Regional meteorological characteristic analysis system and method based on multi-source data fusion

Abstract

The invention discloses a regional meteorological characteristic analysis system and method based on multi-source data fusion, and relates to the technical field of meteorological monitoring and analysis, and the method comprises the steps: a data set collection module collects a historical meteorological analysis data set of a station; the trend analysis module analyzes the data set to generate a meteorological characteristic trend factor set; the monitoring data set acquisition module acquires a real-time meteorological monitoring data set; the time sequence difference analysis module analyzes the time sequence difference; the correction meteorological monitoring data set generation module processes the real-time meteorological monitoring data set; and the regional meteorological characteristic analysis result generation module analyzes the meteorological characteristic trend factor set and the correction meteorological monitoring data set to generate a regional meteorological characteristic analysis result of the target region. The technical problem of inaccurate meteorological characteristic analysis caused by time sequence errors of real-time monitoring data in the existing meteorological analysis technology is solved, and the technical effect of improving the accuracy and reliability of regional meteorological characteristic analysis is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of meteorological monitoring and analysis, and in particular to a regional meteorological characteristic analysis system and method based on multi-source data fusion. Background Art

[0002] In the field of meteorological monitoring and analysis, the impact of meteorological information on many industries is becoming increasingly critical, and the demand for accuracy and timeliness of meteorological data is becoming increasingly prominent. Accurate grasp of regional meteorological characteristics has become an important basis for many decisions. Traditional meteorological monitoring and analysis methods have obvious defects. The data source is usually relatively single, and may rely on only a small number of monitoring stations, resulting in an insufficiently comprehensive and accurate portrayal of regional meteorological characteristics. Moreover, in real-time meteorological monitoring, due to differences in monitoring equipment and interference from environmental factors, data has time series differences, which can easily introduce errors. In addition, meteorological data analysis often focuses on the current situation, and insufficient exploration of historical meteorological characteristics and trends makes it difficult to capture the laws and internal mechanisms of meteorological changes.

[0003] At the current stage, there is a technical problem in the relevant technologies that meteorological analysis technology has inaccurate meteorological characteristic analysis due to the single data source and timing errors in real-time monitoring data. Summary of the invention

[0004] The present application provides a regional meteorological feature analysis system and method based on multi-source data fusion, wherein a data set acquisition module is used to collect historical meteorological analysis data sets (including station location identifiers) of K stations in a target area; a trend analysis module analyzes these data sets to generate a meteorological feature trend factor set; a monitoring data set acquisition module calls the monitoring sensor array of the K stations to acquire a real-time meteorological monitoring data set; a timing difference analysis module traverses the sensor array to analyze timing differences to generate a delay coefficient set; a correction meteorological monitoring data set generation module processes the real-time meteorological monitoring data set based on the delay coefficient set to generate a correction meteorological monitoring data set; a regional meteorological feature analysis result generation module uses a feature analysis unit to analyze the meteorological feature trend factor set and the correction meteorological monitoring data set to generate a regional meteorological feature analysis result of the target area, thereby achieving the technical effect of improving the accuracy and reliability of regional meteorological feature analysis through multi-source data fusion, timing difference correction and historical meteorological trend analysis, and providing strong support for accurate meteorological prediction, optimal resource allocation and risk reduction.

[0005] The present application provides a regional meteorological characteristics analysis system based on multi-source data fusion, comprising:

[0006] A data set acquisition module, the data set acquisition module is used to collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers; a trend analysis module, the trend analysis module is used to perform meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively, and generate K meteorological characteristic trend factor sets; a monitoring data set acquisition module, the monitoring data set acquisition module is used to call the K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring, and obtain K real-time meteorological monitoring data sets; a time series difference analysis module, the time series difference analysis module is used to traverse the K monitoring sensor arrays Perform sensor timing difference analysis to generate K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor; a correction meteorological monitoring data set generation module, the correction meteorological monitoring data set generation module is used to perform pre-sequence data asynchronous extraction analysis on the K real-time meteorological monitoring data sets based on the K delay coefficient sets, and generate K correction meteorological monitoring data sets; a regional meteorological feature analysis result generation module, the regional meteorological feature analysis result generation module is used to use a feature analysis unit to perform meteorological feature analysis on the K meteorological feature trend factor sets and the K correction meteorological monitoring data sets, and generate regional meteorological feature analysis results for the target area.

[0007] The present application provides a method for analyzing regional meteorological characteristics based on multi-source data fusion, including:

[0008] Collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers; perform meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively, and generate K meteorological characteristic trend factor sets; call K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring, and obtain K real-time meteorological monitoring data sets; traverse the K monitoring sensor arrays to perform sensor timing difference analysis, and generate K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor; perform asynchronous extraction analysis of pre-sequence data on the K real-time meteorological monitoring data sets based on the K delay coefficient sets, and generate K deviation-correcting meteorological monitoring data sets; use a feature analysis unit to perform meteorological characteristic analysis on the K meteorological characteristic trend factor sets and the K deviation-correcting meteorological monitoring data sets, and generate regional meteorological characteristic analysis results of the target area.

