Meteorological and hydrological analysis management cloud platform

Abstract

The invention relates to the technical field of data analysis and management, in particular to a meteorological and hydrological analysis and management cloud platform, which firstly realizes data storage and management through a meteorological information storage unit and a hydrological information storage unit in a meteorological and hydrological data storage module, and then realizes visual display of data through a data retrieval module. Particularly, the passive query unit automatically obtains abnormal data and retrieves related data, so that data support is provided for abnormal judgment; the anomaly analysis module obtains abnormal meteorological features and abnormal hydrological features, and calculates risk feature indexes in combination with forest vegetation density to judge whether the features are abnormal, and finally the management module carries out risk grading and detection task distribution on abnormal areas. According to the meteorological and hydrological analysis management cloud platform provided by the invention, efficient forest vegetation coverage monitoring and management are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis management, and in particular to a meteorological and hydrological analysis management cloud platform. Background Art

[0002] With the frequent occurrence of global climate change and extreme weather events, forest fires and floods have become a major issue that seriously threatens the ecological environment and social and economic development. As an important natural ecosystem, forests play an important role in maintaining carbon balance, regulating climate and protecting biodiversity. However, forest fires and floods not only cause huge losses of forest resources, but may also trigger secondary disasters, threatening the lives and property of residents in surrounding areas.

[0003] Traditional forest fire and flood monitoring mainly relies on manual inspections and fixed-site monitoring methods. However, due to the widespread distribution of forests and complex terrain, this method is inefficient and has limited coverage, making it difficult to capture early signals of sudden disasters in a timely manner. In addition, meteorological and hydrological conditions are key factors in causing fires and floods. For example, high temperature, low humidity, and strong winds can easily lead to the spread of fires, while continuous rainfall and rising river water levels may cause floods. Therefore, relying solely on a single data source is not enough to fully understand the dynamic environmental changes of forests. How to use advanced technical means, combined with meteorological and hydrological information, to achieve abnormal monitoring and rapid response to forest vegetation, thereby detecting forest disasters, has become an important issue that needs to be solved urgently.

[0004] Therefore, a meteorological and hydrological analysis and management cloud platform is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a meteorological and hydrological analysis and management cloud platform to improve the monitoring and management efficiency of forest vegetation coverage. First, the meteorological information storage unit and the hydrological information storage unit in the meteorological and hydrological data storage module are used to store and manage data, and then the data retrieval module is used to visualize the data, especially the passive query unit automatically obtains abnormal data and retrieves related data to provide data support for abnormal judgment; the abnormal analysis module obtains abnormal meteorological characteristics and abnormal hydrological characteristics, and calculates risk characteristic indicators in combination with forest vegetation density to determine whether it is abnormal. Finally, the management module performs risk classification and detection task allocation on abnormal areas, thereby realizing efficient monitoring and management of forest vegetation coverage.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A meteorological and hydrological analysis and management cloud platform, comprising:

[0008] The meteorological and hydrological data storage module includes a meteorological information storage unit and a hydrological information storage unit; among them, the meteorological information storage unit includes a meteorological hot data storage unit and a meteorological cold data storage unit, and the hydrological information storage unit includes a hydrological hot data storage unit and a hydrological cold data storage unit, and constructs data priorities according to forest vegetation density and collection time for data sorting;

[0009] Furthermore, the meteorological information storage unit is used to store data from all monitoring stations, and sorts the data according to the data priorities constructed based on forest vegetation density and collection time. The calculation formula for data priority is:

[0010]

[0011] where P i j represents the priority of the i-th meteorological data of the j-th monitoring station, and α 1 represents the weight coefficient of forest vegetation, represents the number of forest vegetation pixel points of the th monitoring station, represents the total number of pixel points of the j-th monitoring station, S represents the area represented by a pixel point, α, represents the weight coefficient of sampling time, e represents the natural logarithm base, k represents the time decay coefficient, t now represents the current time, and t i represents the collection time of the i-th data;

[0012] Determine the storage location according to data priority and threshold:

[0013]

[0014] where P threshold represents the threshold, and P i represents the data priority of the i-th data.

[0015] Furthermore, for the optimization of storage resources for meteorological and hydrological data, the allocation ratio is dynamically adjusted according to the importance of the data:

[0016]

[0017] where R represents the storage resource ratio of hot data to cold data, S h represents the storage resources currently used by the hot data unit, S t represents the total storage resources, P′ m represents the data priority of the m-th hot data, M represents the number of hot data records, P″ n represents the data priority of the n-th data, and N represents the total number of data records.

