Rainfall intelligent monitoring and early warning system based on phased array radar

Abstract

The present application relates to a kind of based on phased array radar's rainfall intelligent monitoring early warning system, belong to phased array radar rainfall detection data analysis processing technical field.The system first determines the moving track of rainfall single body to complete the moving direction prediction and moving distance prediction of each monitoring point after current time, so that the deviation between the direction angle between the monitoring point to be investigated and other monitoring points and the predicted motion direction of other monitoring points, the distance deviation between the predicted position of the monitoring point to be investigated and other monitoring points and the current monitoring value of other monitoring points can be determined, the influence of other monitoring points on the monitoring point to be investigated at next monitoring time is determined, the future weather of the monitoring point to be investigated influenced by other monitoring points under the condition that traditional prediction method cannot be characterized under the condition of weather rapid change is characterized by the influence, so that the monitoring value prediction of the monitoring point to be investigated can be more accurately completed, and the rainfall monitoring and early warning accuracy of phased array radar under the condition of weather rapid change is improved.

Description

Technical Field

[0001] This invention relates to the field of phased array radar rainfall detection data analysis and processing technology, and in particular to a smart rainfall monitoring and early warning system based on phased array radar. Background Technology

[0002] Phased array radar is a radar system composed of multiple transmitting / receiving units. By adjusting the phase of the transmitted signals of each unit, it can quickly complete scanning and positioning. Compared with traditional radar, phased array radar has a faster scanning speed and higher accuracy. In rainfall monitoring and early warning systems, it has higher real-time performance, accuracy, and flexibility, making it an increasingly important component of rainfall early warning systems.

[0003] However, in scenarios of sudden, short-duration heavy rainfall, rain cells generally exhibit rapid movement and rapid generation, dissipation, merging, and splitting. Weather conditions change rapidly and differ significantly from historical weather patterns. Although phased array radar is more flexible and precise, capable of forming higher-resolution monitoring points during meteorological monitoring and obtaining more accurate monitoring values ​​at each point, thus acquiring more detailed and accurate historical data, it still suffers from insufficient accuracy in rainfall monitoring and early warning in such scenarios. This is because long-term historical data lacks reference value for predicting current rapidly changing weather scenarios, while recent historical data is insufficient in terms of data volume and interpretation of current rapidly changing weather scenarios.

[0004] Therefore, how to further improve the accuracy of phased array radar in monitoring and warning of rainfall in rapidly changing weather scenarios has become an urgent technical problem to be solved. Summary of the Invention

[0005] In view of this, the present invention provides a smart rainfall monitoring and early warning system based on phased array radar to solve the technical problem of insufficient accuracy of rainfall monitoring and early warning by phased array radar in scenarios of rapid weather changes.

[0006] The present invention provides a smart rainfall monitoring and early warning system based on phased array radar, comprising:

[0007] The historical data acquisition module is used to acquire the monitoring values ​​of each monitoring point in the monitoring area by the phased array radar within a set time period, and to determine the rainfall cells at each monitoring time.

[0008] The rainfall cell trajectory determination module is used to take any rainfall cell at the current monitoring time as the target cell, determine the proximity of the target cell to each rainfall cell at the previous monitoring time, and determine the movement trajectory of the target cell at the set time period based on the proximity of the positions.

[0009] a prediction value affected degree determination module, configured to determine a moving direction prediction value and a moving distance prediction value of each monitoring point in the target single cell at a next monitoring time according to the moving trajectory, thereby determining the moving direction prediction value and the moving distance prediction value of the monitoring point in the single cell at the current monitoring time, and further determining an affected degree of any monitoring point at the current monitoring time by other monitoring points at the current monitoring time;

[0010] a prediction value correction and early warning module, configured to determine a corrected prediction value of any monitoring point at the next monitoring time according to the prediction value of the monitoring point at the next monitoring time and the affected degree at the current monitoring time, thereby completing rainfall early warning.

[0011] Further, the determination of the single cell at each monitoring time comprises:

[0012] clustering the monitoring points in the monitoring area with a monitoring value greater than a preset monitoring threshold, and taking a clustering cluster with an area greater than a preset area threshold in the obtained clustering cluster as the single cell.

[0013] Further, the determination of the position proximity degree of the target single cell to each single cell at the previous monitoring time comprises:

[0014] continuously selecting a preset number of edge monitoring points closest to the single cell to be judged at the previous monitoring time from the edge monitoring points of the target single cell to form a target edge section, determining a corresponding edge section of the single cell to be judged at the previous monitoring time corresponding to the target edge section, and determining the position proximity degree of each single cell at the previous monitoring time to the target single cell according to a mean value of distances between corresponding edge monitoring points in the target edge section and the corresponding edge section.