[0009] A regional meteorological feature analysis system and method based on multi-source data fusion proposed in this application, first, a data set acquisition module acquires a historical meteorological analysis data set of K stations in the target area; a trend analysis module analyzes these data sets to generate a meteorological feature trend factor set; a monitoring data set acquisition module calls the monitoring sensor array of K stations to acquire a real-time meteorological monitoring data set; a timing difference analysis module traverses the sensor array to analyze the timing difference and generates a delay coefficient set; a correction meteorological monitoring data set generation module processes the real-time meteorological monitoring data set based on the delay coefficient set to generate a correction meteorological monitoring data set; a regional meteorological feature analysis result generation module uses a feature analysis unit to analyze the meteorological feature trend factor set and the correction meteorological monitoring data set to generate a regional meteorological feature analysis result of the target area, and through multi-source data fusion, timing difference correction and historical meteorological trend analysis, the accuracy and reliability of regional meteorological feature analysis are improved, and a technical effect of providing strong support for accurate meteorological prediction, optimal resource allocation and risk reduction is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic diagram of the structure of a regional meteorological characteristics analysis system based on multi-source data fusion provided in an embodiment of the present application;

[0012] Figure 2 A flowchart of a method for analyzing regional meteorological characteristics based on multi-source data fusion is provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: data set acquisition module 10 , trend analysis module 20 , monitoring data set acquisition module 30 , time series difference analysis module 40 , deviation correction meteorological monitoring data set generation module 50 , regional meteorological characteristic analysis result generation module 60 . DETAILED DESCRIPTION

[0014] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0017] The present application embodiment provides a regional meteorological characteristics analysis system based on multi-source data fusion, such as Figure 1 As shown, the system comprises:

[0018] The data set acquisition module 10 is used to collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers. Specifically, the data set acquisition module 10 defines the target area, and the demarcation of the area is based on specific needs and research purposes. For example, if the meteorological research is aimed at a certain city, the target area will be set to the city and a certain range around it. In the target area, there are K stations, and the distribution of the stations is planned and selected to ensure that the meteorological conditions of the area can be fully and accurately reflected. Each station has a unique location identifier, which helps to accurately locate and classify the collected data. For example, the location identifier may include the latitude and longitude coordinates of the station, the name of the administrative division where it is located, or a specific code. When collecting data, the data set acquisition module 10 obtains the historical meteorological analysis data set from the K stations. The data set includes Contains rich meteorological element information, such as temperature, humidity, air pressure, wind speed, wind direction, rainfall, etc. The data collection collection module 10 communicates and interacts with the meteorological monitoring equipment and data storage system of each station, using specific interface protocols, data format conversion and other technical means to ensure that various types of data from different stations can be read and integrated smoothly. The collected data will be sorted and stored to form a structured K historical meteorological analysis data set. For example, if the meteorological conditions of an agricultural production area are studied, then the K stations in the target area may be distributed in different farmlands, orchards and other key locations. The collected historical meteorological data set can help analyze the impact of different seasons and different weather conditions on crop growth in the past, thereby providing decision support for agricultural production.

[0019] The trend analysis module 20 is used to perform meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively, and generate K meteorological characteristic trend factor sets. Specifically, the trend analysis module 20 conducts in-depth and detailed analysis on the K historical meteorological analysis data sets, and accurately identifies the key points of meteorological characteristics such as temperature, irradiance, humidity, and rainfall. With the help of advanced data analysis and processing technology, the core components of each meteorological characteristic are extracted, and their evolution laws and mutual correlations in the time dimension are observed. Based on the results of this in-depth analysis, the development trend of meteorological characteristics is clarified, and K meteorological characteristic trend factor sets are generated. As far as temperature is concerned, the trend analysis module 20 carefully sorts out the temperature values ​​at different times, calculates the mean, extreme values ​​and other key indicators in a specific period, and explores its fluctuations in seasons and years. For example, through rigorous analysis, the subtle changes in winter temperature in a certain area in the past few years are detected. For irradiance, the trend analysis module 20 focuses on the dynamics of solar radiation intensity in different periods, and examines the differences in irradiance at different times of the day, in different seasons, and even in different years. For example, the analysis revealed a unique evolution pattern of irradiance in spring and autumn in a certain region. For humidity, the trend analysis module 20 rigorously examined the fluctuations of humidity values, analyzing the difference between morning and evening within a day, the humidity range in different seasons, and the increase and decrease trend of humidity over time. For example, it was found that the humidity in a certain inland area changed significantly in a specific month. As for rainfall, the module 20 carefully counted the rainfall data in different months, seasons and years, paying attention to the changes in the peak, valley and average rainfall. For example, to determine the obvious changing trend of rainfall in a specific area in summer within a few years, through careful analysis of meteorological characteristics, the trend analysis module 20 uses complex mathematical algorithms and statistical models to extract the core trend information of each meteorological feature, and finally creates a set of K meteorological characteristic trend factors. For example, for a specific station, its temperature trend factor may indicate the future temperature rise and fall trend, the irradiance trend factor shows the possible change trend of irradiance in a specific season, the humidity trend factor suggests the potential law of humidity fluctuation, and the precipitation trend factor indicates the possible increase or decrease trend of precipitation in a specific period. The trend factor set provides crucial reference for subsequent meteorological forecasts, disaster prevention and related decision-making and deployment.

[0020] In a possible implementation, the trend analysis module 20 includes: an indicator feature sequence set generation unit, which is used to cluster and identify the K historical meteorological analysis data sets according to a preset meteorological indicator set to generate K historical meteorological indicator feature sequence sets. Specifically, the indicator feature sequence set generation unit analyzes the K historical meteorological analysis data sets one by one according to a preset set of rich meteorological indicators such as temperature, irradiance, humidity, rainfall, wind speed, etc. For each historical meteorological analysis data set, it is accurately classified strictly according to each meteorological indicator. For example, for the temperature indicator, all temperature-related data are accurately extracted from the data set to form an independent temperature indicator feature sequence. For other indicators such as irradiance and humidity, such precise separation operations are also performed, and finally a unique feature sequence is generated for each meteorological indicator, thereby obtaining K historical meteorological indicator feature sequence sets.

