anomaly level according to the anomaly type, and the anomaly level is divided into different levels according to the degree of difference.

[0093] Based on the anomaly types identified in the previously determined anomaly distribution areas, the anomaly level is determined by combining the degree of difference between each anomaly candidate node and the historical fused meteorological feature vector. Corresponding anomaly level classification rules are formulated for different anomaly types.

[0094] Taking temperature anomalies as an example, if the temperature feature vector corresponding to an anomaly candidate node has a small difference from the historical vector, the temperature anomaly level is classified as low within a pre-defined low range, indicating that the temperature anomaly is relatively mild. As the difference increases, if it exceeds the low range but is within another medium range, the temperature anomaly level is raised to medium, indicating that the temperature anomaly has worsened. When the difference further increases, exceeding the medium range and reaching a high range, the temperature anomaly level is set to high, meaning that the temperature anomaly is quite severe.

[0095] A similar approach is used for humidity and wind speed anomalies, classifying them into different anomaly levels within a defined range based on the degree of difference between their individual feature vectors and historical vectors. This method of classifying anomalies according to the magnitude of difference allows for a more accurate description of the severity of different anomaly types.

[0096] Step S143: Collect the real-time fused meteorological feature vectors and historical fused meteorological feature vectors of each grid node in the abnormal distribution area to generate an abnormal feature comparison data set.

[0097] For each anomalous distribution area, all grid nodes within it are traversed. For each grid node, the real-time fused meteorological feature vector of that node is obtained from the current fused meteorological feature set, and the corresponding fused meteorological feature vector of that node in the historical period is obtained from the historical fused meteorological feature database.

[0098] For example, for a grid node P within an anomalous distribution area, the real-time fused meteorological feature vector corresponding to P is found from the current fused meteorological feature set. This real-time fused meteorological feature vector reflects the multi-element meteorological characteristics of the node at the current moment. Then, from the historical fused meteorological feature database, based on the grid node coordinates and timestamp of P, the fused meteorological feature vector of the node for the corresponding historical time period is extracted.

[0099] The real-time fused meteorological feature vector and the historical fused meteorological feature vector of each grid node are combined to form a data pair. All data pairs from all grid nodes are collected to form an anomaly feature comparison dataset, which contains comparative information on the meteorological characteristics of each grid node within the anomaly distribution area at the current moment and in historical periods.

[0100] Step S144: Analyze the changing trends of elements in the abnormal feature comparison data set, and generate an abnormal process description including the abnormal start time, duration, and evolution direction.

[0101] In the anomaly feature comparison dataset, the changing trends of various meteorological elements are analyzed using time as a clue. For each grid node data pair, the differences in the characteristics of each element in the real-time fused meteorological feature vector and the historical fused meteorological feature vector are observed over time.

[0102] For example, by comparing parameters (such as average value and rate of change) in the single-factor feature vector of temperature at different time points, the trend of temperature change can be determined. If it is found that the average temperature has been rising continuously since a certain time point, and this upward trend continues in subsequent time points, and combined with historical data, it is determined that this rise exceeds the normal fluctuation range, then the starting time of the temperature anomaly can be determined as the time point when the abnormal rise begins. By continuing to observe the data at subsequent time points until the temperature returns to a relatively normal range or other significant changes occur, the duration of the anomaly can be determined.

[0103] The direction of evolution is determined by analyzing the changes in parameters within the single-element eigenvector of temperature. If the rate of temperature change remains positive, it indicates that the temperature is continuously rising, and the direction of evolution is upward. If the rate of change changes from positive to negative, it indicates that the temperature first rises and then falls, and the direction of evolution has changed.

[0104] Similarly, the same analysis was performed on humidity and wind speed elements to determine the onset time, duration, and evolution direction of humidity and wind speed anomalies, respectively. This information on the onset time, duration, and evolution direction of temperature, humidity, and wind speed anomalies was then integrated to generate a comprehensive textual description of the anomaly process, detailing its development across various meteorological elements.

[0105] Step S145: Integrate the spatial coverage area, geometric center coordinates, anomaly level, and anomaly process description according to a preset report format to generate a meteorological monitoring and analysis report containing spatiotemporal positioning information and anomaly feature analysis.

