Meteorological early warning system based on unmanned aerial vehicle platform

Abstract

The invention relates to the technical field of meteorological early warning, and discloses a meteorological early warning system based on an unmanned aerial vehicle platform. The system comprises a data acquisition module, a rule identification module and an early warning control module. The data acquisition module acquires meteorological abnormal data of the meteorological monitoring equipment in a plurality of historical operation cycles, obtains a risk performance value through processing and analysis, compares the risk performance value with a risk performance threshold, generates a high risk signal if the risk performance value is greater than the threshold, and generates a low risk signal if not. And the rule identification module obtains an abnormal rule performance value based on the high-risk signal in combination with the meteorological abnormal time point and the meteorological abnormal data, judges whether a regular change exists or not, and generates a rule signal if the regular change exists. The early warning control module obtains a meteorological abnormal time period based on the rule signal, compares the meteorological abnormal time period with the current time point, controls the early warning process of the unmanned aerial vehicle, and generates a control stop signal or an early warning continuous signal. The system can optimize the early warning process of the unmanned aerial vehicle and improve the meteorological early warning effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of meteorological warning, in particular to a meteorological warning system based on a UAV platform. BACKGROUND

[0002] In the field of meteorological monitoring, timely and accurate warning is crucial for preventing natural disasters and ensuring the safety of production and life. Traditional meteorological warning systems rely on fixed meteorological monitoring equipment. These devices are distributed in specific areas and achieve warning functions by continuously collecting meteorological data. However, fixed monitoring equipment has many limitations in actual operation. Since the installation location is fixed, its monitoring range is often limited, making it difficult to cover complex terrain or areas with less human activity, resulting in meteorological anomalies in some areas that cannot be captured in time. Fixed equipment is exposed to the natural environment for a long time and is affected by extreme weather, equipment aging, and other factors, which can cause operational abnormalities and affect the accuracy and continuity of data collection.

[0003] When meteorological monitoring equipment malfunctions, traditional systems often fail to detect and handle it in a timely manner. In most cases, device abnormalities can only be detected by manual inspection, which not only consumes a lot of manpower and resources, but also delays the handling of abnormal situations due to long inspection cycles, causing meteorological warning to lag. In addition, the traditional system's judgment of device abnormalities is based on single real-time data feedback, lacking deep analysis of historical operation data, and unable to identify whether device abnormalities have regular changes. For example, some devices may frequently malfunction at certain times of the year or during certain time periods, but the traditional system cannot mine these time patterns, resulting in frequent false positives or false negatives in the warning process. In terms of warning control, the traditional system usually performs warning operations according to a fixed cycle or mode, lacking the ability to dynamically adjust according to device abnormality patterns. When monitoring equipment is prone to abnormalities at certain times, the system will still operate according to the regular process, which not only wastes resources such as UAVs for warning, but also may distort the warning results due to the continuous transmission of abnormal data. At the same time, for the application of UAV platforms in meteorological warning, existing technologies focus on the flight control and data collection functions of UAVs, but lack effective solutions on how to optimize the warning process in combination with device abnormality patterns. With the increase in the frequency and complexity of meteorological disasters, the limitations of traditional warning systems are increasingly evident. In the face of sudden meteorological anomalies, due to the inability to quickly identify device anomaly patterns, unmanned aerial vehicle platforms often fail to adjust warning strategies in a timely manner, resulting in low warning efficiency. For example, before the arrival of extreme weather such as typhoons and heavy rain, some monitoring devices may frequently exhibit data anomalies due to their own performance degradation. If the system cannot identify the time patterns of these anomalies, the unmanned aerial vehicle may continue to perform warning during the device anomaly period, delivering false information and affecting the accuracy of disaster prevention decisions. In addition, in traditional systems, the evaluation of device operation risks relies heavily on manual judgment, which is highly subjective and inefficient, making it difficult to meet the needs of large-scale meteorological monitoring networks. The existence of these problems seriously affects the timeliness and reliability of meteorological warnings, and cannot meet the requirements of modern society for precise meteorological services. SUMMARY

[0004] The purpose of the present application is to provide a meteorological warning system based on an unmanned aerial vehicle platform to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides a meteorological warning system based on an unmanned aerial vehicle platform, which comprises: a data acquisition module that acquires meteorological anomaly data of meteorological monitoring devices in a meteorological warning coverage area over multiple historical operation periods, processes and analyzes the meteorological anomaly data to obtain a risk performance value, judges the operation risk of the meteorological monitoring devices based on the risk performance value, and generates a risk signal, wherein the signal includes a high-risk signal and a low-risk signal; comparing the risk performance value with a risk performance threshold value; if the risk performance value is greater than the risk performance threshold value, a high-risk signal is generated; if the risk performance value is less than or equal to the risk performance threshold value, a low-risk signal is generated; a pattern recognition module that, based on the high-risk signal, acquires time data of meteorological anomalies of the meteorological monitoring devices over multiple historical operation periods, combines and analyzes the time data and the meteorological anomaly data to obtain an anomaly pattern performance value, judges whether the meteorological anomalies of the meteorological monitoring devices have regular changes based on the anomaly pattern performance value, and if so, generates a pattern signal; wherein the time data of meteorological anomalies includes time points of meteorological anomalies; a warning control module that, based on the pattern signal, acquires a meteorological anomaly period within a meteorological warning period and compares it with the current time point of the warning, controls the meteorological warning process of the unmanned aerial vehicle platform according to the comparison result, and generates a control signal, wherein the control signal includes a control stop signal and a warning continuation signal.