[0021] A change curve set construction unit is used to construct K historical meteorological indicator characteristic change curve sets based on the K historical meteorological indicator characteristic sequence sets. Specifically, the change curve set construction unit sequentially obtains the K historical meteorological indicator characteristic sequence sets generated previously, and takes the temperature indicator characteristic sequence set as an example, with time as the horizontal axis and the temperature value as the vertical axis, and carefully draws a curve that can clearly reflect the temperature change over time. For other indicator characteristic sequence sets such as irradiance, humidity, rainfall, wind speed, etc., the same standard method is used to construct their respective corresponding change curves, thereby forming K complete historical meteorological indicator characteristic change curve sets.

[0022] A trend analysis unit is used to perform trend analysis on the K historical meteorological indicator characteristic change curve sets to generate the K meteorological characteristic trend factor sets. Specifically, the trend analysis unit processes the K historical meteorological indicator characteristic change curve sets constructed by the change curve set construction unit. When analyzing the trend strength, for each curve, the slope of the trend line is obtained through rigorous calculation. For example, if the slope of a temperature change curve is large, it indicates that the temperature change trend is significant; if the slope is small, it means that the temperature change is relatively stable. When examining the change in trend speed, the acceleration is accurately calculated. If the acceleration of a curve is positive and large, it means that the trend speed is accelerating; if the acceleration is negative, it means that the trend speed is slowing down. Finally, K meteorological characteristic trend factor sets are successfully generated to provide strong support for subsequent work. For example, assuming that the study is about the meteorology of a certain mountainous area, the indicator characteristic sequence set generation unit classifies the historical meteorological data of the mountainous area into sequence sets according to various indicators, and the change curve set construction unit draws the change curves of temperature, humidity, etc. based on this. The trend analysis unit analyzes the curve. For example, if it is found that the slope of the temperature curve is large and the acceleration is positive, it indicates that the temperature in the mountainous area is showing a significant and accelerating upward trend; while the slope of the humidity curve is small and the acceleration is close to zero, indicating that the humidity changes are relatively stable.

[0023] In a possible implementation, the trend analysis unit includes: a historical curve slope set generation subunit, which is used to traverse the K historical meteorological indicator characteristic change curve sets to perform slope analysis and generate K historical curve slope sets. Specifically, the historical curve slope set generation subunit performs a detailed slope analysis on the K historical meteorological indicator characteristic change curve sets one by one, and for each historical meteorological indicator characteristic change curve, by using a specific mathematical algorithm and model, accurately calculates the slope value of the curve at different stages. After a comprehensive analysis of all K curve sets, K historical curve slope sets are finally generated. For example, when analyzing the temperature change curve, the slope value corresponding to the steepness of the temperature rise or fall in different time periods can be obtained.

[0024] The historical curve acceleration set generation subunit is used to traverse the K historical meteorological indicator characteristic change curve sets to perform curve acceleration analysis and generate K historical curve acceleration sets. Specifically, the historical curve acceleration set generation subunit traverses the K historical meteorological indicator characteristic change curve sets to perform curve acceleration analysis. Through analysis, the acceleration of each curve at different positions is determined. After processing all the curves, K historical curve acceleration sets are successfully generated. For example, for the humidity change curve, the acceleration value corresponding to the increase or decrease of its change speed can be calculated.

[0025] The weighted analysis subunit is used to perform weighted analysis on the K historical curve slope sets and the K historical curve acceleration sets to generate the K meteorological characteristic trend factor sets. Specifically, the weighted analysis subunit obtains the K historical curve slope sets and the K historical curve acceleration sets generated previously, and uses the pre-set weight allocation rules and analysis methods to perform a comprehensive weighted analysis on the data in these two sets, fully considering the different degrees of influence of slope and acceleration on meteorological characteristic trends, and finally generates K meteorological characteristic trend factor sets that can accurately reflect meteorological characteristic trends. For example, suppose we study the meteorological conditions of a certain city. The historical curve slope set generation subunit calculates the slope sets of temperature, humidity and other curves. The historical curve acceleration set generation subunit obtains the corresponding acceleration set. The weighted analysis subunit combines these data, such as giving a higher weight to the slope of the temperature curve and a lower weight to the acceleration of the humidity curve. After weighted calculation, a factor set that can reflect the meteorological trend of the city is generated, providing a strong basis for meteorological forecasting and decision-making.

[0026] The monitoring data set acquisition module 30 is used to call the K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring and obtain K real-time meteorological monitoring data sets. Specifically, the monitoring data set acquisition module 30 analyzes the K stations, accurately identifies the unique attributes of each station, determines the key points such as the location distribution and environmental conditions of the stations, and calls the K monitoring sensor arrays equipped with the K stations. When the call instruction is started, the sensor array immediately starts working to monitor meteorological elements in real time. The temperature sensor accurately senses the subtle fluctuations of the current temperature, the humidity sensor keenly captures the instantaneous value of humidity, the wind speed sensor quickly records the strength and direction of the wind speed, and the air pressure sensor accurately measures the changes in air pressure. The sensors work together to continuously transmit the monitored data back. The monitoring data set acquisition module 3 0 conducts fine integration and efficient processing of data from different sensors and different stations to construct K real-time meteorological monitoring data sets, each of which clearly presents the real-time status of multiple meteorological elements such as temperature, humidity, wind speed, and air pressure. Assuming that the focus is on K meteorological stations around a specific area, the monitoring data set acquisition module 30 can quickly and accurately collect real-time meteorological data from the stations, providing the latest and most timely information support for subsequent analysis, judgment and decision-making. For example, in the field of transportation, the monitoring data set acquisition module 30 can control the meteorological conditions of different sections of the road in real time, providing an accurate basis for traffic diversion, travel arrangements, etc.

