Agrometeorological disaster chained early warning method and system based on cloud computing

Abstract

The invention discloses an agricultural meteorological disaster chain type early warning method and system based on cloud computing, and the method comprises the steps: collecting meteorological data in real time through meteorological monitoring equipment, carrying out the preprocessing of the received meteorological data on a cloud computing platform, carrying out the disaster chain recognition and analysis of the processed meteorological data, and carrying out the early warning of the disaster chain. According to the method, a possible agro-meteorological disaster chain is identified, an agro-meteorological disaster chain type early warning model is constructed based on historical meteorological data and disaster chain identification results, disaster chain early warning analysis is carried out, and publishing is carried out through a network. The system comprises a data acquisition module, a data processing module, a disaster chain identification module, an early warning model module, an early warning analysis module and an information issuing module. Therefore, the accuracy and timeliness of agricultural meteorological disaster early warning can be remarkably improved, and decision support is provided for agricultural disaster prevention and reduction by collecting meteorological data in real time, analyzing by using the cloud computing platform, constructing an accurate early warning model and issuing early warning information in real time.

Description

Technical Field

[0001] The present invention relates to the field of meteorological information technology, and in particular to a cloud computing-based agricultural meteorological disaster chain early warning method and system thereof. Background Art

[0002] Agricultural production is closely related to meteorological conditions. Meteorological disasters seriously affect the sustainable development of agriculture. my country has a vast territory and agricultural meteorological disasters occur frequently. Every year, the area of ​​crops affected by meteorological disasters reaches hundreds of millions of acres, causing huge economic losses.

[0003] In terms of agricultural meteorological disaster warning, traditional methods are mainly based on threshold judgments of single meteorological elements, such as temperature and precipitation. When a certain meteorological element reaches the set threshold, an early warning is issued. Although this method can play an early warning role to a certain extent, it has obvious limitations. On the one hand, agricultural meteorological disasters are often the result of the combined effects of multiple meteorological factors. A single threshold judgment cannot fully reflect the real risk of disasters. On the other hand, there is a chain reaction in the process of meteorological disasters. For example, heavy rains may cause floods, which in turn lead to a series of secondary disasters such as loss of soil fertility and the breeding of pests and diseases. Traditional early warning methods find it difficult to effectively identify and warn of disaster chains.

[0004] The emergence of cloud computing technology has made it possible to solve these problems. Cloud computing has powerful computing and data storage capabilities, can quickly process massive meteorological data, and analyze them through complex algorithms. At the same time, with the development of Internet of Things technology, meteorological monitoring equipment can collect meteorological data more widely and accurately, providing a data basis for cloud computing-based agricultural meteorological disaster chain warning. Therefore, it is of great practical significance to study the cloud computing-based agricultural meteorological disaster chain warning method and system. Summary of the invention

[0005] The present invention aims to solve the technical problems in the above-mentioned technology at least to some extent.

[0006] To this end, the present invention discloses, on one hand, a cloud computing-based agricultural meteorological disaster chain early warning method, comprising the following steps: S1. Collecting meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas, and transmitting the meteorological data to a cloud computing platform; S2. Preprocessing the received meteorological data on the cloud computing platform, including data cleaning, format unification and data verification; S3. Using computing resources of the cloud computing platform, the processed meteorological data is subjected to disaster chain identification analysis, wherein the analysis is conducted through a comprehensive spatial and temporal analysis of the meteorological data, and a comparison and pattern matching with historical disaster data to identify possible agricultural meteorological disaster chains; S4. Train and build an agricultural meteorological disaster chain warning model through machine learning algorithms based on historical meteorological data and disaster chain identification results as input; S5. Input the real-time meteorological data into the early warning model. When the probability of the real-time meteorological data being the starting disaster is greater than the set threshold, a disaster chain early warning analysis is performed. The analysis uses the knowledge graph technology and is based on the real-time meteorological data as the starting disaster to construct a disaster chain. The graph search algorithm and the causal relationship between disasters are used to predict whether the disaster chain will occur. When the disaster chain is predicted to occur, early warning information is generated according to the early warning information template. S6. Release early warning information to agricultural practitioners and agricultural departments in real time through mobile communication networks or the Internet.