[0018] The data retrieval module includes an active query unit and a passive query unit. The active query unit includes querying according to keywords and displaying according to data priorities. The passive query unit includes obtaining the first meteorological information and the first hydrological information collected by the data monitoring station according to event triggers, associatively retrieving the second meteorological information according to the first meteorological information, associatively retrieving the second hydrological information according to the first hydrological information, interactively retrieving the third hydrological information according to the first meteorological information, and interactively retrieving the third meteorological information according to the first hydrological information; constructing an abnormal meteorological data vector according to the first, second, and third meteorological information, and constructing an abnormal hydrological data vector according to the first, second, and third hydrological information;

[0019] Further, the active query unit further includes: the active query unit is used to receive the keywords, time range, and geographical range actively input by the user and perform queries, and optimize the priority sorting and display in combination with the forest vegetation area and density; the functions of the active query unit further include: keyword retrieval and multi-condition cross-screening, and the multi-condition cross-screening includes supporting compound queries based on keywords, time range, monitoring station location, and forest vegetation coverage rate.

[0020] Further, the passive query unit further includes:

[0021] The formula for associatively retrieving the second meteorological information according to the first meteorological information is:

[0022]

[0023] Wherein, represents the second meteorological information, represents the first meteorological information, F weather represents meteorological characteristics, corr() represents the correlation function, and θ 1 represents the first meteorological correlation threshold;

[0024] The formula for interactively retrieving the third hydrological information according to the first meteorological information is:

[0025]

[0026] Wherein, represents the third hydrological information, F hydro represents hydrological characteristics, f() represents the interaction function, and θ 2 represents the first interaction threshold;

[0027] The formula for associatively retrieving the second hydrological information according to the first hydrological information is:

[0028]

[0029] Wherein, Represents the second hydrological information, Represents the first hydrological information, θ 3 Represents the second hydrological correlation threshold;

[0030] The formula for interactively retrieving the third meteorological information based on the first hydrological information is:

[0031]

[0032] Wherein, Represents the third meteorological information, θ 4 Represents the second interaction influence threshold.

[0033] Anomaly analysis module, configured to obtain the abnormal meteorological data vector and the abnormal hydrological data vector, extract features according to the temporal difference of the data, obtain abnormal meteorological features and abnormal hydrological features, and combine the forest vegetation density to obtain a risk feature index, and determine whether there is an anomaly according to the risk feature index;

[0034] Further, the specific steps for constructing the abnormal meteorological feature and the abnormal hydrological feature include:

[0035] Obtain the abnormal meteorological data vector and extract abnormal hot data and abnormal cold data to obtain an abnormal meteorological hot data component and an abnormal meteorological cold data component; obtain the abnormal meteorological hot data component collected within a plurality of time windows and use a sliding window to calculate the data change amount of adjacent time periods to obtain an abnormal meteorological hot data change matrix, extract features from the abnormal meteorological hot data change matrix using a trend analysis model to obtain abnormal meteorological hot data features, and obtain the abnormal meteorological hot data features and the abnormal meteorological cold data component to obtain abnormal meteorological features;

[0036] Obtain the abnormal hydrological data vector and extract abnormal hot data and abnormal cold data to obtain an abnormal hydrological hot data component and an abnormal hydrological cold data component; obtain the abnormal hydrological hot data component collected within a plurality of time windows and use a sliding window to calculate the data change amount of adjacent time periods to obtain an abnormal hydrological hot data change matrix, extract features from the abnormal hydrological hot data change matrix using a trend analysis model to obtain abnormal hydrological hot data features, and obtain the abnormal hydrological hot data features and the abnormal hydrological cold data component to obtain abnormal hydrological features.

[0037] Further, the calculation formula of the risk feature index is:

[0038]

[0039] Wherein, Dan represents the risk feature index, ω 1 Represents the weight coefficient of the abnormal meteorological feature, H represents the length of the abnormal meteorological feature, x hThe h-th component representing abnormal meteorological characteristics, μ x The mean of normal meteorological data, σ x The standard deviation of normal meteorological data, ω 2 The weight coefficient representing abnormal hydrological characteristics, G represents the length of abnormal hydrological characteristics, y g The g-th component representing abnormal hydrological characteristics, μ y The mean of normal hydrological data, σ y The standard deviation of normal hydrological data, ω 3 The weight coefficient representing forest vegetation density, ΔF density Represents the change in forest vegetation density.

[0040] The management module is used to classify according to risk characteristic indicators, and perform task allocation and status tracking.