[0027] The timing difference analysis module 40 is used to traverse the K monitoring sensor arrays to perform sensor timing difference analysis and generate K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor. Specifically, the timing difference analysis module 40 analyzes the K monitoring sensor arrays, obtains the characteristics of each sensor array, clarifies the time performance of each sensor in the data collection process, traverses the K monitoring sensor arrays, and analyzes the time sequence and interval regularity of each sensor's data collection. In view of the influence of multiple factors, such as the performance differences of the sensors themselves, the delays in the data signal transmission process, the interference of the surrounding environment, etc., different sensors are likely to have deviations in the time nodes of data collection. The timing difference analysis module 40 determines the delay status of each sensor relative to the standard time or a specific reference time by analyzing and determining the delay status of each sensor for each monitoring sensor. Sensor, accurately calculate the corresponding delay coefficient, the delay coefficient truthfully reflects the time delay degree of the sensor during data collection, successfully generate K delay coefficient sets, each set is closely associated with a specific monitoring sensor, for example, assuming that in a meteorological monitoring system, there are four sensors located in different positions, the timing difference analysis module 40 analyzes and finds that the first sensor has a 3-second delay due to network transmission problems; the second sensor has a delay of 1.5 seconds due to the performance of internal components; the third sensor has a delay of 2 seconds due to environmental electromagnetic interference; the fourth sensor has basically no delay, and the delay coefficient is 0. In this way, a set covering these four delay coefficients is generated. The delay coefficient set is of great significance for subsequent data processing and analysis, and can effectively correct the time deviation of data, greatly improving the accuracy and reliability of meteorological monitoring data.

[0028] In a possible implementation, the timing difference analysis module 40 includes: a signal transmission test unit, which is used to perform signal transmission tests on the K monitoring sensor arrays according to the signal test instructions, and obtain K test time point sets, wherein each test time point is a time node when the monitoring sensor array sends real-time test data to the corresponding station after receiving the signal test instruction, and the real-time test data arrives at the data processing center of the corresponding station. Specifically, the signal transmission test unit receives the signal test instruction, operates the K monitoring sensor arrays one by one, and when each monitoring sensor array receives the instruction, it sends real-time test data to the corresponding station, and the specific time node when the real-time test data arrives at the data processing center of the corresponding station is recorded, thereby forming a set of K test time points, ensuring that the data transmission time information of each sensor array is fully obtained.

[0029] A difference calculation unit is used to calculate the difference between the K test time point sets and the sending time node of the signal test instruction, and obtain K test feedback time difference sets. Specifically, the difference calculation unit obtains the K test time point sets generated by the signal transmission test unit, performs a difference operation on each test time point and the sending time node of the signal test instruction, and obtains K test feedback time difference sets through precise calculation, providing basic data for subsequent analysis.

[0030] A delay benchmark analysis space construction unit, the delay benchmark analysis space construction unit is used to construct K delay benchmark analysis spaces based on the K test feedback time difference value sets, wherein the K delay benchmark analysis spaces include K test particle point sets, and each test particle point corresponds to a test feedback time difference value. Specifically, the delay benchmark analysis space construction unit constructs a delay benchmark analysis space for each set based on the K test feedback time difference value sets obtained by the difference calculation unit, and the space includes K test particle point sets, and each test particle point corresponds to a specific test feedback time difference value, thereby constructing a spatial structure for in-depth analysis.

[0031] The difference fluctuation analysis unit is used to traverse the K test feedback time difference sets to perform difference fluctuation analysis and determine K analysis steps. Specifically, the difference fluctuation analysis unit performs in-depth difference fluctuation analysis on the K test feedback time difference sets one by one, calculates the difference variance in each set, and evaluates the degree of data dispersion. If the variance is large, it means that the data is more discrete, and a larger analysis step will be determined at this time. Specifically, the variance is calculated first, and then the K difference fluctuation variances are compared with the preset difference fluctuation variances, and then the ratio is multiplied by the preset distribution step, so as to determine the K analysis step.

[0032] A target delay center generation unit, the target delay center generation unit is used to obtain the K spatial centers of the K delay benchmark analysis spaces, and to search the K delay benchmark analysis spaces according to the K analysis steps starting from the K spatial centers to generate K target delay centers, wherein the K spatial centers are the centroids of the K delay benchmark analysis spaces. Specifically, the target delay center generation unit obtains the K spatial centers of the K delay benchmark analysis spaces, and the spatial centers are actually the centroids of each delay benchmark analysis space. Starting from the spatial centers, the corresponding delay benchmark analysis spaces are searched according to the determined K analysis steps, and finally K target delay centers are generated, providing key reference points for subsequent calculations and analyses.

[0033] A delay coefficient set generation unit, the delay coefficient set generation unit is used to use the test feedback time difference values ​​corresponding to the K target delay centers as K benchmark test feedback time difference values, respectively calculate the difference between the K test feedback time difference value sets and the K benchmark test feedback time difference values, and divide the calculation result by the K benchmark test feedback time difference values ​​to generate the K delay coefficient sets. Specifically, the delay coefficient set generation unit uses the test feedback time difference values ​​corresponding to the K target delay centers as K benchmark test feedback time difference values, respectively calculates the difference between the K test feedback time difference value sets and the K benchmark test feedback time difference values, and divides the calculation result by the K benchmark test feedback time difference values, thereby generating K delay coefficient sets, which provides an important basis for the performance evaluation and optimization of the entire system. For example, assuming that there are 5 monitoring sensor arrays, the signal transmission test unit obtains 5 different test time point sets. The difference calculation unit calculates the corresponding 5 test feedback time difference value sets. The delay benchmark analysis space construction unit constructs a delay benchmark analysis space for each set. The difference fluctuation analysis unit determines 5 analysis steps. The target delay center generation unit generates 5 target delay centers. The delay coefficient set generation unit finally obtains 5 delay coefficient sets for evaluating the signal transmission performance of the 5 monitoring sensor arrays.