[0007] In addition, the cloud computing-based agricultural meteorological disaster chain early warning method disclosed in the present invention may also have the following additional technical features: In one embodiment of the present invention, the meteorological monitoring equipment includes but is not limited to a temperature sensor, a humidity sensor, a wind speed sensor, a rainfall sensor and an air pressure sensor.

[0008] In one embodiment of the present invention, the early warning model construction step also includes training and verifying the early warning model.

[0009] Another aspect of the present invention discloses a cloud computing-based agricultural meteorological disaster chain early warning system, comprising: A data collection module is used to collect meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas; A data processing module, used for pre-processing the received meteorological data; The disaster chain identification module uses the computing resources of the cloud computing platform to perform disaster chain identification analysis on the processed meteorological data; The early warning model module builds an agricultural meteorological disaster chain early warning model based on historical meteorological data and disaster chain identification results; The early warning analysis module inputs real-time meteorological data into the early warning model to conduct disaster chain early warning analysis and generate early warning information when a disaster chain is predicted to occur; The information release module is used to release early warning information to agricultural practitioners and agricultural departments in real time.

[0010] In addition, the cloud computing-based agricultural meteorological disaster chain early warning system disclosed in the present invention may also have the following additional technical features: In one embodiment of the present invention, a user interaction module is further included, which is used to receive a query request input by a user.

[0011] In one embodiment of the present invention, the data collection module further includes a data transmission unit for transmitting the collected meteorological data to the cloud computing platform via wired or wireless means.

[0012] In one embodiment of the present invention, the early warning model module further includes a model updating unit for regularly updating and optimizing the early warning model according to new meteorological data and disaster chain identification results.

[0013] The cloud computing-based agricultural meteorological disaster chain warning method and system disclosed in the present invention can significantly improve the accuracy and timeliness of agricultural meteorological disaster warnings. By collecting meteorological data in real time and analyzing it using a cloud computing platform, an accurate warning model can be constructed, and warning information can be released in real time to provide decision support for agricultural disaster prevention and mitigation.

[0014] Additional contents and advantages of the present invention will be given in the following description or will be understood through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The technical solutions and beneficial effects of the present invention will become apparent and easily understood from the following contents in conjunction with the accompanying drawings, wherein: Figure 1 It is a flow chart of the agricultural meteorological disaster chain early warning method based on cloud computing of the present invention; Figure 2 The present invention is a schematic diagram of the agricultural meteorological disaster chain early warning method based on cloud computing; Figure 3 This is an architecture diagram of the agricultural meteorological disaster chain early warning system based on cloud computing of the present invention. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0017] The agricultural meteorological disaster chain early warning method and system based on cloud computing disclosed in the present invention will be described below with reference to the accompanying drawings.