[0015] Further, the determination of the corresponding edge section of the single cell to be judged at the previous monitoring time corresponding to the target edge section comprises:

[0016] randomly selecting the preset number of continuous edge monitoring points from the edge monitoring points of the single cell to be judged at the previous monitoring time to form a to-be-judged edge section, determining a distance value between the target edge section and the to-be-judged edge section according to a mean value of distances between corresponding edge monitoring points in the target edge section and the to-be-judged edge section, and taking a to-be-judged edge section corresponding to a maximum distance value as the corresponding edge section in the single cell to be judged at the previous monitoring time.

[0017] Further, the determination of the moving trajectory of the target single cell in the set time length according to the position proximity degree comprises:

[0018] selecting, at a previous monitoring time, a set number of rainfall monomers closest to the position of the to-be-analyzed monomer as to-be-determined rainfall monomers, selecting, at a monitoring time prior to the previous monitoring time, a set number of rainfall monomers closest to the position of the new to-be-analyzed monomer as new to-be-determined rainfall monomers, and repeating the process of determining the to-be-determined rainfall monomers corresponding to the to-be-analyzed monomers until all monitoring times in the set time length are traversed;

[0019] determining each monomer combination formed by the target monomer and the to-be-determined rainfall monomers corresponding to each monitoring time, determining the angle change amount and the position change amount of the rainfall monomers at any two adjacent monitoring times in the current monomer combination, and taking the monomer combination with the smallest angle change amount and position change amount as the movement trajectory.

[0020] Further, the method for determining the movement direction prediction value and the movement distance prediction value of each monitoring point in the target monomer at the next monitoring time according to the movement trajectory comprises:

[0021] predicting the predicted angle change amount of the target monomer at the next monitoring time relative to the current monitoring time according to the angle change amount of the target monomer at any two adjacent monitoring times in the movement trajectory, determining the movement direction prediction value according to the movement direction of the target monomer at the current monitoring time relative to the target monomer at the previous monitoring time and the predicted angle change amount, and determining the movement distance prediction value according to the average of the movement distances of the target monomer at any two adjacent monitoring times in the movement trajectory.

[0022] Further, the method for predicting the predicted angle change amount of the target monomer at the next monitoring time relative to the current monitoring time according to the angle change amount of the target monomer at any two adjacent monitoring times in the movement trajectory comprises:

[0023] arranging the angle change amounts of the target monomer at any two adjacent monitoring times in the movement trajectory in a time sequence to form an angle change amount sequence and performing curve fitting to obtain an angle change amount fitting curve, and taking the value of the angle change amount fitting curve at the next monitoring time as the predicted angle change amount.

[0024] Further, the method for further determining the influence degree of any monitoring point at the current monitoring time on other monitoring points at the current monitoring time comprises:

[0025] determine a direction angle from any rainfall cell monitoring point at the current monitoring time to the target monitoring point, and calculate a direction deviation between the direction angle and a moving direction prediction value of the any rainfall cell monitoring point at the next monitoring time;

[0026] determine a distance deviation between the target monitoring point and the any rainfall cell monitoring point at the next monitoring time according to a distance value between the target monitoring point and the any rainfall cell monitoring point at the current monitoring time and a moving distance prediction value of the any rainfall cell monitoring point at the next monitoring time;

[0027] determine the influence degree according to the direction deviation, the distance deviation and a monitoring value of the any rainfall cell monitoring point at the current monitoring time.

[0028] Further, the influence degree is:

[0029]

[0030] wherein H i represents an influence degree of the i-th monitoring point at the current monitoring time on the next monitoring time by other monitoring points, N i represents a total number of remaining monitoring points in the monitoring area except for the rainfall cell where the i-th monitoring point is located, represents a direction angle from the e-th monitoring point to the i-th monitoring point at the current monitoring time, represents a moving direction prediction value of the e-th monitoring point at the next monitoring time at the current monitoring time, represents a direction deviation between d and d , d i,e represents a distance value between the i-th monitoring point and the e-th monitoring point at the current monitoring time, D e represents a moving distance prediction value of the e-th monitoring point at the next monitoring time at the current monitoring time, d i,e -D e represents a distance deviation between d i,e and d e , d i,e represents an absolute value of d e , and ε is used to ensure that the fraction is meaningful, Z e represents a monitoring value of the e-th monitoring point at the current monitoring time, and norm() represents a normalization function.

[0031] Further, the predicted value of any monitoring point at the next monitoring time is determined, comprising:

[0032] Based on the monitoring value of any monitoring point at each monitoring time within the set time length at the current monitoring time, the ARIMA algorithm is used to obtain the predicted value of any monitoring point at the next monitoring time at the current monitoring time.