[0034] In a possible implementation, the target delay center generation unit includes: an iterative test particle point generation subunit, the iterative test particle point generation subunit is used to take the K spatial centers as the starting point, move in the K delay benchmark analysis spaces according to the K analysis steps, and generate K iterative test particle points. Specifically, the iterative test particle point generation subunit takes the previously acquired K spatial centers as the starting point, and moves in the corresponding K delay benchmark analysis spaces according to the determined K analysis steps, and generates K iterative test particle points through the movement. For example, assuming that the first spatial center is located at a certain position, the first iterative test particle point is obtained by moving a certain distance according to the corresponding first analysis step, and so on, all K iterative test particle points are generated.

[0035] Iteration subunit, the iterative subunit is used to determine whether the difference in particle aggregation density between the K iterative test particle points and the K space centers meets the preset requirements. If not, the K iterative test particle points are updated as the starting point, and the iteration is continued in the K time delay benchmark analysis spaces. Specifically, the iterative subunit will judge the generated K iterative test particle points, calculate the difference in particle aggregation density between each iterative test particle point and the corresponding space center, and compare it with the preset requirements. The particle aggregation density refers to the ratio of the number of particles in the area constructed with the iterative test particle point as the center and the analysis step length as the radius to the area of ​​the area, reflecting the degree of particle aggregation distributed around each iterative test particle point. If the preset requirements are not met, the K iterative test particle points are updated as new starting points, and the next round of iteration is continued in the K time delay benchmark analysis spaces.

[0036] The target delay center determination subunit is used to stop iteration if so, and use the K spatial centers as the K target delay centers. Specifically, the target delay center determination subunit determines that when the iteration subunit determines that the difference in particle aggregation density between the K iterative test particle points and the K spatial centers meets the preset requirements, the target delay center determination subunit will stop the iteration operation. And these K spatial centers are determined as K target delay centers. For example, in a specific scenario, the difference in particle aggregation density between the initially generated iterative test particle points and the spatial center does not meet the requirements, so the iteration continues. After multiple iterations, the preset requirements are finally met. At this time, the spatial center is determined as the target delay center, which provides a key positioning point for subsequent processing.

[0037] The correction meteorological monitoring data set generation module 50 is used to perform asynchronous extraction and analysis of the preceding data of the K real-time meteorological monitoring data sets based on the K time delay coefficient sets to generate K correction meteorological monitoring data sets. Specifically, the correction meteorological monitoring data set generation module 50 analyzes K real-time meteorological monitoring data sets, identifies the key characteristics of the meteorological data in each set, and obtains the potential rules of the data in the time series. First, K delay coefficient sets and K real-time meteorological monitoring data sets are obtained. For each real-time meteorological monitoring data set, according to the corresponding delay coefficient, asynchronous extraction and in-depth analysis of the preceding data are implemented, and the preceding data that precedes the current data in the time dimension and is closely related to the current data are carefully screened out. With the help of the preceding data and the current real-time monitoring data, the scientific method of mean calculation is adopted to process them. Through mean calculation, the preceding data and the current data can be comprehensively considered, and the errors that may be caused by time differences can be effectively compensated. After processing the K real-time meteorological monitoring data sets one by one, K correction meteorological monitoring data sets are generated. For example, assuming that for the real-time meteorological monitoring data of a specific monitoring sensor, its corresponding delay coefficient indicates that there is a significant time delay phenomenon. When performing the correction operation, the relevant previous data in the past will be accurately extracted and the mean will be calculated together with the current data, so that the final correction meteorological monitoring data obtained will be more accurate and reliable, providing a solid data foundation for subsequent meteorological analysis and applications.

[0038] In a possible implementation, the correction meteorological monitoring data set generation module 50 includes: a coefficient comparison unit, the coefficient comparison unit is used to compare each delay coefficient in the K delay coefficient sets with the sum of the coefficients of the corresponding delay coefficient set, and subtract the calculation result from 1 to obtain K extraction factor sets. Specifically, the coefficient comparison unit divides each delay coefficient in each delay coefficient set by the sum of the coefficients of the delay coefficient set, and subtracts the calculation result from 1 to obtain an extraction factor. By performing such processing on all delay coefficients in each delay coefficient set, K extraction factor sets are finally obtained. For example, for a certain delay coefficient in the first delay coefficient set, the ratio of the delay coefficient to the sum of the coefficients of the set is calculated, and then the ratio is subtracted from 1 to obtain the corresponding extraction factor, and so on, the extraction factors of all sets are obtained.

[0039] The step-size set extraction unit is used to obtain a preset preamble window, multiply the K extraction factor sets by the preset preamble window, and obtain K preamble data extraction step-size sets. Specifically, the step-size set extraction unit obtains a preset preamble window, which specifies the time period in which the preamble data of the monitoring sensor can be extracted, and multiplies the K extraction factor sets obtained by the coefficient comparison unit by the preset preamble window respectively. Through the multiplication operation, K preamble data extraction step-size sets are obtained. The smaller the extraction factor, the smaller the step-size obtained by multiplying the preset preamble window, the smaller the time length between two adjacent extractions, and the more data the overall extracted data will be.

[0040] An asynchronous extraction unit, wherein the asynchronous extraction unit is used to asynchronously extract the preceding data of the K real-time meteorological monitoring data sets according to the K preceding data extraction step sets, and generate K preceding meteorological monitoring data sets. Specifically, the asynchronous extraction unit asynchronously extracts the preceding data of the K real-time meteorological monitoring data sets according to the K preceding data extraction step sets generated by the step set extraction unit. Since the larger the delay coefficient, the more correction data is required, the more data will be extracted, and finally, K preceding meteorological monitoring data sets are generated.