[0018] like Figure 1 , Figure 2 and Figure 3 As shown, a cloud computing-based agricultural meteorological disaster chain early warning method includes the following steps: S1. Collect meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas, and transmit the meteorological data to the cloud computing platform; Specifically, suppose the data set collected by the meteorological sensor network is , where each data point Contains multiple dimension information, such as timestamp , space coordinates , meteorological element values ( Indicates different meteorological elements, such as temperature, humidity, wind speed, etc.); S2. Preprocess the received meteorological data on the cloud computing platform, including data cleaning, format unification and data verification; Specifically, in the data cleaning phase, the cloud computing platform identifies and removes abnormal values ​​and erroneous values ​​in the data, such as correcting or deleting temperature or rainfall data that are obviously deviated from the normal range. In the process of format unification, data from different sensors and in different formats are converted into a standard format to facilitate subsequent data analysis and processing; Data verification is based on meteorological principles and data statistical laws, and conducts detailed verification of the rationality, consistency and integrity of the data. For example, based on the statistical characteristics of local historical meteorological data, it is determined whether the currently collected meteorological element values ​​are within a reasonable range. In addition, as a possibility, after completing the preprocessing of meteorological data, in order to ensure the effectiveness and accuracy of subsequent analysis, relevant staff can configure a dynamic feedback mechanism for the system, allowing continuous adjustment and improvement of data preprocessing strategies based on actual application effects, thereby ensuring that the data entering the disaster chain identification stage is always in the best state; S3. Using computing resources of the cloud computing platform, the processed meteorological data is subjected to disaster chain identification analysis, wherein the analysis is conducted through a comprehensive spatial and temporal analysis of the meteorological data, and a comparison and pattern matching with historical disaster data to identify possible agricultural meteorological disaster chains; Specifically, suppose the historical disaster dataset is , where each disaster record Included disaster types , occurrence time series , related meteorological element sequences ( is the meteorological element vector at the corresponding time point), geographic information wait; As a possibility, when analyzing the continuous rainfall data of a certain area, factors such as the intensity, duration, range of rainfall, and topography of the surrounding area will be considered at the same time. By comparing and matching with historical disaster data, it is determined whether the current meteorological conditions are likely to trigger a similar disaster chain. For example, combined with the water level change data of the surrounding area, it is determined whether flood disasters are likely to occur, and further analyzed whether flood disasters will lead to secondary disasters such as soil erosion and the breeding of crop diseases and pests, so as to determine the complete disaster chain; In addition, it should be noted that after using the computing resources of the cloud computing platform to perform disaster chain identification analysis on the processed meteorological data, the system needs to consider a variety of factors to enhance the accuracy and practicality of the analysis; Relevant staff can introduce environmental and socio-economic factors as auxiliary information, including but not limited to soil type, vegetation coverage, land use pattern, population density, etc. By combining these non-meteorological factors, the sensitivity of different regions to specific types of disasters and the possible impact range can be more accurately assessed; For example, in some areas with loose soil texture and low vegetation coverage, even if the precipitation does not reach the traditional flood standard, it is possible that local debris flow or landslide and other geological disasters may be caused due to severe soil erosion. In densely populated areas with a high degree of urbanization, special attention should be paid to the risk of heat island effect exacerbating extreme high temperature events. In addition, socioeconomic factors such as crop planting structure and the completeness of irrigation facilities will also affect the final disaster risk assessment results, and therefore are also included in the comprehensive considerations. S4. Train and build an agricultural meteorological disaster chain warning model through machine learning algorithms based on historical meteorological data and disaster chain identification results as input; Specifically, a disaster chain recognition model based on machine learning is constructed, such as a convolutional neural network or a recurrent neural network using deep learning, and the model input is the current preprocessed meteorological data sequence. and geographic information (obtained from the geographic information system), the output is the probability distribution of the starting disaster of the potential disaster chain ( is the total number of known disaster types); During training, the loss function can use cross entropy loss ,in, is the real disaster label (1 if it is the initial disaster type, otherwise 0), and the model parameters are continuously adjusted through the back propagation algorithm so that the model can accurately identify the initial disaster; As a possibility, relevant staff can use stochastic gradient descent, adaptive learning rate adjustment, etc. to improve the training efficiency and accuracy of the model; S5. Input the real-time meteorological data into the early warning model. When the probability of the real-time meteorological data being the starting disaster is greater than the set threshold, a disaster chain early warning analysis is performed. The analysis uses the knowledge graph technology and is based on the real-time meteorological data as the starting disaster to construct a disaster chain. The graph search algorithm and the causal relationship between disasters are used to predict whether the disaster chain will occur. When the disaster chain is predicted to occur, early warning information is generated according to the early warning information template. Specifically, once the initial hazard is identified (Probability greater than the set threshold ), using knowledge graph technology to build a disaster chain, assuming that the knowledge graph is , including the causal relationship between disasters ,in, For disaster type, Describe the associations and triggering conditions between them. Starting from the initial disaster, traverse the knowledge graph through graph search algorithms (such as depth-first search or breadth-first search) to build a disaster chain ,in, For secondary or derivative disasters, the time delay parameters between each disaster node are determined (Indicates that the To disaster average delay time); Assume that the meteorological disaster prediction model consists of a numerical weather forecast model and a statistical forecast model; Numerical weather forecast models are based on physical equations to describe atmospheric movement, water vapor transport and other processes, and input global meteorological reanalysis data. (such as global-scale data on atmospheric circulation, temperature, humidity, air pressure, etc.), topographic data (including altitude, slope, aspect, etc.), current weather data By solving a series of partial differential equations, the evolution of meteorological elements in the future is simulated. The simulated future meteorological element field is assumed to be ,in, For the future time, The calculation process is complex and involves a large amount of computing resources. It is implemented through parallel computing on the cloud computing platform. The statistical prediction model is based on historical similar meteorological processes and disaster cases. Assume that the historical data set is , for the disasters in the currently identified disaster chain , search for records of disasters occurring under similar meteorological conditions in historical data, and construct feature vectors (including meteorological element characteristics, time characteristics, geographical characteristics, etc.) and corresponding disaster intensity or probability labels , using regression analysis (such as linear regression, ridge regression, etc.) or time series analysis (such as Model) and other methods to train the model and obtain the prediction function , predict disasters The probability of occurrence and intensity estimates ; As a possibility, the linear regression method can be used to train the model and obtain the prediction function , specifically, It is a constructed feature vector, which includes meteorological element features, time features, geographical features, etc. are the eigenvalues ​​in the eigenvector, is the intercept term, are regression coefficients, which indicate the influence of each feature on the prediction result. By summing the products of each eigenvalue and the corresponding regression coefficient and adding the intercept term, we can get the predicted value of the disaster occurrence probability. The probability of occurrence ; Similarly, for the intensity estimate , based on historical data, using the constructed feature vector and the corresponding disaster intensity or probability label , based on the linear relationship ,in, is the regression coefficient parameter, which can be calculated by the least squares method. After obtaining the parameter value, the disasters in the current disaster chain are The corresponding eigenvector Substituting this formula, we can find the strength estimate ; In addition, as another possibility, time series analysis (such as Model) trains the model, specifically, In the model, is the autoregressive order, representing the number of lags of historical observations used, is the difference order, used to make the time series stationary, is the moving average order, reflecting the error term dependency structure; For disasters The time series data is first tested for stationarity. If it is not stationary, it is processed by difference. Suppose the sequence after stationarization is , establish the corresponding model ,in, is the autoregressive coefficient, is the moving average coefficient, is a white noise sequence; The model predicts The predicted value of The predicted value is converted into disaster The probability of occurrence and intensity estimates ; Taking flood disasters as an example, the probability of flood disasters is , intensity estimate The empirical formula is ,in, is the statistical analysis coefficient; The numerical weather forecast model and the statistical prediction model results are combined by weighted fusion. The weight of the numerical weather forecast model is set as , the weight of the statistical prediction model is , the probability of disaster occurrence after fusion is The intensity estimate is , and obtain the final meteorological disaster prediction result set ; It should be noted that is the probability of disaster occurrence predicted by the numerical weather forecast model, is the intensity estimate predicted by the numerical weather prediction model, and The data can be obtained from the website of the Meteorological Bureau through the standard API interface. The original data packet adopts the GRIB2 binary encoding format. is the probability of disaster occurrence predicted by the statistical prediction model, is the intensity estimate predicted by the statistical prediction model, and The specific value of is calculated from the previous step.