[0106] First, define the pre-defined report format. This format specifies the order and manner in which the spatial coverage area, geometric center coordinates, anomaly level, and anomaly process description are presented in the report. For example, it might require listing the rectangular boundary coordinates of the spatial coverage area first, followed by the geometric center coordinates, then the anomaly level, and finally a detailed description of the anomaly process.

[0107] Following the preset report format, integrate all previously calculated and generated information. Accurately fill in the rectangular boundary coordinates of the spatial coverage area into the corresponding positions in the report, followed by the geometric center coordinates. Then, based on the determined anomaly type and anomaly level classification results, clearly describe the corresponding anomaly level information in the report. Finally, fully enter the generated anomaly process description, including the anomaly start time, duration, and evolution direction, into the report.

[0108] Through the above integration process, a meteorological monitoring and analysis report is generated, which includes spatiotemporal positioning information (representing spatial coverage and geometric center coordinates) and anomaly characteristic analysis (representing anomaly level and anomaly process description). This report comprehensively and in detail presents relevant information on meteorological element anomalies within the monitoring area, providing valuable reference materials for meteorological researchers, policymakers, and relevant departments, enabling them to better understand meteorological anomalies and make corresponding decisions and countermeasures.

[0109] In the above embodiments, the meteorological element monitoring system based on multi-source data fusion for executing the above method embodiments has at least one processor, a control module (chipset) coupled to at least one of the processors, a memory coupled to the control module, a non-volatile memory (NVM) / storage device coupled to the control module, at least one load to / output device coupled to the control module, and a network interface coupled to the control module.

[0110] The processor may include at least one single-core or multi-core processor, and may include any combination of general-purpose processors or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). For some alternative implementations, a meteorological element monitoring system based on multi-source data fusion can serve as an electronic device such as the gateway described in the embodiments of this application.

[0111] In some alternative implementations, a meteorological element monitoring system based on multi-source data fusion may include at least one computer-readable medium (e.g., a memory or NVM / storage device) having instructions and at least one processor fused with the at least one computer-readable medium and configured to execute the instructions to implement the module thereby performing the actions described in this disclosure.

[0112] In one embodiment, the control module may include any suitable interface controller to provide any suitable interface to at least one of the processors and / or any suitable device or component communicating with the control module.

[0113] The control module may include a memory controller module to provide an interface to the memory. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0114] The memory can be used, for example, to load and store data and / or instructions for a meteorological element monitoring system based on multi-source data fusion. In one embodiment, the memory may include any suitable volatile memory, such as suitable DRAM.

[0115] In one embodiment, the control module may include at least one load-to-output controller to provide an interface to the NVM / storage device and (at least one) load-to-output device.

[0116] For example, an NVM / storage device can be used to store data and / or instructions. An NVM / storage device may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable (at least one) non-volatile storage device (e.g., at least one hard disk drive (HDD), at least one optical disc (CD) drive, and / or at least one digital universal optical disc (DVD) drive).

[0117] NVM / storage devices may include storage resources that are physically part of a device installed on a meteorological element monitoring system based on multi-source data fusion, or that can be accessed by the device without being part of it. For example, NVM / storage devices may be accessed over a network via (at least one) load to / output device.

[0118] At least one loading / output device may provide an interface for the meteorological element monitoring system based on multi-source data fusion to communicate with any other suitable device. The loading / output device may include communication components, pinyin components, sensor components, etc. A network interface may provide an interface for the meteorological element monitoring system based on multi-source data fusion to communicate via at least one network. The meteorological element monitoring system based on multi-source data fusion may wirelessly communicate with at least one component of a wireless network based on at least one wireless network prior and / or any prior and / or protocol, such as accessing a wireless network based on communication priors.

[0119] In one embodiment, at least one of the processors may be integrated with the logic of at least one controller of the control module (e.g., a memory controller module). In one embodiment, at least one of the processors may be integrated with the logic of at least one controller of the control module to generate a system-level load. In one embodiment, at least one of the processors may be fused with the logic of at least one controller of the control module on the same die. In one embodiment, at least one of the processors may be fused with the logic of at least one controller of the control module on the same die to generate a system-on-a-chip (SoC).

[0120] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0121] This invention discloses a computer read storage medium that stores a computer program for electronic data exchange, wherein the computer program causes a computer to execute the steps in the meteorological element monitoring method based on multi-source data fusion described in the foregoing embodiments.