[0041] A mean calculation unit, the mean calculation unit is used to perform mean calculation on the K preceding meteorological monitoring data sets and the K real-time meteorological monitoring data sets respectively, to generate the K deviation-correcting meteorological monitoring data sets. Specifically, the mean calculation unit performs mean calculation on the K preceding meteorological monitoring data sets generated by the asynchronous extraction unit and the original K real-time meteorological monitoring data sets respectively. Through the mean calculation, the preceding data and the real-time data are integrated, thereby generating K deviation-correcting meteorological monitoring data sets to make up for possible errors and improve the accuracy and reliability of the data. For example, assuming that there are three delay coefficient sets, the coefficient comparison unit calculates the corresponding extraction factor set. The step set extraction unit uses a preset preceding window to obtain a preceding data extraction step set. The asynchronous extraction unit extracts the preceding data from the real-time meteorological monitoring data set according to the step set to form a preceding meteorological monitoring data set. The mean calculation unit performs mean calculation on these data to obtain the final deviation-correcting meteorological monitoring data set.

[0042] The regional meteorological characteristic analysis result generation module 60 is used to use the characteristic analysis unit to perform meteorological characteristic analysis on the K meteorological characteristic trend factor sets and the K correction meteorological monitoring data sets, and generate the regional meteorological characteristic analysis results of the target area. Specifically, the regional meteorological characteristic analysis result generation module 60 obtains K meteorological characteristic trend factor sets and K correction meteorological monitoring data sets. For the meteorological characteristic trend factor sets, each factor is first classified according to the meteorological element it represents, such as temperature, humidity, wind speed, etc., and then its numerical changes over a period of time are observed. If the temperature factor value continues to increase, it is judged that the temperature has risen. If the humidity factor value is stable, it is determined that the humidity has not changed significantly. If the wind speed factor value gradually decreases, it indicates that the wind speed has decreased, so as to clearly distinguish the characteristics of the direction of meteorological changes. For the correction meteorological monitoring data set, it is first grouped according to the chronological order or meteorological element category. Calculate the mean, median, standard deviation and other statistics of each set of data, draw a line graph or bar graph to intuitively present its changes in the time dimension, compare the data of different locations or time periods to find differences and similarities, and the feature analysis unit comprehensively considers the data and features of these two sets, combines the change direction characteristics obtained from the meteorological characteristic trend factor set with the specific numerical changes in the correction meteorological monitoring data set, and uses correlation analysis, regression analysis and other methods and appropriate data models for deep fusion and cross-analysis, and finally obtains regional meteorological characteristic analysis results that can comprehensively and accurately reflect the actual meteorological conditions and change trends. For example, suppose it is found that the temperature factor value in the meteorological characteristic trend factor set continues to rise, and the actual temperature data in the correction meteorological monitoring data set is also gradually rising. Combining these can produce an accurate analysis result that the temperature is rising.

[0043] In a possible implementation, the regional meteorological characteristic analysis result generation module 60 includes: a construction data acquisition unit, the construction data acquisition unit is used to acquire multiple sample meteorological characteristic trend factor sets, multiple sample deviation correction meteorological monitoring data sets and multiple sample regional meteorological characteristic analysis results as construction data. Specifically, the construction data acquisition unit acquires data through multiple channels, such as from professional meteorological databases, relevant research institutions or specific monitoring sites, and selects samples covering different geographical regions, different time periods and different meteorological conditions for multiple sample meteorological characteristic trend factor sets. The meteorological characteristic trend factors in the samples include the change trend of temperature, the increase and decrease trend of humidity, the fluctuation trend of air pressure, etc. For multiple sample deviation correction meteorological monitoring data sets, data that can more accurately reflect the real meteorological situation after correction and processing are collected, and the data involves multiple aspects such as wind speed, rainfall, light intensity, etc., and each sample is representative, and multiple sample regional meteorological characteristic analysis results are obtained. The results are obtained by professional meteorological personnel or advanced analysis systems, covering comprehensive judgments and detailed analysis of various complex meteorological conditions. Through targeted collection, the construction data acquisition unit prepares rich and comprehensive basic data for subsequent work.

[0044] A supervised training unit, the supervised training unit is used to supervise the preset neural network framework based on the constructed data to generate the feature analysis unit. Specifically, the supervised training unit is based on a large amount of data acquired by the constructed data acquisition unit, and inputs multiple sample meteorological feature trend factor sets, multiple sample deviation correction meteorological monitoring data sets and multiple sample regional meteorological feature analysis results into the preset neural network framework. During the training process, the neural network will continuously adjust the internal parameters and weights according to the input data. For example, when the input meteorological feature trend factor shows that the temperature continues to rise, and the corresponding deviation correction meteorological monitoring data also supports this trend, and the sample regional meteorological feature analysis results also show that the temperature rises, the neural network will learn this association pattern. Through multiple iterations and optimizations, the neural network gradually grasps the laws and features in the data. After continuous supervised training, a feature analysis unit that can accurately analyze meteorological features is finally generated. For example, assuming that the constructed data acquisition unit acquires rich sample data from 500 different regions covering four seasons, the supervised training unit uses these data to train the preset neural network for 80 rounds, and finally successfully creates a feature analysis unit that can accurately analyze the newly input meteorological data.

[0045] In the above, refer to Figure 1 A regional meteorological characteristics analysis system based on multi-source data fusion according to an embodiment of the present invention is described in detail. Figure 2A method for analyzing regional meteorological characteristics based on multi-source data fusion according to an embodiment of the present invention is described.

[0046] A regional meteorological characteristics analysis method based on multi-source data fusion, such as Figure 2 As shown, the method includes:

[0047] Collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers; perform meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively, and generate K meteorological characteristic trend factor sets; call K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring, and obtain K real-time meteorological monitoring data sets; traverse the K monitoring sensor arrays to perform sensor timing difference analysis, and generate K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor; perform asynchronous extraction analysis of pre-sequence data on the K real-time meteorological monitoring data sets based on the K delay coefficient sets, and generate K deviation-correcting meteorological monitoring data sets; use a feature analysis unit to perform meteorological characteristic analysis on the K meteorological characteristic trend factor sets and the K deviation-correcting meteorological monitoring data sets, and generate regional meteorological characteristic analysis results of the target area.