[0019] Assume that the agricultural planting distribution information dataset is , where each record Included crop varieties , planting area , Growth Stage , spatial location information wait; Agricultural impact assessment model based on meteorological disaster prediction results and agricultural planting information For each crop, an assessment is made , and its growth model is (Describe the growth and development patterns of crops at different growth stages over time, based on The disaster tolerance threshold function is: (Indicates crops Tolerance to disasters, such as lower temperature limit, upper humidity limit, etc.); When a disaster is predicted When a disaster occurs, , Occurrence time and crop growth stages , calculate the yield loss ratio For example, if the disaster is low temperature damage and occurs during the flowering period of crops, the relationship between yield loss and low temperature intensity and duration can be obtained through experimental data or empirical formulas. ,in, is the loss coefficient determined according to the variety, The duration of the disaster, and then the total loss of crop yield in the whole region is estimated to be At the same time, the impact of disasters on agricultural operations is assessed, such as based on soil moisture models Considering factors such as precipitation, evaporation, and irrigation, we predict the change of soil moisture content, determine the irrigation timing and make adjustment suggestions, and (Combined with the relationship between meteorological conditions and the breeding of pests and diseases), provide suggestions on advancing or postponing pest and disease control, etc.; S6. Release early warning information to agricultural practitioners and agricultural departments in real time through mobile communication networks or the Internet; Specifically, the warning information template is , which includes the disaster chain details (Describe the disaster chain in text form: initial disaster - secondary disaster 1 - secondary disaster 2...), expected time of occurrence (specific time interval or probability distribution), impact range (accurate to geographical area, such as township, village list), affected crop varieties and area (list varieties and corresponding areas), range of yield loss (estimated minimum and maximum values), suggestions for agricultural operations (List the response measures in separate items, such as irrigation water volume, fertilizer types, etc.) and other fields; Fill in the warning information template to generate specific warning information based on the results of the agricultural impact assessment , let the warning release channel set be Such as SMS platform, mobile APP, electronic display screen, rural radio, etc.), each channel corresponds to a publishing function , used to push warning information to the target user group, and set a feedback confirmation function , used to receive user feedback and count feedback. If a user does not provide feedback confirmation, an alert will be issued again according to the preset strategy (such as a second SMS reminder, increasing the frequency of APP push notifications, etc.) to ensure that the information is effectively delivered; It should be noted that meteorological monitoring equipment includes but is not limited to temperature sensors, humidity sensors, wind speed sensors, rainfall sensors and air pressure sensors, and the early warning model construction step also includes training and verifying the early warning model.