[0041] Furthermore, the steps of classifying according to risk characteristic indicators and performing task allocation and status tracking also include: obtaining abnormal risk indicators and dividing the abnormal risk levels according to the priorities, and allocating detection time, detection personnel and detection tools according to the abnormal risk levels and historical detection data.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] 1. Design data priorities by combining the proportion of vegetation pixel points in the area where the monitoring station is located and the collection time of meteorological and hydrological data, and sort the data, so that the monitoring data in areas with high forest vegetation density is more likely to be processed preferentially, meeting the actual needs of forest anomaly monitoring; by comparing the data priorities and thresholds, store high-priority data in the hot data unit and low-priority data in the cold data unit, realizing the reasonable allocation of storage resources and improving the storage efficiency.

[0044] 2. Through the association retrieval and interactive retrieval mechanisms, the dynamic association and interactive retrieval between meteorological and hydrological data are realized, and relevant data can be quickly found, improving the efficiency and accuracy of retrieval; decompose the abnormal data vector into hot data and cold data components, realizing the fine processing of data, which helps to capture potential anomalies more precisely; through the sliding window and trend analysis model, extract the change matrix and trend characteristics from time series data, which can better reflect the dynamic change characteristics of meteorological and hydrological data, and construct comprehensive risk characteristic indicators, improving the risk assessment ability.

[0045] 3. By integrating the forest vegetation density into all aspects of the storage, retrieval, analysis, and management of meteorological and hydrological data, the cloud platform can more accurately identify high-risk areas, reasonably allocate resources, improve the efficiency of anomaly monitoring and disaster prevention and control, demonstrating the deep integration of data and forest ecological characteristics, reflecting the intelligence and innovation of the cloud platform, realizing real-time supervision and management of forest vegetation, and improving the early warning efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 FIG. is a schematic structural diagram of a meteorological and hydrological analysis and management cloud platform provided by an embodiment of the present invention;

[0047] Figure 2 FIG. is a schematic structural diagram of a meteorological and hydrological data storage module provided by an embodiment of the present invention;

[0048] Figure 3 FIG. is a flowchart for constructing abnormal meteorological characteristics and abnormal hydrological characteristics provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] Embodiment 1

[0051] A certain company introduced a meteorological and hydrological analysis and management cloud platform provided by the present invention during the process of monitoring and managing forest vegetation to improve management efficiency. The structure of the cloud platform is as Figure 1 shown, and the specific implementation is as follows:

[0052] First, a vegetation remote sensing supervision module is also introduced in the cloud platform, which is used to obtain forest remote sensing image data, process it, and obtain the forest vegetation density of the data monitoring station according to the pixel points of the remote sensing image.

[0053] The structure of the meteorological and hydrological data storage module is as Figure 2 shown, and it is used to store meteorological information and hydrological information, including a meteorological information storage unit and a hydrological information storage unit; wherein the meteorological information storage unit includes a meteorological hot data storage unit and a meteorological cold data storage unit; the hydrological information storage unit includes a hydrological hot data storage unit and a hydrological cold data storage unit, and the data is sorted according to the forest vegetation density and collection time of the data monitoring station; Table 1 shows some of the hydrological information stored in the hydrological information storage unit.

[0054] Table 1, Partial Hydrological Information

[0055] Monitoring station number Hydrological hot data Hydrological cold data A001 Water level: 5.3 m, flow velocity: 2.1 m / s, …… Average water level: 4.8 m, annual average flow velocity: 1.9 m / s, …… A002 Water level: 3.8 m, flow velocity: 1.5 m / s, …… Average water level: 3.5 m, annual average flow velocity: 1.4 m / s, …… A003 Water level: 6.7 m, flow velocity: 2.8 m / s, …… Average water level: 6.2 m, annual average flow velocity: 2.5 m / s, …… A004 Water level: 4.1 m, flow velocity: 1.8 m / s, …… Average water level: 3.9 m, annual average flow velocity: 1.6 m / s, …… A005 Water level: 5.9 m, flow velocity: 2.3 m / s, …… Average water level: 5.5 m, annual average flow velocity: 2.1 m / s, ……

[0056] Further, the meteorological information storage unit is specifically used to store data from all monitoring stations, and dynamically sort all the data according to the data priority constructed by combining the forest vegetation density and the collection time, which is beneficial for subsequent data display and call. The calculation formula of the data priority is:

[0057]

[0058] Among them, P i j represents the data priority of the i-th meteorological data of the j-th monitoring station, and α 1 represents the weight coefficient of the forest vegetation, represents the number of forest vegetation pixel points of the j-th monitoring station, represents the total number of pixel points of the j-th monitoring station, S represents the area represented by a pixel point, and α 2 represents the weight coefficient of the sampling time, e represents the natural logarithm base, k represents the time decay coefficient, t now represents the current time, and t i represents the collection time of the i-th data;

[0059] Further, determine the storage location of the meteorological data according to the data priority and the threshold:

[0060]

[0061] Among them, P threshold represents the threshold, and P i represents the data priority of the i-th meteorological data;

[0062] Further, the hydrological information storage unit is similar to the meteorological information storage unit, and both use the data priority method for storage and display.