[0122] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the meteorological element monitoring method based on multi-source data fusion described in the foregoing embodiments.

[0123] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any inventive effort.

[0124] Finally, it should be noted that the above-disclosed embodiments are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring meteorological elements based on multi-source data fusion, characterized in that, The method includes: Acquire a multi-source meteorological data set, which includes observation data units of meteorological elements with spatiotemporal markers collected by different types of monitoring equipment; Feature extraction is performed based on the multi-source meteorological data set to generate a fused meteorological feature set containing multi-element correlation features; The fused meteorological feature set is used to identify meteorological element anomalies, thereby determining the distribution area and type of anomalies of meteorological elements within the monitoring area. A meteorological monitoring and analysis report containing spatiotemporal positioning information is generated based on the abnormal distribution area and abnormal type. The step of performing feature extraction based on the multi-source meteorological data set to generate a fused meteorological feature set containing multi-element correlation features includes: Different types of meteorological element data are extracted from the multi-source meteorological data set to generate temperature element data subsets, humidity element data subsets, and wind speed element data subsets; Spatiotemporal feature extraction processing is performed on each subset of the element data to generate a single-element feature vector containing continuous spatiotemporal variation characteristics; Establish a multi-factor correlation analysis model and calculate the correlation coefficients between the feature vectors of temperature, humidity and wind speed. The weight parameters of each single-element feature vector in the fusion process are determined based on the correlation coefficient, and the weight parameters are positively correlated with the correlation coefficient. The single-element feature vectors are weighted and fused according to the weight parameters to generate a fused meteorological feature vector containing the spatiotemporal correlation characteristics of multiple elements. All fused meteorological feature vectors are combined to generate the fused meteorological feature set. The establishment of a multi-factor correlation analysis model, and the calculation of the correlation coefficients between the feature vectors of the temperature element, humidity element, and wind speed element, include: The temperature feature vector, humidity feature vector, and wind speed feature vector are aligned according to timestamps to form a multi-element time series vector group with equal time intervals. For each timestamp, calculate the point-state correlation index of each feature vector in the spatial dimension; A sliding window integral is performed on the point-state correlation index of multiple consecutive time stamps to generate a dynamic correlation coefficient matrix that characterizes the spatiotemporal coupling degree of multiple factors. Based on the fluctuation range of the correlation coefficients between each element in the dynamic correlation coefficient matrix, the time-sensitive parameters of the multi-factor association analysis model are determined. By using the time-sensitive parameter to weight and correct the correlation coefficients of different timestamps, a target correlation coefficient reflecting the spatiotemporal dynamic correlation characteristics of multiple elements is obtained. The process of using the fused meteorological feature set to identify meteorological element anomalies and determine the distribution area and type of anomalies within the monitoring area includes: A historical fusion meteorological feature database for a preset time period is established, wherein the historical fusion meteorological feature database contains the fusion meteorological feature vectors of each grid node within the historical period; For each fused meteorological feature vector in the fused meteorological feature set at the current moment, calculate its difference degree with the historical fused meteorological feature vector of the corresponding grid node; Grid nodes whose difference exceeds the anomaly determination threshold are marked as anomaly candidate nodes; Analyze the spatial distribution pattern of the anomalous candidate nodes and identify continuous areas of anomalous candidate nodes as areas of anomalous distribution of meteorological elements. Based on the composition of the fused meteorological feature vector corresponding to the abnormal candidate node, the meteorological element anomaly type of the abnormal distribution area is determined. The anomaly type includes one or more combinations of temperature anomaly, humidity anomaly, and wind speed anomaly.

2. The meteorological element monitoring method based on multi-source data fusion according to claim 1, characterized in that, The step of performing spatiotemporal feature extraction processing on each subset of element data to generate a single-element feature vector containing spatiotemporally continuously changing characteristics includes: The subset of element data is divided into multiple time windows according to time order, and each time window contains meteorological element data corresponding to multiple consecutive standardized timestamps. For meteorological element data within each time window, calculate its mean, standard deviation, and extreme values ​​in the spatial dimension as spatial statistical characteristic parameters; Analyze the changing trends of meteorological element data within adjacent time windows, and calculate the rate and direction of element change within the time interval as time series characteristic parameters; The spatial statistical feature parameters and time series feature parameters are normalized, and the normalized spatial statistical feature parameters and time series feature parameters are concatenated in a preset order to generate a single-element feature vector containing spatiotemporal continuous change characteristics.