[0006] Preferably, the meteorological anomaly data comprises a meteorological anomaly frequency and a meteorological anomaly duration; Based on the analysis of the meteorological anomaly frequency and the meteorological anomaly duration, an anomaly frequency performance value and an anomaly time performance value are obtained respectively; The anomaly frequency performance value and the anomaly time performance value are summed to obtain a risk performance value.

[0007] Preferably, the meteorological anomaly frequencies of the meteorological monitoring equipment in all historical running periods are summed and averaged to obtain a meteorological anomaly frequency average value, the meteorological anomaly frequency average value is processed by ratio with the length of the historical running period to obtain a meteorological anomaly frequency, and the meteorological anomaly frequency is processed by ratio with a meteorological anomaly frequency threshold to obtain an anomaly frequency performance value.

[0008] Preferably, the meteorological anomaly durations of the meteorological monitoring equipment in all historical running periods are summed to obtain a total meteorological anomaly duration, the total meteorological anomaly duration is processed by ratio with the number of historical running periods to obtain an average meteorological anomaly duration in the historical running periods, and the average meteorological anomaly duration in the historical running periods is processed by ratio with the length of the historical running period to obtain an anomaly time performance value.

[0009] Preferably, the anomaly rule performance value is obtained in the following manner: The first rule performance value is marked as a first value; The second rule performance value is marked as a second value; The anomaly rule performance value is obtained by comprehensively processing the first value and the second value through a preset proportion coefficient.

[0010] Preferably, the first rule performance value and the second rule performance value are obtained in the following manner: After obtaining the frequency normal historical running period and the duration normal historical running period, a historical running period that is both the frequency normal historical running period and the duration normal historical running period is marked as a coincidence normal historical running period, and the meteorological anomaly time points in each coincidence normal historical running period are numbered and marked in time sequence; Based on the processing and analysis of the numbered and marked meteorological anomaly time points, a time interval deviation is obtained; The obtained time interval deviation is compared with a time interval deviation threshold respectively, and if the time interval deviation is less than or equal to the time interval deviation threshold, the time interval deviation is marked as a normal time interval deviation; Obtaining the number of normal time interval deviations, processing the number of normal time interval deviations and the number of time interval deviations by ratio, obtaining the normal time interval deviation number ratio, comparing the normal time interval deviation number ratio with the normal time interval deviation number ratio threshold value, if the normal time interval deviation number ratio is greater than the normal time interval deviation number ratio threshold value, marking the coincident normal historical running period as a comprehensive normal historical running period; Obtaining the number of comprehensive normal historical running periods, and processing the number of comprehensive normal historical running periods and the number of historical running periods by ratio, obtaining the first regularity performance value; Obtaining all interval reference times in the comprehensive normal historical running period, integrating the interval reference times into an interval reference time group, obtaining the variance value of all interval reference time groups and summing and averaging the variance values to obtain the variance mean value, marking the variance mean value as the second regularity performance value.

[0011] Preferably, the frequency abnormal historical running period is obtained in the following manner: Integrating the meteorological anomaly frequencies of the meteorological monitoring equipment in all historical running periods into a meteorological anomaly frequency data group, obtaining the mode of the meteorological anomaly frequency data group and marking the mode as a meteorological anomaly reference frequency, comparing the meteorological anomaly frequencies in the meteorological anomaly frequency data group with the meteorological anomaly reference frequency, if the meteorological anomaly frequencies are equal to the meteorological anomaly reference frequency, marking the historical running period corresponding to the meteorological anomaly frequencies as a frequency normal historical running period, if the meteorological anomaly frequencies are not equal to the meteorological anomaly reference frequency, marking the historical running period corresponding to the meteorological anomaly frequencies as a frequency abnormal historical running period; The duration abnormal historical running period is obtained in the same manner as the frequency abnormal historical running period.

[0012] Preferably, the abnormal regularity performance value is compared with the abnormal regularity performance threshold value; If the abnormal regularity performance value is greater than the abnormal regularity performance threshold value, a regularity signal is generated; If the abnormal regularity performance value is less than or equal to the abnormal regularity performance threshold value, no operation is performed.