[0063] By the method of sorting according to the data priority constructed by combining the forest vegetation density of the monitoring station and the data collection time, the priority of meteorological data and hydrological data can be divided, ensuring that key data can be stored and called preferentially, improving the data management efficiency while making full use of storage resources. Using the priority and the threshold to divide the storage location enables high-priority data to be stored in the fast access area and low-priority data to be stored in the low-cost area, significantly improving the performance and resource utilization rate of the storage system. At the same time, the division of hot and cold data provides support for the subsequent extraction of abnormal features.

[0064] Further, for the optimization of the storage resources of meteorological and hydrological data, the allocation ratio is dynamically adjusted according to the importance of the data:

[0065]

[0066] Among them, R represents the storage resource ratio of hot data to cold data, S h represents the storage resource currently used by the hot data unit, S t represents the total storage resource, P′ m represents the data priority of the m-th hot data, M represents the number of hot data records, P″ n represents the data priority of the n-th data, and N represents the total number of data records.

[0067] Dynamically adjusting the storage resource ratio of hot data and cold data according to the data priority can achieve reasonable allocation of storage resources, ensure that critical data can be accessed quickly, and enhance the adaptability and flexibility of data storage.

[0068] A data retrieval module, used to query data according to user needs, including an active query unit and a passive query unit. The active query unit includes querying according to keywords and performing priority retrieval and display according to the data priority of the data; the passive query unit includes obtaining the first meteorological information and the first hydrological information collected by the data monitoring station according to event triggering, associatively retrieving the second meteorological information according to the first meteorological information, associatively retrieving the second hydrological information according to the first hydrological information, interactively retrieving the third hydrological information according to the first meteorological information, interactively retrieving the third meteorological information according to the first hydrological information, constructing an abnormal meteorological data vector according to the first meteorological information, the second meteorological information and the third meteorological information, and constructing an abnormal hydrological data vector according to the first hydrological information, the second hydrological information and the third hydrological information;

[0069] Further, the active query unit further includes:

[0070] The active query unit supports users to actively input keywords, time range and geographical range for query, and performs sorting and display of data in combination with the data priority. Its functions include: keyword retrieval and multi-condition cross-screening, and the multi-condition cross-screening includes supporting compound queries based on keywords, time range, monitoring station location and forest vegetation coverage rate.

[0071] The active query unit combines the multi-condition cross-screening and data priority sorting functions, significantly improving the accuracy and efficiency of data retrieval, and at the same time enhancing the intelligence and flexibility of the system.

[0072] Further, obtaining the first meteorological information and the first hydrological information collected by the data monitoring station according to event triggering means that if some data indicators of the monitoring station, such as temperature and air pressure, exceed the set threshold, then the first meteorological information and the first hydrological information are obtained.

[0073] Furthermore, the passive query unit further includes:

[0074] The formula for associatively retrieving the second meteorological information based on the first meteorological information is:

[0075]

[0076] Wherein, represents the second meteorological information, represents the first meteorological information, F weather represents the meteorological feature, corr() represents the correlation function, and θ 1 represents the first meteorological correlation threshold;

[0077] Furthermore, the calculation formula of the correlation function is:

[0078]

[0079] Wherein, corr() represents the correlation function, X′ and Y′ respectively represent two groups of data vectors, and x′ q represents the q-th component of the first group of data vectors, and y′ q represents the q-th component of the second group of data vectors, Q represents the length of the data vector, and respectively represent the means of the two groups of data vectors.

[0080] The formula for interactively retrieving the third hydrological information based on the first meteorological information is:

[0081]

[0082] Wherein, represents the third hydrological information, F hydro represents the hydrological feature, f() represents the interaction function, and θ 2 represents the first interaction threshold;

[0083] Furthermore, the calculation formula of the interaction function is:

[0084]

[0085] Wherein, ρ q represents the weight coefficient of the q-th component, e represents the natural logarithm base, and κ represents the adjustment parameter for controlling the attenuation rate of the interaction intensity.