[0033] Compared with the prior art, the present application has the following beneficial effects:

[0034] The system first determines the moving direction prediction and moving distance prediction of each monitoring point at the predicted time after the current time by determining the moving track of the rainfall single body, so as to determine the influence degree of other monitoring points on the monitoring point to be investigated at the next monitoring time by combining the deviation between the direction angle of the monitoring point to be investigated and the predicted motion direction of other monitoring points, the distance deviation between the predicted positions of the monitoring point to be investigated and other monitoring points, and the monitoring value of other monitoring points at the current time, which represents the influence of other monitoring points on the future weather change of the monitoring point to be investigated under the condition that the traditional prediction method cannot represent, so that the monitoring value prediction of the monitoring point to be investigated can be more accurately completed after considering the influence, and the rainfall monitoring and early warning accuracy of the phased array radar under the condition of rapid weather change is improved. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0036] Figure 1 is a structural block diagram of a rainfall intelligent monitoring and early warning system based on a phased array radar provided by the first embodiment of the present application. DETAILED DESCRIPTION

[0037] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" or "in other embodiments" or "in some embodiments" in various places throughout this specification are not necessarily referring to the same embodiment, unless otherwise specified. Furthermore, the particular features, structures, or characteristics can be combined in any suitable manner on one or more embodiments. The terms "including," "containing," and "having" are intended to be open-ended terms that are to be interpreted in the context of this application, and are not intended to be limiting. The terms "a" and "an" are defined as one or more unless explicitly stated otherwise herein. Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated in the specification as if it were individually recited herein.

[0038] It should be understood that the size of the serial number of each step in the following embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the application.

[0039] In order to further illustrate the technical solutions of the present application, the following will be described by specific embodiments.

[0040] System embodiment:

[0041] Reference Figure 1 is a structural block diagram of a rainfall intelligent monitoring and early warning system based on a phased array radar provided by Embodiment One of the present application, as Figure 1 shown, the system comprises a monitoring historical data acquisition module 11, a rainfall single track determination module 12, a prediction value affected degree determination module 13, and a prediction value correction and early warning module 14, which will be introduced respectively as follows:

[0042] The monitoring historical data acquisition module 11 is used to acquire the monitoring values of each monitoring point in the monitoring area by the phased array radar within a set time length, and to determine the rainfall single at each monitoring time.

[0043] By installing a phased array radar at the center of the early warning area, the radar can scan the three-dimensional space according to the set scanning interval (such as 1 min), at the set multiple elevation angle levels and azimuth angle, and each scan can obtain the echo intensity data of the area, the data is in the form of a three-dimensional data body, each data point or monitoring point includes the coordinate data and echo intensity (reflectivity) of the point, and the echo intensity value is the monitoring value. Among them, the elevation angle is used to control the height of the scan, and the azimuth angle is used to control the height of the scan, that is, a 360° scan is performed at each elevation angle layer to obtain the echo intensity data at different heights, and the number of elevation angle layers is usually set to 12-20 layers, and the elevation angle is adjusted according to the actual weather condition, and the local area can be scanned at a higher frequency. The echo intensity data of a set time (such as 15 min) before the prediction time is taken as the historical data for prediction.

[0044] Since the formation of rainfall is often related to the rainfall cell (cloud cluster), and the cloud cluster may change rapidly when affected by wind, air temperature and other factors, thereby causing great changes in the rainfall distribution of the monitoring area. It can be understood that the position in the moving direction of the rainfall cell is more likely to increase the rainfall, and the rainfall and the possibility of rainfall gradually decrease in the direction away from the rainfall cell. Therefore, the echo intensity of the monitoring point in the moving direction of the rainfall cell is more easily affected by the echo intensity of other monitoring points. Therefore, to analyze the influence relationship between the echo intensities of different monitoring points, the moving trend of the rainfall needs to be obtained first.

[0045] The moving trend of the rainfall is the moving trend of the rainfall cell, so it is necessary to identify the rainfall cell in the current area at each monitoring time first. The echo intensity of the rainfall cell is relatively higher, and according to the common echo intensity range of the rainfall cell, when the echo intensity of the monitoring point is greater than a preset monitoring threshold (30 dBZ is preferred in this embodiment), it can be considered that the monitoring point is in the rainfall cell. When the area of the obtained clustering cluster is greater than a preset area threshold, it is considered that the clustering cluster is an effective rainfall cell.

[0046] The rainfall cell trajectory determination module 12 is configured to take any rainfall cell at the current monitoring time as a target cell, determine the position proximity degree of the target cell and each rainfall cell at the previous monitoring time, and determine the moving trajectory of the target cell in the set time according to the position proximity degree.

[0047] Then, according to the change of the position of the rainfall unit and the echo intensity corresponding to the continuous monitoring time, the moving trend of the rainfall unit is analyzed. Because the rainfall unit can grow, split, merge and change, the shape and size of the same rainfall unit in different monitoring times can not be completely the same, and the same rainfall unit in different monitoring times cannot be directly identified according to the shape of the rainfall unit in different monitoring times. However, the moving direction of the rainfall unit usually has certain regularity, so the position of the same rainfall unit in the scanning data in different monitoring times is obtained according to the relevance of the position of the rainfall unit in different monitoring times, and then the moving trend of the rainfall unit, that is, the moving track of the rainfall unit is obtained.

[0048] At the current time, that is, a certain rainfall unit at the last time in the obtained historical data is taken as a target unit for analysis.