[0013] Preferably, the meteorological anomaly time period in the meteorological warning period is obtained in the following manner: Obtaining the start time point and the end time point of the meteorological warning period and marking the time points on a time axis, obtaining the time point of the nearest meteorological monitoring equipment meteorological anomaly before the start time point of the meteorological warning period and marking the time point as a reference point on the time axis; Summing and averaging the interval times in all interval reference time groups to obtain an interval reference time reference value, marking the interval on the time axis with the reference point as the starting point and the interval reference time reference value as the interval distance, and marking the marked point as a meteorological anomaly start point; marking points of the marked intervals on the time axis as meteorological anomaly ending points; marking a time period between the meteorological anomaly starting point and the adjacent meteorological anomaly ending point as a meteorological anomaly time period, marking a time period between the starting time point and the ending time point of the meteorological warning period as a meteorological warning time period, and marking an overlapping time period between the meteorological anomaly time period and the meteorological warning time period as a meteorological anomaly time period in the meteorological warning period.

[0014] Preferably, when the unmanned aerial vehicle platform needs to be warned, the current time point of the warning is obtained, the current time point of the warning is compared with the meteorological anomaly time period in the meteorological warning period, if the current time point of the warning is in the meteorological anomaly time period in the meteorological warning period, a control stop signal is generated, and if the current time point of the warning is not in the meteorological anomaly time period in the meteorological warning period, a to-be-analyzed signal is generated. Based on the to-be-analyzed signal, the current time point of the warning is marked on the time axis, based on the time axis, the nearest meteorological anomaly starting point adjacent to the current time point of the warning is obtained, the corresponding interval time between the current time point of the warning and the nearest meteorological anomaly starting point is obtained, and the interval time is marked as a warning limit time, and the warning limit time is compared with the warning reference time. If the warning limit time is greater than or equal to the warning reference time, a warning duration signal is generated. If the warning limit time is less than the warning reference time, a control stop signal is generated.

[0015] Compared with the prior art, the present application has the following beneficial effects: Through the cooperative work of the data acquisition module, the rule identification module and the warning control module, many problems existing in the traditional meteorological warning system are effectively solved. The data acquisition module can process and analyze the meteorological anomaly data of the meteorological monitoring equipment in multiple historical operation periods, obtain the risk performance value, and generate the corresponding risk signal by comparing with the risk performance threshold. This process enables the system to automatically judge the operation risk of the equipment, without relying on manual inspection, reduces the interference of human factors, and enables the equipment abnormality to be discovered in time. When the equipment is in a high-risk operation state, the system can quickly capture and provide direction for subsequent processing, avoiding the delay of warning due to the failure to detect equipment abnormalities. The regularity identification module analyzes the meteorological anomaly time data of the meteorological monitoring device in combination with the abnormal data based on the high-risk signal to obtain an abnormal regularity performance value, and then judges whether the abnormality has regular changes and generates a regularity signal. This design breaks through the limitation of traditional systems relying only on real-time data for early warning, and through deep mining of historical data, it can discover the potential regularity of device anomalies in time. For example, some devices may frequently appear abnormal at a specific time period in a specific season, and the regularity identification module can accurately identify this regularity to provide a basis for subsequent early warning control. This grasp of regularity enables the system to predict in advance the time period when the device may appear abnormal, avoiding distorted early warning data caused by device anomalies, and making the early warning results more valuable. The early warning control module obtains the meteorological abnormal period according to the regularity signal, compares it with the current time point, and then controls the early warning process of the unmanned aerial vehicle platform and generates a control signal. This mechanism realizes the dynamic adjustment of the unmanned aerial vehicle early warning process. When the current time point is in the meteorological abnormal period, the system can issue a control stop signal to avoid the unmanned aerial vehicle continuing to perform invalid early warning in the device abnormal period, reducing unnecessary resource consumption. In the non-abnormal period, the system issues a warning continuation signal to ensure that the unmanned aerial vehicle can continuously and effectively perform meteorological monitoring and early warning, improving the pertinence and efficiency of the early warning work. Through this intelligent control method, the operation of the unmanned aerial vehicle platform is more reasonable, which can play the maximum role in the effective period and avoid invalid work in the device abnormal period, so that the operation of the entire meteorological early warning system is more efficient and accurate, and better meets the complex and variable meteorological monitoring needs. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A working principle diagram of the meteorological early warning system based on the unmanned aerial vehicle platform is described. Figure 2 A design diagram for calculating the risk performance value is described. Figure 3 A design diagram for calculating the abnormal frequency performance value is described. Figure 4 A design diagram for calculating the abnormal time performance value is described. Figure 5 A design diagram for generating the early warning control signal is described. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0018] Please refer toFigures 1-5 The application provides a weather warning system based on a UAV platform, which comprises a data acquisition module, a rule identification module and a warning control module, and each module works together to realize weather warning based on the UAV platform. The specific implementation steps are as follows: The working process of the data acquisition module is as follows: the weather anomaly data of the weather monitoring equipment in the weather warning coverage area in multiple historical operation cycles is obtained. The risk performance value is obtained by processing and analyzing these weather anomaly data. The risk performance value is compared with the risk performance threshold value, and the operation risk of the weather monitoring equipment is judged according to the comparison result and a risk signal is generated. If the risk performance value is greater than the risk performance threshold value, a high risk signal is generated; if the risk performance value is less than or equal to the risk performance threshold value, a low risk signal is generated.