[0086] The formula for associatively retrieving the second hydrological information based on the first hydrological information is:

[0087]

[0088] Wherein, represents the second hydrological information represents the first hydrological information, θ 3 represents the second hydrological correlation threshold;

[0089] The formula for interactively retrieving the third meteorological information based on the first hydrological information is:

[0090]

[0091] wherein, represents the third meteorological information, θ 4 represents the second interaction influence threshold.

[0092] Associative retrieval can reveal the internal correlation hidden behind the data through the correlation function between data, enhancing the mining of data correlation; introducing interactive retrieval, a two-way influence relationship between meteorological and hydrological data is established, revealing the potential driving effect of meteorological changes on hydrological data and analyzing the feedback effect of hydrological characteristics on meteorology. This two-way interactive data retrieval helps to construct the joint characteristics of meteorology and hydrology, providing data support for the analysis of complex environmental systems.

[0093] The anomaly analysis module obtains the abnormal meteorological data vector and the abnormal hydrological data vector, extracts features according to the temporal difference change of the data to obtain abnormal meteorological features and abnormal hydrological features, and performs feature fusion with the forest vegetation density to obtain a risk feature index, and determines whether there is an anomaly according to the risk feature index;

[0094] Further, the specific steps for constructing the abnormal meteorological features and the abnormal hydrological features are as Figure 3 shown, including:

[0095] Obtain the abnormal meteorological data vector and extract the abnormal hot data and abnormal cold data therein to obtain the abnormal meteorological hot data component and the abnormal meteorological cold data component; obtain the abnormal meteorological hot data component collected in the recent several time windows and use a sliding window to compare the data change amounts in adjacent time periods to obtain an abnormal meteorological hot data change matrix, extract features from the abnormal meteorological hot data change matrix using a trend analysis model to obtain abnormal meteorological hot data features, and obtain the abnormal meteorological hot data features and the abnormal meteorological cold data component to obtain abnormal meteorological features;

[0096] Obtain the abnormal hydrological data vector, extract the abnormal heat data and abnormal cold data from it to obtain the abnormal hydrological heat data component and the abnormal hydrological cold data component; obtain the abnormal hydrological heat data components collected in several recent time windows and use a sliding window to compare the data change amounts in adjacent time periods to obtain the abnormal hydrological heat data change matrix, extract features from the abnormal hydrological heat data change matrix using a trend analysis model to obtain the abnormal hydrological heat data features, and obtain the abnormal hydrological heat data features and the abnormal hydrological cold data components to obtain the abnormal hydrological features.

[0097] Further, the trend analysis model can be existing models such as recurrent neural network models, convolutional neural network models, and autoregressive integrated moving average models for feature extraction.

[0098] First, the separation of heat data and cold data helps to analyze the data with significant short-term changes and the abnormal data with small long-term fluctuations but possible importance separately. Then, using a sliding window to compare the data change amounts in adjacent time periods can capture the dynamic evolution characteristics of abnormal data in a short time. Finally, extracting features through a trend analysis model effectively compresses the data dimension and retains key information; finally, fusing the abnormal heat data features and the abnormal cold data components forms a more comprehensive feature, enhancing the comprehensiveness of the abnormal features.

[0099] Further, the calculation formula for the risk feature index is:

[0100]

[0101] where Dan represents the risk feature index, ω 1 represents the weight coefficient of the abnormal meteorological feature, H represents the length of the abnormal meteorological feature, x h represents the h-th component of the abnormal meteorological feature, μ x represents the mean of the normal meteorological data, σ x represents the standard deviation of the normal meteorological data, ω 2 represents the weight coefficient of the abnormal hydrological feature, G represents the length of the abnormal hydrological feature, y g represents the g-th component of the abnormal hydrological feature, μ y represents the mean of the normal hydrological data, σ y represents the standard deviation of the normal hydrological data, ω 3 represents the weight coefficient of the forest vegetation density, ΔF density represents the change amount of the forest vegetation density. As shown in Table 2, the risk feature indexes of some monitoring stations all show normal.

[0102] Table 2. Risk Feature Indexes of Some Monitoring Stations

[0103] Monitoring station number Risk characteristic index A001 0.85 A002 0.72 A003 0.68 A004 0.76 A005 0.81

[0104] The risk characteristic index combines three key factors: abnormal meteorological characteristics, abnormal hydrological characteristics, and forest vegetation density, which helps to identify extreme weather events and abnormal hydrological changes, ensures the multi-dimensionality and reliability of the risk assessment results, improves the comprehensiveness, adaptability, and accuracy of risk identification, and provides strong support for disaster warning, ecological protection, and emergency decision-making.