[0049] Firstly, the proximity degree of each rainfall unit in the previous monitoring time and a certain rainfall unit in the next monitoring time is determined. Taking the proximity degree between the target unit in the current monitoring time and a certain rainfall unit to be judged in the previous monitoring time as an example, the proximity degree of the rainfall unit is not obtained by the centroid of the rainfall unit in this embodiment, but the distance value between the same position on the edge of the rainfall unit in adjacent time is used to determine the proximity degree. The process is as follows: first, a target edge section is formed by continuously selecting the nearest preset number (20 is preferred in this embodiment) of edge monitoring points of the target unit to the rainfall unit to be judged in the previous monitoring time on the edge of the target unit, that is, in the edge monitoring points of the target unit, then a corresponding edge section corresponding to the target edge section on the rainfall unit to be judged in the previous monitoring time is determined, and the proximity degree of the rainfall unit to be judged in the previous monitoring time and the target unit is determined according to the average distance between the corresponding edge monitoring points in the target edge section and the corresponding edge section.

[0050] In which, the determination method of the corresponding edge section is:

[0051] In the previous monitoring moment, 20 continuous edge monitoring points in the edge of the to-be-judged rainfall cell are randomly selected to form a to-be-judged edge segment, the distance value between the target edge segment and the to-be-judged edge segment is determined according to the average distance between the target edge segment and the corresponding edge monitoring points in the to-be-judged edge segment, and the to-be-judged edge segment corresponding to the maximum distance value is taken as the corresponding edge segment corresponding to the target edge segment in the to-be-judged rainfall cell at the previous monitoring moment. The reason why the to-be-judged edge segment corresponding to the maximum distance value is taken as the corresponding edge segment corresponding to the target edge segment in the to-be-judged rainfall cell at the previous monitoring moment is that the purpose here is to determine the proximity of the rainfall cell at the adjacent moment, so only the moving distance of the same position on the rainfall cell at the adjacent moment needs to be considered, such as the moving distance of the left position of the rainfall cell at the adjacent two moments, and the situation that the edge points on the corresponding edge segment are not the original edge points on the target edge segment due to rotation or deformation of the rainfall cell during movement is not considered, that is, the process of determining the corresponding edge segment is not the process of finding the substantially corresponding edge points on the rainfall cell at the adjacent two moments, but only the process of determining the same position on the rainfall cell at the adjacent two moments. Since the embodiment first selects the edge segment closest to the to-be-judged rainfall cell at the previous moment on the target cell as the target edge segment, the distance between the target edge segment and the corresponding edge segment should be greater than the distance between the target edge segment and other edge segments on the edge of the to-be-judged rainfall cell at the previous moment, so as to realize accurate determination of the corresponding edge segment.

[0052] The specific implementation of determining the corresponding edge segment can be:

[0053] Taking 20 continuous monitoring points as the window length, all edge monitoring points of the to-be-judged rainfall cell at the previous monitoring moment are sequentially traversed, and the average distance between the edge monitoring points in each window and the target edge segment of the target cell at the current monitoring moment is calculated:

[0054]

[0055] wherein s w represents the average distance between the edge monitoring points in the wth window and the target edge segment, n w represents the number of edge monitoring points contained in the wth window (here n w = 20), and d w,j represents the distance between the jth edge monitoring point in the wth window and the corresponding jth edge monitoring point in the target edge segment.

[0056] wherein s wThe maximum value corresponds to the edge monitoring point in the window which is considered to be the monitoring point in the corresponding edge segment corresponding to the target edge segment of the target cell at the current monitoring time. According to the arrangement order of the edge monitoring points in the window and the arrangement order of the monitoring points in the target edge segment, such as the clockwise or counterclockwise order, the edge monitoring points in the two edge segments can be corresponded one by one to form 20 groups of edge monitoring points.

[0057] Based on the obtained 20 groups of edge monitoring points on the target edge segment and the corresponding edge segment corresponding thereto, the position proximity between the rainfall cell where the corresponding edge segment is located and the target cell where the target edge segment is located is determined according to the mean value of the distance between the corresponding edge monitoring points (i.e., the 20 groups of edge monitoring points) in the target edge segment and the corresponding edge segment, so as to determine the position proximity between each rainfall cell and the target cell at the previous monitoring time:

[0058]

[0059] wherein c u represents the position proximity between the u th rainfall cell and the target cell at the previous monitoring time, n c represents the number of edge monitoring points in the target edge segment, i.e., the number of groups of corresponding edge monitoring point pairs on the target edge segment and the corresponding edge segment corresponding thereto (herein n c = 20), and Δd u,j represents the distance value between the j th edge monitoring point in the corresponding edge segment of the u th rainfall cell and the corresponding j th edge monitoring point in the target edge segment at the previous monitoring time. represents that the smaller the distance is, the closer the positions of the two rainfall cells are. norm() represents a normalization function, such as norm normalization, etc.