[0019] The working process of the rule identification module is as follows: based on the high risk signal generated by the data acquisition module, the time data of weather anomalies of the weather monitoring equipment in multiple historical operation cycles is obtained, wherein the time data of weather anomalies includes the time points of weather anomalies. The time data and the weather anomaly data are combined for analysis to obtain an abnormal rule performance value. Whether the weather anomaly of the weather monitoring equipment has regular changes is judged according to the abnormal rule performance value, and if so, a rule signal is generated.

[0020] The working process of the warning control module is as follows: based on the rule signal generated by the rule identification module, the weather anomaly period in the weather warning cycle is obtained, and the period is compared with the current time point of the warning. The weather warning process of the UAV platform is controlled according to the comparison result, and a control signal is generated, including a control stop signal and a warning continuation signal.

[0021] Embodiment 1: The weather anomaly data covers the weather anomaly frequency and the weather anomaly duration. In the process of processing these data to obtain the risk performance value by the data acquisition module, the weather anomaly frequency and the weather anomaly duration are analyzed first to obtain the abnormal frequency performance value and the abnormal time performance value, and then the two performance values are added to obtain the risk performance value.

[0022] In processing the weather anomaly frequency to obtain the abnormal frequency performance value, the weather anomaly frequency recorded by the weather monitoring equipment in all historical operation cycles needs to be collected first. These frequencies are added and then divided by the number of historical operation cycles to obtain the average value of weather anomaly frequency. Then, the average value of weather anomaly frequency is compared with the length of a single historical operation cycle to obtain the weather anomaly frequency. Then, the weather anomaly frequency is compared with the preset weather anomaly frequency threshold value, and the result obtained is the abnormal frequency performance value.

[0023] For the processing of the duration of meteorological anomalies to obtain the anomaly time performance value, the total duration of meteorological anomalies occurring in the meteorological monitoring equipment in all historical running periods needs to be accumulated first. Then, the total duration of meteorological anomalies is divided by the number of historical running periods to obtain the average duration of meteorological anomalies in each historical running period. Then, the average duration of meteorological anomalies is divided by the length of a single historical running period, and the result is the anomaly time performance value.

[0024] After the data acquisition module calculates the anomaly frequency performance value and the anomaly time performance value respectively through the above-mentioned manner, it performs a summation operation on the two values, and the sum obtained is the risk performance value. Then, the data acquisition module compares the risk performance value with the risk performance threshold value. When the risk performance value is greater than the risk performance threshold value, a high-risk signal is generated; when the risk performance value is less than or equal to the risk performance threshold value, a low-risk signal is generated.

[0025] Through such a processing flow, the data acquisition module can comprehensively consider the frequency and duration of meteorological anomalies as two key factors to evaluate the running risk of the meteorological monitoring equipment. The frequency of meteorological anomalies reflects the frequency of occurrence of anomalies, while the duration of meteorological anomalies reflects the length of time affected by anomalies. The combination of the two makes the calculation of the risk performance value more comprehensive, so as to more accurately judge whether the meteorological monitoring equipment is in a high-risk running state and generate the corresponding risk signal, providing basic data for the work of the subsequent rule identification module.

[0026] In actual operation, the number of historical running periods can be set according to actual needs, for example, the running periods in the past month, quarter or year can be selected as the source of historical data. The length of each historical running period can also be determined according to the accuracy requirements of meteorological monitoring, which can be one day, one week, or other appropriate lengths of time. The meteorological anomaly frequency threshold value and the risk performance threshold value are values pre-set according to the characteristics of the past meteorological data in the region, the performance parameters of the meteorological monitoring equipment and the relevant meteorological warning standards. The setting of these threshold values needs to consider multiple factors to ensure the rationality and accuracy of risk assessment.

[0027] In addition, in the process of summing and averaging the frequency of meteorological anomalies, it is necessary to ensure that the frequency of meteorological anomalies in all historical running periods is accurately counted, avoiding omission or repeated calculation. Similarly, when calculating the total duration of meteorological anomalies, it is also necessary to ensure that the duration of meteorological anomalies in each historical running period is accurately recorded, so as to obtain reliable average duration of meteorological anomalies and ensure the authenticity and effectiveness of the calculation result of the anomaly time performance value.

[0028] In the embodiment 2, the abnormal regularity performance value is obtained by combining the first regularity performance value and the second regularity performance value with a preset proportion coefficient, wherein the first regularity performance value is marked as a first value, and the second regularity performance value is marked as a second value.

[0029] When obtaining the first regularity performance value and the second regularity performance value, firstly, the frequency normal historical running period and the duration normal historical running period need to be determined. The frequency normal historical running period refers to a historical running period in which the frequency of meteorological anomalies is equal to the reference frequency of meteorological anomalies, and the duration normal historical running period refers to a historical running period in which the duration of meteorological anomalies is equal to the reference duration of meteorological anomalies. The historical running period that belongs to both the frequency normal historical running period and the duration normal historical running period is marked as a coincidence normal historical running period.