3. The meteorological element monitoring method based on multi-source data fusion according to claim 1, characterized in that, The establishment of a historical fused meteorological feature database for a preset time period, wherein the database contains fused meteorological feature vectors of each grid node within the historical period, including: Collect a set of historical multi-source meteorological data within a preset time period, wherein the set of historical multi-source meteorological data and the set of currently monitored multi-source meteorological data have the same spatiotemporal coordinate system; Perform the same feature extraction as the current monitoring on the historical multi-source meteorological data set to generate a historical fused meteorological feature set; The historical fused meteorological feature set is classified and stored according to grid node coordinates and timestamps, and a historical fused meteorological feature database is established with grid node coordinates as the index. The historical fused meteorological feature database is updated regularly, deleting historical fused meteorological feature vectors that have exceeded a preset time period and supplementing them with the latest fused meteorological feature vectors.

4. The meteorological element monitoring method based on multi-source data fusion according to claim 3, characterized in that, The analysis of the spatial distribution patterns of the anomalous candidate nodes, identifying contiguous areas of anomalous candidate nodes as areas of anomalous meteorological element distribution, includes: Construct a spatial adjacency table for grid nodes in the monitoring area, wherein the spatial adjacency table defines the set of directly adjacent nodes for each grid node; For each abnormal candidate node, count the number of abnormal candidate nodes in its direct neighboring nodes to generate a spatial anomaly clustering index. Abnormal candidate nodes whose spatial abnormal clustering index exceeds the spatial clustering threshold and their adjacent abnormal candidate nodes are classified into the same abnormal clustering unit. Spatial connectivity detection is performed on all anomalous clusters, and interconnected anomalous clusters are merged to form a complete meteorological element anomalous distribution area. The spatial continuity level of the abnormal distribution area is determined based on the density of abnormal candidate nodes of each grid node within the abnormal distribution area.

5. The meteorological element monitoring method based on multi-source data fusion according to claim 1, characterized in that, The process of generating a meteorological monitoring and analysis report containing spatiotemporal positioning information based on the abnormal distribution area and abnormal type includes: Extract the set of grid node coordinates for the abnormal distribution area, and calculate the spatial coverage and geometric center coordinates of the set of grid node coordinates. The anomaly type is used to determine the corresponding meteorological element anomaly level, and the anomaly level is divided into different levels according to the degree of difference. Collect real-time fused meteorological feature vectors and historical fused meteorological feature vectors of each grid node within the abnormal distribution area to generate an abnormal feature comparison data set; Analyze the changing trends of elements in the comparison dataset of the aforementioned abnormal features to generate an abnormal process description that includes the start time, duration, and direction of evolution of the anomaly; The spatial coverage area, geometric center coordinates, anomaly level, and anomaly process description are integrated according to a preset report format to generate a meteorological monitoring and analysis report containing spatiotemporal positioning information and anomaly feature analysis.

6. The meteorological element monitoring method based on multi-source data fusion according to claim 5, characterized in that, The step of extracting the set of grid node coordinates for the abnormal distribution area and calculating the spatial coverage and geometric center coordinates of the set of grid node coordinates includes: Extract the grid node coordinates of all candidate abnormal nodes from the abnormal distribution area to generate a set of abnormal node coordinates; Determine the maximum horizontal coordinate, minimum horizontal coordinate, maximum vertical coordinate, and minimum vertical coordinate in the set of abnormal node coordinates; The horizontal and vertical spans of the abnormal distribution area are calculated based on the maximum horizontal coordinate, minimum horizontal coordinate, maximum vertical coordinate, and minimum vertical coordinate, and the rectangular boundary coordinates of the spatial coverage area are generated. The arithmetic mean of all the horizontal and vertical coordinates in the set of abnormal node coordinates is used to obtain the horizontal and vertical coordinates of the geometric center of the abnormal distribution area. The rectangular boundary coordinates and geometric center coordinates are combined to generate a spatial positioning information description that includes the boundary range and center position.

7. A meteorological element monitoring system based on multi-source data fusion, characterized in that, The method includes a processor and a computer-readable storage medium storing machine-executable instructions, which, when executed by a computer, implement the meteorological element monitoring method based on multi-source data fusion as described in any one of claims 1-6.

Images