[0020] A cloud computing-based agricultural meteorological disaster chain early warning system 100, comprising: The data collection module 101 is used to collect meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas; Specifically, the meteorological monitoring equipment includes but is not limited to temperature sensors, humidity sensors, wind speed sensors, rainfall sensors and air pressure sensors, etc. These sensors are distributed in various key locations in the agricultural area to form a comprehensive meteorological monitoring network. The data acquisition module also includes a data transmission unit for transmitting the collected meteorological data to the cloud computing platform via wired or wireless means; The data processing module 102 is used to pre-process the received meteorological data; Specifically, the received meteorological data are preprocessed, including data cleaning, format unification, and data verification; In addition, through a strict data verification mechanism, based on meteorological principles and data statistical laws, the rationality, consistency and integrity of the data are carefully verified; The relevant staff can also integrate an intelligent diagnostic unit into the data processing module 102, which can automatically identify and mark potential data quality issues, such as sensor failures or abnormal readings. For the marked data points, the system will initiate a secondary verification process to determine its validity by comparing the data of adjacent sensors or other historical records; The disaster chain identification module 103 uses the computing resources of the cloud computing platform to perform disaster chain identification analysis on the processed meteorological data; Specifically, for example, by analyzing the continuous rainfall data of a certain area and combining it with the water level change data of the surrounding areas, we can determine whether flood disasters are likely to occur, and further analyze whether flood disasters will lead to secondary disasters such as soil erosion and the breeding of crop pests and diseases, thereby determining the complete disaster chain; The early warning model module 104 builds an agricultural meteorological disaster chain early warning model based on historical meteorological data and disaster chain identification results; Specifically, machine learning algorithms, such as deep learning convolutional neural networks or recurrent neural networks, can be used here to take the current preprocessed meteorological data sequence and geographic information as model inputs and output the probability distribution of the starting disaster of the potential disaster chain; In addition, as a possibility, relevant staff can regularly retrain and update the model to adapt to changing meteorological conditions and disaster patterns; The early warning analysis module 105 inputs the real-time meteorological data into the early warning model to perform disaster chain early warning analysis, and generates early warning information when a disaster chain is predicted to occur; Specifically, when a disaster chain is predicted to occur, detailed warning information is generated. The warning analysis module comprehensively considers the results of the numerical weather forecast model and the statistical forecast model, and uses a weighted fusion method to improve the accuracy of the forecast; Numerical weather forecast models simulate the evolution of future meteorological elements based on physical equations, while statistical forecast models make predictions based on similar historical meteorological processes and disaster cases. The two complement each other. As a possibility, relevant staff can reasonably adjust the weights of the two models according to the climate characteristics and historical disasters in different regions to improve the accuracy of prediction. In monsoon climate areas, for example, statistical prediction models may be given higher weights in certain seasons because monsoon climate has certain periodicity and regularity; The information release module 106 is used to release the early warning information to agricultural practitioners and agricultural departments in real time; Specifically, the information release module fills in specific warning content based on the preset warning information template, including disaster chain details, expected occurrence time, impact range, affected crop varieties and areas, yield loss range, and suggestions for agricultural response operations; The warning release channels are diversified, covering SMS platforms, mobile apps, electronic display screens, rural broadcasts, etc. Each channel corresponds to a release function to ensure that the warning information can be pushed to the target user group in a timely and accurate manner; At the same time, the information release module also has a feedback confirmation function, which is used to receive user feedback and count the feedback. If a user does not provide feedback confirmation, an early warning will be issued again according to the preset strategy, such as a second SMS reminder, increasing the frequency of APP push notifications, etc., to ensure that the information is effectively delivered.