[0030] For each coincidence normal historical running period, the time points of all meteorological anomalies in the coincidence normal historical running period are collected, and the time points are numbered in chronological order. Based on the numbered time points, the time interval between adjacent two time points is calculated, and the time interval is compared with a preset standard time interval to obtain a time interval deviation. Each time interval deviation is compared with a time interval deviation threshold value, and if the time interval deviation is less than or equal to the time interval deviation threshold value, the time interval deviation is marked as a normal time interval deviation.

[0031] The number of normal time interval deviations in all time interval deviations is counted, and the number is divided by the total number of time interval deviations to obtain a normal time interval deviation ratio. The normal time interval deviation ratio is compared with a preset normal time interval deviation ratio threshold value, and if the normal time interval deviation ratio is greater than the normal time interval deviation ratio threshold value, the corresponding coincidence normal historical running period is marked as a comprehensive normal historical running period.

[0032] The number of comprehensive normal historical running periods is calculated, and the number is divided by the total number of historical running periods to obtain the first regularity performance value. At the same time, the interval reference times in all comprehensive normal historical running periods are collected, and the interval reference times are grouped by period to form a plurality of interval reference time groups. The variance value of each interval reference time group is calculated, and the sum of all variance values is taken to obtain the mean value of the variance, which is the second regularity performance value.

[0033] After obtaining the first regularity performance value and the second regularity performance value, the two values are comprehensively calculated according to a preset proportion coefficient to obtain the abnormal regularity performance value. The preset proportion coefficient can be adjusted according to the emphasis degree of time interval stability and overall distribution consistency in actual application. For example, if more attention is paid to the stability of the time interval, the proportion of the first regularity performance value in the comprehensive calculation can be increased; if more emphasis is placed on the consistency of the overall distribution, the weight of the second regularity performance value can be increased.

[0034] In Example 3, the acquisition of the frequency abnormality historical operation period needs to go through multiple steps. First, collect the frequency of meteorological abnormalities recorded by the meteorological monitoring device in all historical operation periods, and integrate these frequencies into a meteorological abnormality frequency data set. Each element in the data set corresponds to the frequency of meteorological abnormalities in a historical operation period. For example, if the meteorological monitoring device has 5 meteorological abnormalities in a certain historical operation period, the corresponding element in the data set for this period is 5. Then, count the number of the most frequent value in the meteorological abnormality frequency data set. This value is the mode of the meteorological abnormality frequency data set, which is marked as the meteorological abnormality reference frequency.

[0035] After determining the meteorological abnormality reference frequency, compare each meteorological abnormality frequency in the meteorological abnormality frequency data set with the meteorological abnormality reference frequency. If the meteorological abnormality frequency of a certain historical operation period is equal to the meteorological abnormality reference frequency, then this historical operation period is marked as a frequency normal historical operation period. If the meteorological abnormality frequency of a certain historical operation period is not equal to the meteorological abnormality reference frequency, then this historical operation period is marked as a frequency abnormality historical operation period. For example, if the meteorological abnormality reference frequency is 3, then all historical operation periods with a meteorological abnormality frequency of 3 are frequency normal historical operation periods, while historical operation periods with a meteorological abnormality frequency of 2 or 4 are frequency abnormality historical operation periods.

[0036] The acquisition method of the duration abnormality historical operation period is the same as that of the frequency abnormality historical operation period. First, integrate the meteorological abnormality duration of the meteorological monitoring device in all historical operation periods into a meteorological abnormality duration data set. Each element in the data set represents the meteorological abnormality duration in a historical operation period, which can be in minutes, hours, etc. Then, find the mode of the meteorological abnormality duration data set and mark it as the meteorological abnormality reference duration.

[0037] Subsequently, compare each meteorological abnormality duration in the meteorological abnormality duration data set with the meteorological abnormality reference duration. If the meteorological abnormality duration of a certain historical operation period is equal to the meteorological abnormality reference duration, then this historical operation period is marked as a duration normal historical operation period. If the meteorological abnormality duration of a certain historical operation period is not equal to the meteorological abnormality reference duration, then this historical operation period is marked as a duration abnormality historical operation period. For example, if the meteorological abnormality reference duration is 60 minutes, then all historical operation periods with a meteorological abnormality duration of 60 minutes are duration normal historical operation periods, while historical operation periods with a meteorological abnormality duration of 40 minutes or 80 minutes are duration abnormality historical operation periods.

[0038] In the above process, the determination of the mode is a key link, and its calculation formula is: M = argmax x {count(x)} Wherein, M represents the mode, x represents the element in the data set, count(x) represents the number of times that the element x appears in the data set, and argmax x represents finding the corresponding x value when count(x) reaches the maximum value. Through the formula, the highest frequency value can be filtered out from the data set, which is used as a reference standard to distinguish between normal and abnormal historical running periods.