[0048] In a possible implementation, the regional meteorological characteristics analysis method based on multi-source data fusion also includes: clustering and identifying the K historical meteorological analysis data sets according to a preset meteorological indicator set to generate K historical meteorological indicator characteristic sequence sets; constructing K historical meteorological indicator characteristic change curve sets based on the K historical meteorological indicator characteristic sequence sets; performing trend analysis on the K historical meteorological indicator characteristic change curve sets to generate the K meteorological characteristic trend factor sets.

[0049] In a possible implementation, the regional meteorological characteristics analysis method based on multi-source data fusion also includes: traversing the K historical meteorological indicator characteristic change curve sets to perform slope analysis and generate K historical curve slope sets; traversing the K historical meteorological indicator characteristic change curve sets to perform curve acceleration analysis and generate K historical curve acceleration sets; performing weighted analysis on the K historical curve slope sets and the K historical curve acceleration sets to generate the K meteorological characteristic trend factor sets.

[0050] In a possible implementation, the regional meteorological characteristics analysis method based on multi-source data fusion also includes: according to the signal test instruction, respectively performing signal transmission tests on the K monitoring sensor arrays to obtain K test time point sets, wherein each test time point is a time node when the monitoring sensor array sends real-time test data to the corresponding station after receiving the signal test instruction, and the real-time test data arrives at the data processing center of the corresponding station; calculating the difference between the K test time point sets and the sending time node of the signal test instruction to obtain K test feedback time difference sets; constructing K delay benchmark analysis spaces based on the K test feedback time difference sets, wherein the K delay benchmark analysis spaces include K test particle point sets, each test particle A point corresponds to a test feedback time difference; traverse the K test feedback time difference sets to perform difference fluctuation analysis and determine K analysis steps; obtain the K spatial centers of the K delay benchmark analysis spaces, take the K spatial centers as the starting point, search in the K delay benchmark analysis spaces according to the K analysis steps, and generate K target delay centers, wherein the K spatial centers are the centroids of the K delay benchmark analysis spaces; use the test feedback time differences corresponding to the K target delay centers as K benchmark test feedback time differences, calculate the differences between the K test feedback time difference sets and the K benchmark test feedback time differences, and compare the calculation results with the K benchmark test feedback time differences to generate the K delay coefficient sets.

[0051] In a possible implementation, the regional meteorological characteristics analysis method based on multi-source data fusion also includes: taking the K spatial centers as the starting point, moving according to the K analysis steps in the K time delay benchmark analysis spaces, and generating K iterative test particle points; judging whether the particle aggregation density difference between the K iterative test particle points and the K spatial centers meets the preset requirements; if not, updating the K iterative test particle points as the starting point, and continuing to iterate in the K time delay benchmark analysis spaces; if so, stopping the iteration and taking the K spatial centers as the K target time delay centers.

[0052] In a possible implementation, the regional meteorological characteristics analysis method based on multi-source data fusion also includes: respectively dividing each delay coefficient in the K delay coefficient sets by the sum of the coefficients of the corresponding delay coefficient set, and subtracting the calculation result from 1 to obtain K extraction factor sets; obtaining a preset preamble window, multiplying the K extraction factor sets by the preset preamble window, and obtaining K preamble data extraction step length sets; performing preamble data asynchronous extraction on the K real-time meteorological monitoring data sets according to the K preamble data extraction step length sets, and generating K preamble meteorological monitoring data sets; performing mean calculation on the K preamble meteorological monitoring data sets and the K real-time meteorological monitoring data sets, and generating the K deviation-corrected meteorological monitoring data sets.

[0053] In a possible implementation, the regional meteorological feature analysis method based on multi-source data fusion also includes: obtaining multiple sample meteorological feature trend factor sets, multiple sample correction meteorological monitoring data sets and multiple sample regional meteorological feature analysis results as construction data; and performing supervised training on a preset neural network framework based on the construction data to generate the feature analysis unit.

[0054] A regional meteorological feature analysis system based on multi-source data fusion provided in an embodiment of the present invention can execute a regional meteorological feature analysis method based on multi-source data fusion provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0055] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0056] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A regional meteorological characteristics analysis system based on multi-source data fusion, characterized in that: The system comprises: A data set acquisition module, the data set acquisition module is used to collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers; A trend analysis module, the trend analysis module is used to perform meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively, and generate K meteorological characteristic trend factor sets; A monitoring data set acquisition module, wherein the monitoring data set acquisition module is used to call the K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring to obtain K real-time meteorological monitoring data sets; A timing difference analysis module, the timing difference analysis module is used to traverse the K monitoring sensor arrays to perform sensor timing difference analysis and generate K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor; A deviation correction meteorological monitoring data set generation module, the deviation correction meteorological monitoring data set generation module is used to perform asynchronous extraction and analysis of the preceding data of the K real-time meteorological monitoring data sets based on the K time delay coefficient sets to generate K deviation correction meteorological monitoring data sets; The regional meteorological characteristic analysis result generation module is used to use the characteristic analysis unit to perform meteorological characteristic analysis on the K meteorological characteristic trend factor sets and the K correction meteorological monitoring data sets to generate the regional meteorological characteristic analysis results of the target area.

2. The system according to claim 1, characterized in that The trend analysis module is used to perform the following steps: An indicator feature sequence set generation unit, the indicator feature sequence set generation unit is used to perform clustering identification on the K historical meteorological analysis data sets according to a preset meteorological indicator set to generate K historical meteorological indicator feature sequence sets; A change curve set construction unit, the change curve set construction unit is used to construct K historical meteorological indicator characteristic change curve sets based on the K historical meteorological indicator characteristic sequence sets; A trend analysis unit is used to perform trend analysis on the K historical meteorological indicator characteristic change curve sets to generate the K meteorological characteristic trend factor sets.

