Gale disaster weather index prediction system based on long short-term memory network model

Abstract

The application provides a gale disaster weather index prediction system based on a long short-term memory network model, relates to the technical field of weather monitoring, and specifically comprises the following steps: collecting historical meteorological data of an area to be evaluated and a disaster area and duration of a gale disaster event occurring at a corresponding time; establishing an index prediction model based on a long short-term memory network, inputting the historical meteorological data into the model as a training set, inputting the disaster area and duration of the corresponding gale disaster as labels, and training the model; collecting meteorological data of the area to be evaluated in real time, obtaining a predicted disaster area and a predicted duration; collecting terrain height and slope data to correct the predicted disaster area and the predicted duration, comparing the corrected predicted disaster area and the predicted duration with preset threshold values, and dividing a risk level according to a result. The application of the long short-term memory network model can learn deep correlations between factors such as wind speed, humidity and temperature and the gale disaster event from historical meteorological data, and the accuracy of prediction is improved.

Description

[0001] This application is a divisional application of application No. 202410923710.5, filed on July 10, 2024, with the original title of "A gale disaster weather index prediction system and method based on deep learning". TECHNICAL FIELD

[0002] The present application relates to the technical field of weather monitoring, in particular to a gale disaster weather index prediction system based on a long short-term memory network model. BACKGROUND

[0003] Gale disasters pose a serious threat to human society and the economy, so it is crucial to accurately predict the impact of gale disasters. Traditional weather forecasting methods mainly rely on numerical weather prediction models and empirical formulas, which cannot fully capture the complex effects of meteorological factors and terrain on gale disasters. In recent years, deep learning technology, especially long short-term memory networks, has shown excellent performance in handling time series data, effectively capturing the temporal dependence and nonlinear characteristics of meteorological data. Therefore, applying LSTM to gale disaster weather index prediction can improve the accuracy of the prediction.

[0004] With climate change, the frequency and intensity of extreme weather events have increased, placing higher demands on accurate predictions of gale disasters. However, traditional prediction methods are limited in their ability to predict multivariate, multiscale, and nonlinear meteorological data. In particular, in areas with complex terrain, wind distribution is significantly affected by factors such as terrain height and slope, and traditional models often struggle to accurately reflect these effects, resulting in significant deviations between predicted results and actual disaster conditions. By using a gridding method to collect terrain height and slope data in the area to be evaluated, and by modifying the terrain data to fully consider the regulatory effects of terrain on gale disasters, the prediction results are more meaningful. In addition, with the development of sensor technology and data collection technology, more abundant and detailed meteorological data can be obtained, but how to effectively utilize these data to improve the performance of the prediction model is a major challenge facing the field of meteorological prediction.

[0005] The above information disclosed in the BACKGROUND section only serves to enhance the understanding of the background of the present disclosure, and thus it can include information that does not constitute the prior art known to those of ordinary skill in the art. SUMMARY

[0006] The purpose of the present application is to provide a gale disaster weather index prediction system based on a long short-term memory network model to solve the problems raised in the background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A gale disaster weather index prediction system based on a long short-term memory network model, specifically comprising:

[0009] A data acquisition module is configured to acquire the disaster area and duration of a gale disaster event that occurred in the past in an evaluation area, and acquire historical weather data of the gale disaster event that occurred in the past in the evaluation area one week ago, wherein the weather data includes wind speed, humidity, temperature, air pressure, precipitation and cloud cover.

[0010] A model construction module is configured to establish an index prediction model based on a long short-term memory network, input the historical weather data as a training set into the index prediction model, input the disaster area and duration of the corresponding gale disaster as labels, train the model, and obtain a trained index prediction model.

[0011] A prediction module is configured to acquire weather data of the evaluation area in the past one week, input the weather data into the trained index prediction model, obtain a predicted disaster area and a predicted duration of a gale disaster weather, and acquire terrain height and slope data of the evaluation area by using a gridding method.

[0012] A comprehensive evaluation module is configured to correct the predicted disaster area and the predicted duration of the gale disaster weather by using the acquired terrain height and slope data of the evaluation area, obtain a comprehensive weather index, compare the comprehensive weather index with a preset evaluation threshold, and divide a risk level according to a comparison result.

[0013] Further, the specific logic for acquiring the wind speed, humidity, temperature and air pressure of the evaluation area one week before a gale disaster event that occurred in the past is as follows:

[0014] The wind speed, humidity, temperature and air pressure of the evaluation area one week before a gale disaster event that occurred in the past are obtained through weather data recorded by a meteorological bureau;

[0015] n measurement points around the evaluation area are selected, covering the four directions of east, west, south and north, and the measurement points are selected according to equal amount and equal distance in each direction;

[0016] The specific logic for acquiring weather data of the evaluation area in the past one week is as follows:

[0017] Anemometers, hygrometers, thermometers and barometers are installed at the measurement points;

[0018] The wind speed data, humidity data, temperature data and air pressure data are recorded at different measurement points to obtain the average wind speed, average humidity, average temperature and average air pressure of the evaluation area;

[0019] The formula for obtaining the average wind speed is as follows:

[0020]

[0021] wherein, is the average wind speed, v i represents the wind speed measured at the i-th measuring point;

[0022] The formula for obtaining the average humidity is:

[0023]

[0024] wherein, is the average humidity, RH i represents the humidity measured at the i-th measuring point;

[0025] The formula for obtaining the average temperature is:

[0026]

[0027] wherein, is the average temperature, Temp i represents the temperature measured at the i-th measuring point;

[0028] The formula for obtaining the average air pressure is:

[0029]

[0030] wherein, is the average air pressure, PRES i represents the air pressure measured at the i-th measuring point.