[0039] In this way, from a large number of historical running period data, periods that exhibit common characteristics in terms of meteorological anomaly frequency and duration, as well as those that have obvious differences, can be quickly filtered out. These classification results will provide basic data for subsequent determination of coincident normal historical running periods, comprehensive normal historical running periods, etc., so that the subsequent rule identification process can be based on more representative data samples, thereby more accurately capturing the potential rules of meteorological anomalies. At the same time, the unified acquisition logic ensures that the judgment criteria for frequency and duration anomaly periods are consistent, reduces errors caused by method differences, and makes the entire system's data analysis process more rigorous.

[0040] Embodiment 4: After the rule identification module obtains the abnormal rule performance value, it compares it with the preset abnormal rule performance threshold. The abnormal rule performance threshold is a benchmark value preset according to the historical meteorological data characteristics of the meteorological warning coverage area, the running characteristics of the meteorological monitoring equipment, and the actual warning requirements.

[0041] When the abnormal rule performance value is greater than the abnormal rule performance threshold, the rule identification module generates a rule signal. The core of this process is to determine whether the meteorological anomaly presents a sufficiently significant rule through comparison. For example, in the meteorological monitoring of a certain area, if the abnormal rule performance value obtained through analysis is 0.75, and the preset abnormal rule performance threshold is 0.6, since 0.75 is greater than 0.6, the rule identification module will generate a rule signal. The generated rule signal will be transmitted to the warning control module as an important basis for subsequent warning control operations.

[0042] If the abnormal rule performance value is less than or equal to the abnormal rule performance threshold, the rule identification module does not perform any operation. In this case, it means that the meteorological anomaly regularity of the meteorological monitoring equipment is not obvious enough to support targeted warning adjustments. For example, when the abnormal rule performance value is 0.55 and the abnormal rule performance threshold is 0.6, 0.55 is less than 0.6, and the rule identification module will not generate a rule signal, and the warning control module will perform meteorological warning control according to the normal mode.

[0043] The setting of the abnormal regularity performance threshold value needs to consider multiple factors. Different meteorological warning coverage areas have different meteorological conditions. Some areas have stable meteorological changes, and the abnormal regularity is easy to appear, so the threshold value can be appropriately increased. Some areas have complex meteorological changes, and it is difficult to capture the abnormal regularity, so the threshold value can be appropriately reduced. At the same time, the type and performance of the meteorological monitoring equipment will also affect the setting of the threshold value. High-precision equipment can capture more subtle regularities, and the threshold value can be adjusted accordingly to adapt to its characteristics.

[0044] When the rule identification module compares, it relies on the accuracy of the abnormal regularity performance value calculated in advance. The abnormal regularity performance value is calculated by synthesizing the first regularity performance value and the second regularity performance value according to the preset proportion coefficient, and its value directly reflects the strength of the meteorological abnormal regularity. The first regularity performance value is based on the proportion of the number of normal historical operation cycles, which reflects the proportion of cycles that conform to the overall regularity. The second regularity performance value is based on the variance mean of the interval reference time group, which reflects the stability of the time interval distribution. The comprehensive calculation result of the two can comprehensively measure the regularity of meteorological anomalies.

[0045] When the rule identification module generates a regularity signal, the warning control module will immediately start the corresponding processing flow, obtain the meteorological abnormal period within the meteorological warning period based on the regularity signal, and compare it with the current time point, and then generate a control signal. If no regularity signal is generated, the warning control module maintains the original warning logic and does not make additional adjustments. This design allows the system to flexibly adjust the warning strategy according to the actual regularity of meteorological anomalies, ensuring accurate control when there is a clear regularity to follow, and maintaining the regular warning mode when the regularity is not obvious.

[0046] By strictly following the comparison result of the abnormal regularity performance value and the threshold value to perform operations, the system's response to meteorological anomaly regularity is consistent and objective, avoiding the uncertainty brought by human intervention. At the same time, this judgment method based on data and threshold makes the system's operation logic clear and traceable, facilitating subsequent analysis and optimization of the warning process.

[0047] In the meteorological warning period, the meteorological abnormal period needs to be obtained in a series of steps. The starting time point and the ending time point of the meteorological warning period are determined, for example, the period of a certain meteorological warning is set as 2024-07-01 00:00 to 2024-07-07 24:00, and these two time points are marked on the time axis to form a clear warning time range. In the historical data before the starting time point of the meteorological warning period, find the time point closest to the starting time point when the meteorological monitoring equipment appears meteorological anomaly, assuming that the time point is 2024-06-30 18:00, then mark this time point as a reference point on the time axis.

[0048] The interval reference time group is processed, the interval times in all interval reference time groups are collected, the interval times are added and averaged to obtain the interval reference time reference value. If the interval times in the interval reference time group are 12 hours, 14 hours and 10 hours respectively, the total sum is 36 hours, and the average value of the interval reference time reference value is 12 hours. Taking the reference point of June 30, 2024, 18:00 as the starting point, mark the points on the time axis according to the interval distance of 12 hours, and these mark points are the meteorological anomaly starting points, for example, July 1, 2024, 6:00, 18:00, etc. After each meteorological anomaly starting point, mark the meteorological anomaly ending point with the meteorological anomaly reference duration as the interval distance. If the meteorological anomaly reference duration is 2 hours, the meteorological anomaly ending point corresponding to July 1, 2024, 6:00 is 8:00, and the ending point corresponding to 18:00 is 20:00.