[0060] It should be noted that if the positions of the two rainfall cells completely overlap or one covers the other, it is considered that the rainfall cell does not move, and only growth or partial dissipation may occur, so they are considered to be the same rainfall cell, and the corresponding position proximity takes the maximum value (herein the maximum value of c u is 1).

[0061] Because the distance between different rainfall cells is usually significantly greater than the moving distance of a rainfall cell between adjacent two monitoring times, the target cell is first taken as an analysis cell, and then the rainfall cells with the maximum position proximity to the analysis cell at the previous monitoring time are selected as the to-be-determined rainfall cells, i.e., the first three rainfall cells with the highest position proximity to the current analysis rainfall cell at the previous monitoring time are selected, which are considered to be the same rainfall cell as the current analysis cell.

[0062] Then, the determined pending rainfall monomer is taken as a new to-be-analyzed monomer, and a set number (3) of rainfall monomers closest to the position of the new to-be-analyzed monomer at a more previous monitoring time (a previous-previous monitoring time) are selected as new pending rainfall monomers, and the process of determining the pending rainfall monomers corresponding to the to-be-analyzed monomers is repeated until all monitoring times within the set time length are traversed.

[0063] Each monomer combination formed by the target monomer and the pending rainfall monomers corresponding to each monitoring time is determined, the angle change amount and the position change amount of the rainfall monomers at any two adjacent monitoring times in the current monomer combination are determined, and the monomer combination with the smallest angle change amount and position change amount is taken as the moving track.

[0064] Specifically, it is considered that the traditional rainfall monomer track analysis method is obtained according to the centroid position of the rainfall monomer, but because the centroid of the rainfall monomer deviates obviously when merging and decomposing, the analysis of the moving trend of the rainfall is affected. Therefore, the moving direction of the current to-be-analyzed rainfall monomer relative to each suspected same rainfall monomer is calculated according to the relative positions between the 20 closest edge monitoring points:

[0065]

[0066] wherein θ u represents the azimuth angle of the to-be-analyzed monomer at the later monitoring time relative to the u-th pending rainfall monomer at the previous time, n c represents the number of edge monitoring point groups used to calculate the position proximity (herein n c = 20), θ u,j represents the included angle of the line connecting the j-th pair of edge monitoring points after being projected onto the same horizontal plane and the positive direction of the x-axis. α u represents the pitch angle of the to-be-analyzed monomer at the later monitoring time relative to the u-th pending rainfall monomer at the previous time, α u,j represents the inclination angle of the line connecting the j-th pair of edge monitoring points relative to the horizontal plane. represents the relative moving direction between the to-be-analyzed monomer at the later monitoring time and the u-th rainfall monomer at the later monitoring time, because the radar monitoring points are distributed in three-dimensional space, the moving direction is represented by a vector composed of the azimuth angle and the pitch angle corresponding thereto.

[0067] The same rainfall monomer has certain regularity in the moving process, that is, the moving direction of the rainfall monomer changes less between adjacent monitoring times. After the above suspected same rainfall monomers are combined to obtain a plurality of monomer combinations, in any monomer combination, the relative moving direction feature vectors corresponding to adjacent monitoring times are calculated The angle change amount between the two adjacent monitoring time points is calculated, and a fitting curve of the angle change amount is obtained by using the least square method according to the time sequence of each monitoring time point in the set time length (15 min). The better the fitting degree of the fitting curve is, the stronger the regularity of the rainfall monomer in the monomer combination is, and the more likely the monomer combination is the same rainfall monomer. The calculation method of the angle change amount is as follows:

[0068]

[0069] wherein, β represents the angle change amount, and respectively represent the relative movement direction vectors of the rainfall monomers at the two adjacent monitoring time points in the monomer combination, represents the module of the vector obtained by the cross product of the two vectors, and atan2() represents an arctangent function, and the output is an angle value with positive and negative.

[0070] Therefore, the possibility that the rainfall monomers in each monomer combination are the same rainfall monomer is calculated by comprehensively considering the root mean square error value of the fitting curve of the angle change amount of the rainfall monomers in each monomer combination and the position proximity between the rainfall monomers at the adjacent monitoring time points in the combination. The formula is as follows:

[0071]

[0072] wherein, Y m represents the possibility that the rainfall monomers in the mth monomer combination are the same rainfall monomer, σ m represents the root mean square error value of the fitting curve of the angle change amount of the rainfall monomers in the mth monomer combination, represents that the smaller the root mean square error is, the better the fitting degree of the fitting curve is, that is, the stronger the regularity of the movement direction of the rainfall monomers in the monomer combination is.c m represents the number of monitoring time points contained in the mth monomer combination.c m,u represents the position proximity between the u th rainfall monomer and the rainfall monomer at the previous time point in the mth monomer combination. norm() represents a normalization function.

[0073] The monomer combination with the largest possibility is considered as the moving route of the current monomer to be analyzed. Therefore, the moving trajectory of all the rainfall monomers at the current monitoring time point can be obtained.