[0105] The management module is used to obtain the location of the anomaly, define the priority of the task according to the risk characteristic index, and perform task allocation and status tracking.

[0106] Furthermore, the steps of grading according to the risk characteristic index and performing task allocation and status tracking also include: obtaining the abnormal risk index and dividing the abnormal risk level according to the priority, and allocating detection time, detection personnel, and detection tools according to the abnormal risk level and historical detection data.

[0107] By obtaining the abnormal risk index, dividing the priority according to the forest density and risk level, and dynamically allocating detection tasks in combination with historical data, the management module can significantly improve the scientificity, accuracy, and efficiency of the detection work.

[0108] A meteorological and hydrological analysis management cloud platform provided by the present invention first realizes the storage and call of meteorological data and hydrological data through the meteorological and hydrological data storage module, and sorts the data and allocates storage resources according to the data priority based on the forest vegetation density of the monitoring station and the collection time; then the data retrieval module realizes the automatic retrieval of abnormal data, and realizes the comprehensive retrieval of abnormal data through associated retrieval and interactive retrieval; then analyzes the abnormal data, constructs a risk characteristic index through abnormal meteorological characteristics, abnormal hydrological characteristics, and forest vegetation area, realizes the abnormal monitoring of forest vegetation, and improves the management efficiency.

[0109] Embodiment 2

[0110] A certain engineering unit introduced a meteorological and hydrological analysis management cloud platform provided by the present invention when monitoring a certain forest. The specific implementation method is as follows:

[0111] The meteorological and hydrological data storage module includes a meteorological information storage unit and a hydrological information storage unit; wherein the meteorological information storage unit includes a meteorological hot data storage unit and a meteorological cold data storage unit, and the hydrological information storage unit includes a hydrological hot data storage unit and a hydrological cold data storage unit, and sorts the data according to the data priority constructed based on the forest vegetation density and the collection time; Table 3 shows part of the meteorological information.

[0112] Table 3, Part of the Meteorological Information

[0113] Monitoring station number Meteorological hot data Meteorological cold data B001 Temperature: 24.3 °C, humidity: 65%, …… Precipitation: 0 mm, air pressure: 1011 hPa, …… B002 Temperature: 25.4 °C, humidity: 54%, …… Precipitation: 0 mm, air pressure: 1010 hPa, …… B003 Temperature: 24.8 °C, humidity: 59%, …… Precipitation: 0 mm, air pressure: 1009 hPa, …… B004 Temperature: 24.5 °C, humidity: 62%, …… Precipitation: 0 mm, air pressure: 1012 hPa, …… B005 Temperature: 24.1 °C, humidity: 64%, …… Precipitation: 0 mm, air pressure: 1010 hPa, ……

[0114] Furthermore, the meteorological information storage unit is used to store data from all monitoring stations, and constructs data priorities for data sorting by combining forest vegetation density and collection time. The calculation formula for data priority is:

[0115]

[0116] where P i j represents the priority of the i-th meteorological data of the j-th monitoring station, and α 1 represents the weight coefficient of forest vegetation, represents the number of forest vegetation pixel points of the j-th monitoring station, represents the total number of pixel points of the j-th monitoring station, S represents the area represented by a pixel point, and α 2 represents the weight coefficient of the sampling time, e represents the natural base, k represents the time decay coefficient, t now represents the current time, and t i represents the collection time of the i-th data;

[0117] Determine the storage location according to the data priority and the threshold:

[0118]

[0119] where P threshokd represents the threshold, and P i represents the data priority of the i-th data.

[0120] Furthermore, for the optimization of storage resources for meteorological and hydrological data, the allocation ratio is dynamically adjusted according to the importance of the data:

[0121]

[0122] where R represents the storage resource ratio of hot data to cold data, S h represents the storage resources currently used by the hot data unit, S t represents the total storage resources, P′ m represents the data priority of the m-th hot data, M represents the number of hot data records, P″ n represents the data priority of the n-th data, and N represents the total number of data records.

[0123] The data retrieval module includes an active query unit and a passive query unit. The active query unit includes querying according to keywords and displaying according to data priorities. The passive query unit includes obtaining the first meteorological information and the first hydrological information collected by the data monitoring station according to event triggers, associatively retrieving the second meteorological information according to the first meteorological information, associatively retrieving the second hydrological information according to the first hydrological information, interactively retrieving the third hydrological information according to the first meteorological information, and interactively retrieving the third meteorological information according to the first hydrological information; constructing an abnormal meteorological data vector according to the first, second, and third meteorological information, and constructing an abnormal hydrological data vector according to the first, second, and third hydrological information.