[0049] Mark the period between each meteorological anomaly starting point and the adjacent meteorological anomaly ending point as a meteorological anomaly period, such as 6:00-8:00, 18:00-20:00, etc. At the same time, mark the period between the starting time point and the ending time point of the meteorological warning period as the meteorological warning period, i.e. from 00:00 on July 1, 2024 to 24:00 on July 7, 2024. By comparing the meteorological anomaly period and the meteorological warning period, the overlapping part is the meteorological anomaly period in the meteorological warning period, for example, the above-mentioned 6:00-8:00, 18:00-20:00, etc. If it falls within the warning period, it will be marked as the meteorological anomaly period in the period.

[0050] When the unmanned aerial vehicle platform needs to carry out meteorological warning, the current warning time point is obtained, assuming that the current time point is July 1, 2024, 7:00, which is compared with the meteorological anomaly period in the meteorological warning period, since 7:00 is in the meteorological anomaly period of 6:00-8:00, a control stop signal is generated at this time. If the current time point is July 1, 2024, 9:00, the time point is not in any meteorological anomaly period, a signal to be analyzed is generated. Based on the signal to be analyzed, 9:00 is marked on the time axis, the nearest meteorological anomaly starting point adjacent to it is found, i.e. 18:00, and the interval time between them is calculated as 9 hours, which is the warning limit time. The warning limit time of 9 hours is compared with the warning reference time, if the warning reference time is 5 hours, since 9 hours is greater than 5 hours, a warning duration signal is generated; if the warning reference time is 10 hours, since 9 hours is less than 10 hours, a control stop signal is generated.

[0051] During the whole process, the mark of time axis provides intuitive reference for the comparison of each time point and time period, ensuring the accuracy of meteorological anomaly period judgment. The calculation of interval reference time benchmark value and meteorological anomaly reference duration is based on the analysis result of historical data, so that the predicted meteorological anomaly starting point and ending point have certain rationality. Through the comparison of current time point and meteorological anomaly period, and the comparison of early warning limit time and early warning reference time, the system can generate corresponding control signal, realizing effective control of unmanned aerial vehicle platform meteorological early warning process.

[0052] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0053] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, alternatives, and variations can be made in the embodiments without departing from the spirit and scope of the present application as defined by the appended claims and their equivalents.

Claims

1. A weather warning system based on an unmanned aerial vehicle (UAV) platform, characterized in that: include: Data acquisition module: acquires meteorological anomaly data from meteorological monitoring equipment within the meteorological warning coverage area over multiple historical operating cycles, processes and analyzes the meteorological anomaly data to obtain risk performance values, judges the operational risk of meteorological monitoring equipment based on risk performance values, and generates risk signals, including high-risk signals and low-risk signals. Compare the risk performance value with the risk performance threshold; If the risk performance value is greater than the risk performance threshold, a high-risk signal is generated; If the risk performance value is less than or equal to the risk performance threshold, a low risk signal is generated; Pattern recognition module: Based on high-risk signals, it acquires time data of meteorological anomalies from meteorological monitoring equipment over multiple historical operating cycles. By combining and analyzing the time data and meteorological anomaly data, it obtains anomaly pattern performance values. Based on the anomaly pattern performance values, it determines whether there are regular changes in meteorological anomalies from meteorological monitoring equipment. If so, it generates a pattern signal. Among them, the time data of meteorological anomalies includes the time points of meteorological anomalies; Early warning control module: Based on regular signals, it obtains the abnormal weather periods within the weather warning cycle, compares them with the current time of the warning, controls the weather warning process of the UAV platform according to the comparison results, and generates control signals, including control stop signals and warning continuation signals.

2. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 1, characterized in that: Meteorological anomaly data includes the frequency of meteorological anomalies and the duration of meteorological anomalies; Based on the analysis of the frequency and duration of meteorological anomalies, the anomaly frequency performance value and the anomaly time performance value are obtained respectively. The risk performance value is obtained by summing the abnormal frequency performance value and the abnormal time performance value.

3. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 2, characterized in that: The average frequency of meteorological anomalies is obtained by summing the frequencies of meteorological anomalies from all meteorological monitoring equipment during all historical operating cycles. The average frequency of meteorological anomalies is then compared with the duration of the historical operating cycle to obtain the meteorological anomaly frequency. Finally, the meteorological anomaly frequency is compared with the meteorological anomaly frequency threshold to obtain the anomaly frequency performance value.

4. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 2, characterized in that: The duration of meteorological anomalies from all meteorological monitoring equipment within all historical operating cycles is summed to obtain the total duration of meteorological anomalies. The total duration of meteorological anomalies is then compared with the number of historical operating cycles to obtain the average duration of meteorological anomalies within the historical operating cycles. Finally, the average duration of meteorological anomalies within the historical operating cycles is compared with the duration of the historical operating cycles to obtain the anomaly time performance value.

5. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 1, characterized in that: The method for obtaining abnormal pattern performance values ​​is as follows: The value exhibited by the first rule is marked as the first value; The value representing the second pattern is marked as the second value; The abnormal pattern manifestation value is obtained by comprehensively considering the preset proportional coefficients.

6. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 5, characterized in that: The methods for obtaining the performance values ​​of the first and second rules are as follows: The frequency and duration of the normal historical operating cycle are obtained. The historical operating cycle that is both the frequency and duration normal historical operating cycle is marked as the overlapping normal historical operating cycle. The meteorological anomaly time points in each overlapping normal historical operating cycle are numbered and marked in chronological order. Based on the processing and analysis of the meteorological anomaly time points marked with numbers, the time interval deviation is obtained; The obtained time interval deviations are compared with the time interval deviation threshold. If the time interval deviation is less than or equal to the time interval deviation threshold, the time interval deviation is marked as a normal time interval deviation. The number of normal time interval deviations is obtained. The ratio of the number of normal time interval deviations to the number of time interval deviations is processed to obtain the normal time interval deviation ratio. The normal time interval deviation ratio is compared with the normal time interval deviation ratio threshold. If the normal time interval deviation ratio is greater than the normal time interval deviation ratio threshold, the overlapping normal historical operation cycle is marked as the comprehensive normal historical operation cycle. The number of normal historical operating cycles is obtained, and the ratio of this number to the number of historical operating cycles is processed to obtain the first regularity performance value; Obtain all interval reference times within the comprehensive normal historical operating cycle, integrate them into interval reference time groups, obtain the variance values ​​of all interval reference time groups, sum them and take the mean to obtain the variance mean, and mark the variance mean as the second regularity performance value.

7. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 1, characterized in that: The method for obtaining the historical operating cycle of frequency anomalies is as follows: The meteorological anomaly frequencies of all meteorological monitoring equipment within all historical operating cycles are integrated into a meteorological anomaly frequency data group. The mode of the meteorological anomaly frequency data group is obtained and marked as the meteorological anomaly reference frequency. The meteorological anomaly frequencies within the meteorological anomaly frequency data group are compared with the meteorological anomaly reference frequency. If the meteorological anomaly frequency is equal to the meteorological anomaly reference frequency, the historical operating cycle corresponding to the meteorological anomaly frequency is marked as a frequency normal historical operating cycle. If the meteorological anomaly frequency is not equal to the meteorological anomaly reference frequency, the historical operating cycle corresponding to the meteorological anomaly frequency is marked as a frequency anomaly historical operating cycle. The method for obtaining the historical operating cycle of duration anomalies is the same as that for frequency anomalies.

8. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 5, characterized in that: Compare the abnormal pattern performance value with the abnormal pattern performance threshold; If the value of the abnormal pattern is greater than the threshold for abnormal patterns, a pattern signal is generated; If the abnormal pattern value is less than or equal to the abnormal pattern threshold, no action will be taken.

9. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 1, characterized in that: The method for obtaining the abnormal weather periods within the aforementioned weather warning cycle is as follows: The start and end times of the weather warning cycle are obtained and marked on the timeline. The time of the most recent weather anomaly detected by the weather monitoring equipment before the start time of the weather warning cycle is obtained and marked on the timeline as a reference point. Sum the interval times within all interval reference time groups and take the average to obtain the interval reference time baseline value. Using the baseline point as the starting point and the interval reference time baseline value as the interval distance, mark the intervals on the time axis and mark the marked points as the starting points of meteorological anomalies. After each meteorological anomaly start point, interval markers are made on the time axis with the reference duration of the meteorological anomaly as the interval distance, and the marked points are marked as the meteorological anomaly end points; The time period between the start point of a meteorological anomaly and the adjacent end point of the meteorological anomaly is marked as the meteorological anomaly period. The time period between the start and end points of a meteorological warning cycle is marked as the meteorological warning period. The overlapping time period between the meteorological anomaly period and the meteorological warning period is marked as the meteorological anomaly period within the meteorological warning cycle.

10. A weather warning system based on an unmanned aerial vehicle (UAV) platform according to claim 9, characterized in that: When a drone platform needs to issue a weather warning, the current time of the warning is obtained and compared with the abnormal weather period within the weather warning cycle. If the current time of the warning is within the abnormal weather period within the weather warning cycle, a control stop signal is generated; if the current time of the warning is not within the abnormal weather period within the weather warning cycle, a signal to be analyzed is generated. Based on the signal to be analyzed, the current time point of the warning is marked on the time axis. Based on the time axis, the nearest meteorological anomaly start point after and adjacent to the current time point of the warning is obtained. The interval time between the current time point of the warning and the nearest adjacent meteorological anomaly start point is obtained and marked as the warning restriction time. The warning restriction time is compared with the warning reference time. If the warning restriction time is greater than or equal to the warning reference time, a warning continuity signal will be generated; If the warning time limit is less than the warning reference time, a control stop signal will be generated.

Images