[0074] The influence degree determination module 13 is configured to determine the movement direction prediction value and the movement distance prediction value of each monitoring point in the target monomer at the next monitoring time point according to the moving trajectory, so as to determine the movement direction prediction value and the movement distance prediction value of the monitoring point in each rainfall monomer at the current monitoring time point, and further determine the influence degree of each monitoring point at the current monitoring time point on other monitoring points at the current monitoring time point.

[0075] In the process of obtaining the moving track of each rainfall monomer at the current monitoring moment, the moving direction and the angle change amount of the rainfall monomer at each historical monitoring moment relative to the previous monitoring moment are obtained, and the angle change amount fitting function is obtained. Based on the angle change amount fitting function, the value of the angle change amount fitting function at the next monitoring moment after the current monitoring moment is taken as the predicted angle change amount, and the moving direction of the rainfall monomer at the next monitoring moment is obtained by combining the moving direction of the rainfall monomer at the current monitoring moment.

[0076] In addition, in the process of obtaining the moving track of each rainfall monomer at the current monitoring moment, the position of the rainfall monomer at each historical monitoring moment can be determined, so that the moving distance prediction value of the rainfall monomer at the next monitoring moment after the current monitoring moment can be determined according to the average of the moving distances of the rainfall monomer at any two adjacent monitoring moments in the moving track:

[0077]

[0078] Wherein, D represents the moving distance prediction value of the current to-be-analyzed monomer at the next monitoring moment after the current monitoring moment, n D represents the number of monitoring moment intervals contained in the monomer combination corresponding to the current to-be-analyzed rainfall monomer, that is, the number of sampling intervals between each monitoring moment in the set time length, represents the moving distance of the rainfall monomer in the a-th interval.

[0079] Since the structure and meteorological characteristics of the rainfall monomer are uniform, the moving direction prediction value and the moving distance prediction value of the rainfall monomer at the next monitoring moment after the current monitoring moment can be taken as the moving direction prediction value and the moving distance prediction value of all monitoring points in the rainfall monomer at the current monitoring moment. Thus, the moving direction prediction value and the moving distance prediction value of each monitoring point in each rainfall monomer at the current monitoring moment can be determined.

[0080] After obtaining the moving direction prediction value and the moving distance prediction value of each monitoring point in each rainfall monomer at the current monitoring moment, the position where the echo intensity of each monitoring point in the rainfall monomer affects other monitoring points at the next monitoring moment can be predicted. Thus, the affected degree of each monitoring point at the next monitoring moment is calculated by combining the relative positions between each monitoring point and other monitoring points and the echo intensity, and the construction idea is as follows:

[0081] Taking any monitoring point at the current monitoring moment as a target monitoring point, the direction angle from the monitoring point in any rainfall monomer at the current monitoring moment to the target monitoring point is determined, and the directional deviation between the direction angle and the moving direction prediction value of the monitoring point in any rainfall monomer at the next monitoring moment is calculated.

[0082] According to the distance value between the target monitoring point and the monitoring point in any rainfall cell at the current monitoring time and the moving distance prediction value of the monitoring point in any rainfall cell at the current monitoring time at the next monitoring time, the distance deviation of the target monitoring point and the monitoring point in any rainfall cell at the current monitoring time at the next monitoring time is determined.

[0083] According to the direction deviation, the distance deviation and the monitoring value of the monitoring point in any rainfall cell at the current monitoring time, the influence degree is determined.

[0084] The formula is constructed as:

[0085]

[0086] Wherein, H i represents the influence degree of the i th monitoring point at the current monitoring time at the next monitoring time by other monitoring points, N i represents the total number of remaining monitoring points in the monitoring area except the rainfall cell where the i th monitoring point is located (because the rainfall cell has a certain structure, so the influence degree here does not include the monitoring points in the same rainfall cell), it is easy to understand that if the i th monitoring point is not in the rainfall cell at the current monitoring time, then N i represents the total number of other monitoring points in the monitoring area except the monitoring point, represents the relative moving direction vector composed of the azimuth angle and the pitch angle of the i th monitoring point at the current monitoring time, which also represents the direction angle of the i th monitoring point at the current monitoring time, represents the moving direction prediction value of the e th monitoring point at the next monitoring time at the current monitoring time, represents and the direction deviation between them, d i,e represents the distance value between the i th monitoring point and the e th monitoring point at the current monitoring time, D e represents the moving distance prediction value of the e th monitoring point at the next monitoring time at the current monitoring time, d i,e -D e represents the distance deviation between the i th monitoring point and the e th monitoring point at the next monitoring time after the current monitoring time, |d i,e -D e represents the absolute value of d i,e -D e , and ε is used to ensure that the fraction is meaningful, which is specified as ε = 0.01, Z eThis represents the monitoring value of the e-th monitoring point at the current monitoring time. The larger this value is, the greater its influence on the echo intensity of the i-th monitoring point. norm() represents the normalization function. In this embodiment, the normalization in each place can be implemented using the same method, such as linear normalization. This means that when the position, direction, and distance of the i-th monitoring point relative to the e-th monitoring point are similar to the rainfall movement trend of the e-th monitoring point, the i-th monitoring point is more significantly affected by the rainfall of the e-th monitoring point.