3. The system according to claim 2, characterized in that The trend analysis unit is used to perform the following steps: A historical curve slope set generating subunit, the historical curve slope set generating subunit is used to traverse the K historical meteorological indicator characteristic change curve sets to perform slope analysis and generate K historical curve slope sets; A historical curve acceleration set generation subunit, the historical curve acceleration set generation subunit is used to traverse the K historical meteorological indicator characteristic change curve sets to perform curve acceleration analysis and generate K historical curve acceleration sets; A weighted analysis subunit, the weighted analysis subunit is used to perform weighted analysis on the K historical curve slope sets and the K historical curve acceleration sets to generate the K meteorological characteristic trend factor sets.

4. The system according to claim 1, characterized in that The timing difference analysis module is used to perform the following steps: A signal transmission test unit, the signal transmission test unit is used to perform signal transmission tests on the K monitoring sensor arrays respectively according to the signal test instruction, and obtain K test time point sets, wherein each test time point is a time node when the monitoring sensor array sends real-time test data to the corresponding station after receiving the signal test instruction, and the real-time test data arrives at the data processing center of the corresponding station; A difference calculation unit, the difference calculation unit is used to calculate the difference between the K test time point sets and the sending time node of the signal test instruction, and obtain K test feedback time difference sets; A delay benchmark analysis space construction unit, the delay benchmark analysis space construction unit is used to construct K delay benchmark analysis spaces based on the K test feedback time difference value sets, wherein the K delay benchmark analysis spaces include K test particle point sets, each test particle point corresponds to a test feedback time difference value; A difference fluctuation analysis unit, the difference fluctuation analysis unit is used to traverse the K test feedback time difference sets to perform difference fluctuation analysis and determine K analysis step lengths; A target delay center generation unit, the target delay center generation unit is used to obtain K spatial centers of the K delay benchmark analysis spaces, take the K spatial centers as starting points, search in the K delay benchmark analysis spaces according to the K analysis steps, and generate K target delay centers, wherein the K spatial centers are the centroids of the K delay benchmark analysis spaces; A delay coefficient set generation unit, the delay coefficient set generation unit is used to use the test feedback time difference values ​​corresponding to the K target delay centers as K benchmark test feedback time difference values, respectively calculate the difference between the K test feedback time difference value sets and the K benchmark test feedback time difference values, and compare the calculation results with the K benchmark test feedback time differences to generate the K delay coefficient sets.

5. The system according to claim 4, characterized in that The target delay center generation unit is used to perform the following steps: An iterative test particle point generating subunit, the iterative test particle point generating subunit is used to take the K space centers as starting points, move in the K delay benchmark analysis spaces according to the K analysis step sizes, and generate K iterative test particle points; An iterative subunit, the iterative subunit is used to determine whether the particle aggregation density difference between the K iterative test particle points and the K space centers meets the preset requirements, if not, the K iterative test particle points are updated as the starting point, and the iteration is continued in the K time delay benchmark analysis spaces; The target delay center determination subunit is used to stop iteration if yes and use the K spatial centers as the K target delay centers.

6. The system according to claim 1, wherein: The correction meteorological monitoring data set generation module is used to perform the following steps: A coefficient comparison unit, the coefficient comparison unit is used to compare each delay coefficient in the K delay coefficient sets with the sum of coefficients of the corresponding delay coefficient set, and subtract the calculation result from 1 to obtain K extraction factor sets; A step length set extraction unit, the step length set extraction unit is used to obtain a preset preamble window, multiply the K extraction factor sets by the preset preamble window, and obtain K preamble data extraction step length sets; An asynchronous extraction unit, the asynchronous extraction unit being used to perform asynchronous extraction of preceding data on the K real-time meteorological monitoring data sets respectively according to the K preceding data extraction step length sets, to generate K preceding meteorological monitoring data sets; A mean calculation unit is used to perform mean calculation on the K previous meteorological monitoring data sets and the K real-time meteorological monitoring data sets respectively to generate the K deviation-corrected meteorological monitoring data sets.

7. The system according to claim 1, characterized in that The regional meteorological characteristics analysis result generation module is used to perform the following steps: A construction data acquisition unit is used to acquire multiple sample meteorological characteristic trend factor sets, multiple sample deviation correction meteorological monitoring data sets and multiple sample regional meteorological characteristic analysis results as construction data; A supervised training unit is used to perform supervised training on a preset neural network framework based on the construction data to generate the feature analysis unit.

8. A method for analyzing regional meteorological characteristics based on multi-source data fusion, characterized in that: The method is applied to a regional meteorological characteristics analysis system based on multi-source data fusion as described in any one of claims 1 to 7, and the method comprises: Collect K historical meteorological analysis data sets of K stations in the target area, wherein the K stations include K station location identifiers; Performing meteorological characteristic trend analysis on the K historical meteorological analysis data sets respectively to generate K meteorological characteristic trend factor sets; Calling K monitoring sensor arrays of the K stations to perform real-time meteorological data monitoring to obtain K real-time meteorological monitoring data sets; Traversing the K monitoring sensor arrays to perform sensor timing difference analysis, and generating K delay coefficient sets, wherein each delay coefficient corresponds to a monitoring sensor; Based on the K time delay coefficient sets, asynchronous extraction and analysis of preceding data is performed on the K real-time meteorological monitoring data sets to generate K deviation-corrected meteorological monitoring data sets; A feature analysis unit is used to perform meteorological feature analysis on the K meteorological feature trend factor sets and the K deviation-correcting meteorological monitoring data sets to generate a regional meteorological feature analysis result for the target area.

Images