[0124] Further, the formula for associatively retrieving the second meteorological information according to the first meteorological information is:

[0125]

[0126] Where, represents the second meteorological information, represents the first meteorological information, F weather represents meteorological characteristics, corr() represents the correlation function, θ 1 represents the first meteorological correlation threshold;

[0127] Further, the formula for interactively retrieving the third hydrological information according to the first meteorological information is:

[0128]

[0129] Where, represents the third hydrological information, F hydro represents hydrological characteristics, f() represents the interactive influence function, θ 2 represents the first interactive influence threshold;

[0130] Further, the formula for associatively retrieving the second hydrological information according to the first hydrological information is:

[0131]

[0132] Where, represents the second hydrological information, represents the first hydrological information, θ 3 represents the second hydrological correlation threshold;

[0133] Further, the formula for interactively retrieving the third meteorological information according to the first hydrological information is:

[0134]

[0135] Where, represents the third meteorological information, θ 4Represents the second interaction influence threshold.

[0136] Anomaly analysis module, configured to obtain the abnormal meteorological data vector and the abnormal hydrological data vector, extract features based on the time series difference of the data to obtain abnormal meteorological features and abnormal hydrological features, and combine the forest vegetation density to obtain a risk feature index, and determine whether there is an anomaly according to the risk feature index;

[0137] Furthermore, the calculation formula of the risk feature index is:

[0138]

[0139] Where Dan represents the risk feature index, ω 1 represents the weight coefficient of the abnormal meteorological feature, H represents the length of the abnormal meteorological feature, x h represents the h-th component of the abnormal meteorological feature, μ x represents the mean value of the normal meteorological data, σ x represents the standard deviation of the normal meteorological data, ω 2 represents the weight coefficient of the abnormal hydrological feature, G represents the length of the abnormal hydrological feature, y g represents the g-th component of the abnormal hydrological feature, μ y represents the mean value of the normal hydrological data, σ y represents the standard deviation of the normal hydrological data, ω 3 represents the weight coefficient of the forest vegetation density, ΔF density represents the change in the forest vegetation density.

[0140] The management module is configured to classify according to the risk feature index, and perform task allocation and status tracking.

[0141] Furthermore, the formula of the classification function is:

[0142]

[0143] Where L represents the risk level, Dan 1 、Dan 2 and Dan 3 represent the first risk threshold, the second risk threshold and the third risk threshold respectively.

[0144] Table 4. Risk feature indexes of some monitoring stations

[0145] Monitoring station number Risk characteristic index B001 0.79 B002 0.81 B003 0.93 B004 0.78 B005 0.83

[0146] As shown in Table 4, the risk characteristic indicators of some monitoring stations are all less than the threshold values, indicating no risk. Through a meteorological and hydrological analysis and management cloud platform provided by the present invention, real-time monitoring of forest vegetation has been achieved, management efficiency has been improved, and the early warning ability for natural disasters has been enhanced.

[0147] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A meteorological and hydrological analysis and management cloud platform, characterized in that: include: The meteorological and hydrological data storage module includes a meteorological information storage unit and a hydrological information storage unit; wherein the meteorological information storage unit includes a meteorological hot data storage unit and a meteorological cold data storage unit, and the hydrological information storage unit includes a hydrological hot data storage unit and a hydrological cold data storage unit, and the data is sorted according to the data priority constructed according to the forest vegetation density and the collection time; The data retrieval module includes an active query unit and a passive query unit, wherein the active query unit is used to query according to keywords and display according to data priority; the passive query unit includes obtaining the first meteorological information and the first hydrological information collected by the monitoring station according to the event trigger; retrieving the second meteorological information according to the association of the first meteorological information, retrieving the second hydrological information according to the association of the first hydrological information, interactively retrieving the third hydrological information according to the first meteorological information, and interactively retrieving the third meteorological information according to the first hydrological information; constructing an abnormal meteorological data vector according to the first, second and third meteorological information, and constructing an abnormal hydrological data vector according to the first, second and third hydrological information; The abnormal analysis module is used to obtain the abnormal meteorological data vector and the abnormal hydrological data vector and extract features according to the time series difference of the data to obtain abnormal meteorological features and abnormal hydrological features, and obtain risk feature indicators in combination with forest vegetation density, and judge whether there is an abnormality according to the risk feature indicators; The management module is used to classify risk characteristics and to perform task assignment and status tracking.

2. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The meteorological information storage unit is used to store data from all monitoring stations, and to construct data priorities in combination with forest vegetation density and collection time for data sorting. The calculation formula for data priority is: Among them, P i j represents the priority of the ith meteorological data of the jth monitoring station, α1 represents the weight coefficient of forest vegetation, represents the number of forest vegetation pixels at the ,th monitoring station, represents the total number of pixels at the jth monitoring station, S represents the area represented by a pixel, α2 represents the weight coefficient of the sampling time, e represents the natural base, k represents the time attenuation coefficient, and t now Indicates the current time, t i Indicates the collection time of the i-th data; Determine storage location based on data priority and thresholds: Among them, P threshold represents the threshold, P i Indicates the data priority of the i-th data.

3. A meteorological and hydrological analysis and management cloud platform according to claim 2, characterized in that: Storage resources for meteorological and hydrological data are optimized, and the allocation ratio is dynamically adjusted according to the importance of the data: Among them, R represents the storage resource ratio of hot data to cold data, S h Indicates the storage resources currently used by the hot data unit, S t represents the total storage resources, P′ m represents the data priority of the mth hot data, M represents the number of hot data records, and P" n Indicates the data priority of the nth data, and N indicates the total number of data records.

4. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The active query unit also includes: The active query unit is used to receive keywords, time ranges and geographical ranges actively input by users and conduct queries, and prioritize and display the optimization results in combination with forest vegetation area and density; the functions of the active query unit also include: keyword retrieval and multi-condition cross-screening, and the multi-condition cross-screening includes supporting compound queries based on keywords, time ranges, monitoring station locations and forest vegetation coverage.

5. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The passive query unit also includes: The formula for retrieving the second meteorological information based on the association of the first meteorological information is: in, Indicates the second weather information, Indicates the first weather information, F weather represents meteorological characteristics, corr() represents the correlation function, and θ1 represents the first meteorological correlation threshold; The formula for interactively retrieving the third hydrological information based on the first meteorological information is: in, Indicates the third hydrological information, F hydro represents the hydrological characteristics, f() represents the interaction influence function, and θ2 represents the first interaction influence threshold; The formula for retrieving the second hydrological information based on the association of the first hydrological information is: in, represents the second hydrological information, represents the first hydrological information, θ3 represents the second hydrological correlation threshold; The formula for interactively retrieving the third meteorological information based on the first hydrological information is: in, represents the third weather information, and θ4 represents the second interaction influence threshold.

6. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The specific steps of constructing abnormal meteorological characteristics and abnormal hydrological characteristics include: Acquire abnormal meteorological data vectors and extract abnormal hot data and abnormal cold data to obtain abnormal meteorological hot data components and abnormal meteorological cold data components; acquire the abnormal meteorological hot data components collected in several time windows and use a sliding window to calculate the data change amount in adjacent time periods to obtain an abnormal meteorological hot data change matrix, extract features from the abnormal meteorological hot data change matrix using a trend analysis model to obtain abnormal meteorological hot data features, acquire the abnormal meteorological hot data features and the abnormal meteorological cold data components to obtain abnormal meteorological features; Acquire the abnormal hydrological data vector and extract the abnormal hot data and abnormal cold data to obtain the abnormal hydrological hot data component and the abnormal hydrological cold data component; acquire the abnormal hydrological hot data components collected in several time windows and use the sliding window to calculate the data change in adjacent time periods to obtain the abnormal hydrological hot data change matrix, extract features from the abnormal hydrological hot data change matrix using a trend analysis model to obtain the abnormal hydrological hot data features, acquire the abnormal hydrological hot data features and the abnormal hydrological cold data components to obtain the abnormal hydrological features.

7. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The calculation formula of risk characteristic index is: Wherein, Dan represents the risk characteristic index, ω1 represents the weight coefficient of abnormal meteorological characteristics, H represents the length of abnormal meteorological characteristics, x h The hth component representing abnormal meteorological characteristics, μ x represents the mean of normal meteorological data, σ x represents the standard deviation of normal meteorological data, ω2 represents the weight coefficient of abnormal hydrological characteristics, G represents the length of abnormal hydrological characteristics, y g The gth component representing abnormal hydrological characteristics, μ y represents the mean of normal hydrological data, σ y represents the standard deviation of normal hydrological data, ω3 represents the weight coefficient of forest vegetation density, ΔF density Represents the change in forest vegetation density.

8. A meteorological and hydrological analysis and management cloud platform according to claim 1, characterized in that: The steps of grading according to risk characteristic indicators, assigning tasks and tracking status also include: Obtain abnormal risk indicators and classify abnormal risk levels according to priority, and allocate detection time, detection personnel and detection tools according to abnormal risk levels and historical detection data.

Images