[0087] If the e-th monitoring point is not a monitoring point within a rainfall cell, then there is no movement trend, and it is considered that it will not affect the i-th monitoring point. In this case, the setting is... It is 0.

[0088] The prediction correction and early warning module 14 is used to determine the corrected prediction value of any monitoring point at the current monitoring time at the next monitoring time based on the predicted value of any monitoring point at the current monitoring time at the next monitoring time and the degree of influence, so as to complete the rainfall early warning.

[0089] Based on the echo intensity data (monitored value) of each monitoring point 15 minutes prior to the current time, the predicted value Z of that monitoring point at the next monitoring time is obtained using the ARIMA algorithm. i,pre Because direct prediction using the forecasting algorithm cannot predict the movement of rainfall at other monitoring points, the predicted value is corrected by considering the degree of impact on the current monitoring point.

[0090] Z i,pre '=Z i,pre ×(1+H i )

[0091] Among them, Z i,pre Z represents the corrected predicted echo intensity value of the i-th monitoring point at the next monitoring time after the current monitoring time. i,pre This represents the predicted echo intensity value of the i-th monitoring point at the next monitoring time after the current monitoring time.

[0092] By correcting the predicted values ​​based on the echo intensity of each monitoring point at the next monitoring time after the current monitoring time, an early warning of rainfall in the current monitoring area can be issued. In this embodiment, preferably, when 30dBZ≤Z i,pre A heavy rain warning is issued when Z < 40 dBZ; a heavy rain warning is issued when 40 dBZ ≤ Z. i,pre A rainstorm warning is issued when the Z value is less than 50 dBZ; when the Z value is less than 50 dBZ, a rainstorm warning is issued. i,pre When the value is ≥50dBZ, a severe convective weather warning will be issued, as hail, thunderstorms, etc. may occur.

[0093] The embodiment of the present application firstly completes the prediction of the moving direction and the moving distance of each monitoring point at the predicted time after the current time by determining the moving track of the rainfall monomer, so as to determine the influence degree of other monitoring points on the monitoring point to be investigated at the next monitoring time by combining the motion direction deviation, the distance deviation between the monitoring point to be investigated and other monitoring points and the monitoring value of other monitoring points at the current time, and to represent the influence of the future weather change of the monitoring point to be investigated by other monitoring points which cannot be represented by the traditional prediction method under the condition of the fast changing weather, so as to more accurately complete the prediction of the monitoring value of the monitoring point to be investigated and improve the rainfall monitoring and early warning accuracy of the phased array radar under the condition of the fast changing weather.

[0094] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the same; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified or some technical features can be replaced by the equivalent ones; the modification or replacement does not make the essence of the corresponding technical solution deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A smart rainfall monitoring and early warning system based on phased array radar, characterized in that, The system includes: The historical data acquisition module is used to acquire the monitoring values ​​of each monitoring point in the monitoring area by the phased array radar within a set time period, and to determine the rainfall cells at each monitoring time. The monitoring values ​​are the echo intensity values. The rainfall cell trajectory determination module is used to take any rainfall cell at the current monitoring time as the target cell, determine the proximity of the target cell to each rainfall cell at the previous monitoring time, and determine the movement trajectory of the target cell at the set time period based on the proximity of the positions. The module for determining the degree of influence of the predicted value is used to determine the predicted value of the movement direction and the predicted value of the movement distance of each monitoring point in the target unit at the next monitoring time based on the movement trajectory, thereby determining the predicted value of the movement direction and the predicted value of the movement distance of the monitoring points in each rainfall unit at the current monitoring time, and further determining the degree of influence of any monitoring point on other monitoring points besides the rainfall unit where the monitoring point is located at the current monitoring time. The prediction correction and early warning module is used to determine the corrected prediction value of any monitoring point at the current monitoring time at the next monitoring time based on the predicted value of any monitoring point at the current monitoring time at the next monitoring time and the degree of impact, and to complete the rainfall early warning.

2. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, The determination of rainfall cells at each monitoring time includes: The monitoring points in the monitoring area whose monitored values ​​are greater than a preset monitoring threshold are clustered, and the clusters with an area greater than a preset area threshold are taken as the rainfall units.

3. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, Determining the proximity of the target cell to the location of each rainfall cell at the previous monitoring time includes: A predetermined number of edge monitoring points are continuously selected from the edge monitoring points of the target unit to form a target edge segment. The corresponding edge segment on each rainfall unit at the previous monitoring time is determined to be the same as the target edge segment. The proximity of each rainfall unit to the target unit at the previous monitoring time is determined based on the average distance between the target edge segment and the corresponding edge monitoring point in the corresponding edge segment.

4. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 3, characterized in that, The determination of the corresponding edge segment on each rainfall cell at the previous monitoring time that corresponds to the target edge segment includes: In the previous monitoring time, a preset number of consecutive edge monitoring points are randomly selected from the edge monitoring points of the current rainfall cell to form an edge segment to be judged. The distance value between the target edge segment and the edge segment to be judged is determined according to the average distance between the target edge segment and the corresponding edge monitoring points in the edge segment to be judged. The edge segment to be judged corresponding to the maximum distance value is taken as the corresponding edge segment in the current rainfall cell in the previous monitoring time.

5. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, Determining the movement trajectory of the target unit within the set time period based on the proximity of its location includes: The target cell is taken as the cell to be analyzed. At the previous monitoring time, a set number of rainfall cells that are closest to the target cell in location are selected as the undetermined rainfall cells. The undetermined rainfall cells at the previous time are each selected as new cells to be analyzed. At the monitoring time before the previous monitoring time, a set number of rainfall cells that are closest to the new cells to be analyzed are selected as new undetermined rainfall cells. The process of determining the undetermined rainfall cells corresponding to the target cell is repeated until all monitoring times within the set time period are traversed. Each combination of the target cell and the corresponding undetermined rainfall cells at each monitoring time is determined. The angular change and position change of the rainfall cells in any two adjacent monitoring times in the current cell combination are determined. The cell combination with the smallest angular change and position change is taken as the movement trajectory.

6. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, The method for determining the predicted movement direction and predicted movement distance of each monitoring point in the target unit at the next monitoring time based on the movement trajectory includes: Based on the angle change of the target unit at any two adjacent monitoring times in the movement trajectory, predict the predicted angle change of the target unit at the next monitoring time relative to the current monitoring time. Determine the predicted value of the movement direction based on the movement direction of the target unit at the current monitoring time compared to the previous monitoring time and the predicted angle change. Determine the predicted value of the movement distance based on the average movement distance of the target unit at any two adjacent monitoring times in the movement trajectory.

7. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 6, characterized in that, The step of predicting the change in the angle of the target unit relative to the current monitoring time at the next monitoring time based on the change in angle of the target unit at any two adjacent monitoring times in the movement trajectory includes: The angle change of the target unit at any two adjacent monitoring times in the movement trajectory is arranged into an angle change sequence according to time, and curve fitting is performed to obtain an angle change fitting curve. The value of the angle change fitting curve at the next monitoring time is used as the predicted angle change.

8. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, The further determination of the degree of influence of any monitoring point on other monitoring points besides the rainfall cell where the monitoring point is located at the current monitoring time includes: Taking any monitoring point at the current monitoring time as the target monitoring point, determine the direction angle from the monitoring point in any other rainfall cell other than the target monitoring point at the current monitoring time to the target monitoring point, and calculate the direction deviation between the direction angle and the predicted value of the movement direction of the monitoring point in any other rainfall cell other than the target monitoring point at the current monitoring time at the next monitoring time; Based on the distance between the target monitoring point and any monitoring point in any other rainfall cell other than the one where the target monitoring point is located at the current monitoring time, and the predicted movement distance of the monitoring point in any other rainfall cell other than the one where the target monitoring point is located at the current monitoring time at the next monitoring time, determine the distance deviation between the target monitoring point and any monitoring point in any other rainfall cell other than the one where the target monitoring point is located at the current monitoring time at the next monitoring time; The degree of influence is determined based on the directional deviation, the distance deviation, and the monitoring values ​​of monitoring points in any other rainfall cell other than the target monitoring point at the current monitoring time.

9. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 8, characterized in that, The degree of impact is as follows: ， in, This indicates the degree to which the i-th monitoring point is affected by other monitoring points besides the rainfall cell where the i-th monitoring point is located at the current monitoring time, at the next monitoring time. This represents the total number of remaining monitoring points within the monitoring area, excluding the rainfall cell containing the i-th monitoring point. This represents the direction angle from the e-th monitoring point to the i-th monitoring point at the current monitoring time. This represents the predicted movement direction of the e-th monitoring point at the current monitoring time at the next monitoring time. express and directional deviation between This represents the distance between the i-th monitoring point and the e-th monitoring point at the current monitoring time. This represents the predicted distance traveled by the e-th monitoring point at the current monitoring time at the next monitoring time. This represents the distance deviation between the i-th monitoring point and the e-th monitoring point at the next monitoring time after the current monitoring time. Indicates the request The absolute value, Used to ensure that fractions are meaningful This represents the monitoring value of the e-th monitoring point at the current monitoring time. This represents the normalization function.

10. The intelligent rainfall monitoring and early warning system based on phased array radar according to claim 1, characterized in that, Determining the predicted value of any monitoring point at the current monitoring time at the next monitoring time includes: Based on the monitoring values ​​of any monitoring point at the current monitoring time at each monitoring time within the set time period, the ARIMA algorithm is used to obtain the predicted value of any monitoring point at the current monitoring time at the next monitoring time.

Images