[0021] It should be noted that it also includes a user interaction module for receiving query requests input by users, the data collection module also includes a data transmission unit for transmitting the collected meteorological data to the cloud computing platform via wired or wireless means, and the early warning model module also includes a model updating unit for regularly updating and optimizing the early warning model according to new meteorological data and disaster chain identification results; It should also be noted that agricultural practitioners or agricultural departments can query historical meteorological data, disaster warning records, agricultural meteorological disaster-related knowledge and other information through the user interaction module; In addition, relevant staff can add the following functions: when farmers find that the actual weather conditions in the field are different from the warning information, they can provide timely feedback to the system through the user interaction module. The system will adjust and optimize the model based on this feedback information to improve the accuracy of the warning; At the same time, users can also provide feedback on actual weather conditions and disaster impacts, providing a reference for system optimization and improvement.

[0022] In summary, the cloud computing-based agricultural meteorological disaster chain warning method and system disclosed in the present invention can significantly improve the accuracy and timeliness of agricultural meteorological disaster warnings. By collecting meteorological data in real time and analyzing it using a cloud computing platform, an accurate warning model can be constructed, and warning information can be released in real time to provide decision support for agricultural disaster prevention and mitigation.

[0023] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0024] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0025] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0026] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0027] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0028] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A cloud computing-based agricultural meteorological disaster chain early warning method, characterized in that: The following steps are involved: S1. Collecting meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas, and transmitting the meteorological data to a cloud computing platform; S2. Preprocessing the received meteorological data on the cloud computing platform, including data cleaning, format unification and data verification; S3. Using computing resources of the cloud computing platform, the processed meteorological data is subjected to disaster chain identification analysis, wherein the analysis is conducted through a comprehensive spatial and temporal analysis of the meteorological data, and a comparison and pattern matching with historical disaster data to identify possible agricultural meteorological disaster chains; S4. Train and build an agricultural meteorological disaster chain warning model through machine learning algorithms based on historical meteorological data and disaster chain identification results as input; S5. Input the real-time meteorological data into the early warning model. When the probability of the real-time meteorological data being the starting disaster is greater than the set threshold, a disaster chain early warning analysis is performed. The analysis uses the knowledge graph technology and is based on the real-time meteorological data as the starting disaster to construct a disaster chain. The graph search algorithm and the causal relationship between disasters are used to predict whether the disaster chain will occur. When the disaster chain is predicted to occur, early warning information is generated according to the early warning information template. S6. Release early warning information to agricultural practitioners and agricultural departments in real time through mobile communication networks or the Internet.

2. The cloud computing-based agricultural meteorological disaster chain early warning method according to claim 1, characterized in that: The meteorological monitoring equipment includes but is not limited to a temperature sensor, a humidity sensor, a wind speed sensor, a rainfall sensor and an air pressure sensor.

3. The agricultural meteorological disaster chain early warning method based on cloud computing according to claim 1, characterized in that: The early warning model construction step also includes training and verifying the early warning model.

4. A cloud computing-based agricultural meteorological disaster chain warning system, characterized in that: include: A data collection module is used to collect meteorological data in real time through meteorological monitoring equipment distributed in agricultural areas; A data processing module, used for pre-processing the received meteorological data; The disaster chain identification module uses the computing resources of the cloud computing platform to perform disaster chain identification analysis on the processed meteorological data; The early warning model module builds an agricultural meteorological disaster chain early warning model based on historical meteorological data and disaster chain identification results; The early warning analysis module inputs real-time meteorological data into the early warning model to conduct disaster chain early warning analysis and generate early warning information when a disaster chain is predicted to occur; The information release module is used to release early warning information to agricultural practitioners and agricultural departments in real time.

5. The cloud computing-based agricultural meteorological disaster chain early warning system according to claim 4, characterized in that: It also includes a user interaction module for receiving a query request input by a user.

6. The cloud computing-based agricultural meteorological disaster chain early warning system according to claim 4, characterized in that: The data collection module also includes a data transmission unit for transmitting the collected meteorological data to the cloud computing platform via wired or wireless means.

7. The cloud computing-based agricultural meteorological disaster chain early warning system according to claim 4, characterized in that: The early warning model module also includes a model updating unit, which is used to regularly update and optimize the early warning model according to new meteorological data and disaster chain identification results.

Images