[0031] Further, the specific logic for collecting the terrain height H and slope data C of the area to be evaluated using the gridding method is:

[0032] The area to be evaluated is divided into regular small grids according to the size of the area, and then the average height and slope are calculated in each grid unit to generate a continuous terrain surface model, each unit represents an average value, and the height and slope of all grid units are again weighted and averaged, and the final value is obtained as the terrain height H and slope C of the area to be evaluated.

[0033] Further, the specific logic for collecting the precipitation M and cloud cover N of the area to be evaluated one week before the historical gale disaster event is:

[0034] The daily average precipitation M and daily average cloud cover N one week before the occurrence of the gale disaster in the area to be evaluated are obtained from the historical weather data recorded by the meteorological bureau;

[0035] The collected cloud cover N is the proportion of the sky covered by clouds in the disaster area when the gale disaster occurs, expressed in percentage, ranging from 0% to 100%;

[0036] The specific logic behind collecting precipitation and cloud cover data for the area to be evaluated over the past week is as follows:

[0037] The precipitation and cloud cover of the area to be evaluated over the past week were collected using historical weather data recorded by the meteorological bureau and forecast results.

[0038] Furthermore, the process of establishing an exponential prediction model based on a long short-term memory network specifically includes:

[0039] Historical meteorological data are organized into multiple time series data according to type and used as training set. The affected area and duration of the strong wind disaster are used as labels to construct a deep learning model based on long short-term memory network. The training set and labels are input into the deep learning model to train the model and obtain a trained index prediction model. The historical meteorological data includes wind speed, humidity, temperature, air pressure, precipitation and cloud cover of the area to be evaluated.

[0040] Furthermore, the specific logic underlying the acquisition of the predicted affected area S and predicted duration T of severe wind disasters is as follows:

[0041] Meteorological data for the past week in the area to be evaluated is collected, including wind speed, humidity, temperature, air pressure, precipitation and cloud cover. This data is then input into a trained index prediction model, which outputs the predicted affected area S and the predicted duration T of the severe wind disaster.

[0042] Furthermore, by using the collected topographic elevation and slope data of the area to be assessed, the predicted affected area and predicted duration of strong winds are corrected, resulting in the following formula for the comprehensive weather index:

[0043]

[0044] Where QS is the comprehensive weather index, H is the terrain elevation of the area to be assessed, α is its preset scaling factor, C is the terrain slope of the area to be assessed, β is its preset scaling factor, S is the predicted disaster area, T is the predicted duration, and ρ is (S 2 +T 3 The preset proportional coefficients are C1, which is a constant correction exponent, and α, β, and ρ are all greater than zero.

[0045] Furthermore, the process of comparing the composite weather index (QS) with preset assessment thresholds and classifying risk levels based on the comparison results specifically includes: setting a series of thresholds corresponding to different levels of gale hazard; comparing the obtained composite weather index with the preset thresholds; and classifying the hazard levels based on the comparison results.

[0046] When QS<30, it is predicted to be low risk, the public can normally carry out outdoor activities, but need to pay attention to weather changes;

[0047] When 30≤QS<60, it is predicted to be medium risk, it is suggested that the public reduce unnecessary outdoor activities, and pay attention to fixed loose articles to prevent being blown away by the wind;

[0048] When 60≤QS<80, it is predicted to be high risk, it is suggested that the public avoid outdoor activities, it is suggested that the relevant departments strengthen the patrol to ensure the stability of public safety facilities, and the residents should ensure that the doors and windows are closed and the outdoor articles are fixed;

[0049] When QS≥80, it is predicted to be extremely high risk, it is suggested that the public stay indoors and avoid all outdoor activities, and the relevant departments should start emergency response measures, such as evacuating residents in high-risk areas and closing public facilities that may be affected.

[0050] Among them, QS is a comprehensive weather index.

[0051] The application further provides a gale disaster weather index prediction method based on deep learning, which is executed by the gale disaster weather index prediction system.

[0052] Step 1: Collect the disaster area and duration of the gale disaster event occurred in the history of the to-be-evaluated area, and collect the historical meteorological data one week before the gale disaster event occurred in the history of the to-be-evaluated area, the meteorological data including wind speed, humidity, temperature, air pressure, precipitation and cloud amount;

[0053] Step 2: Establish an index prediction model based on a long short-term memory network, input the historical meteorological data into the index prediction model as a training set, input the disaster area and duration of the corresponding gale disaster as a label, train the model, and obtain the trained index prediction model;

[0054] Step 3: Collect the meteorological data of the to-be-evaluated area in the recent one week, input the meteorological data into the trained index prediction model, obtain the predicted disaster area and predicted duration of the gale disaster weather, and collect the terrain height and slope data of the to-be-evaluated area by using a gridding method;

[0055] Step 4: Correct the predicted disaster area and predicted duration of the gale disaster weather by using the collected terrain height and slope data of the to-be-evaluated area, obtain a comprehensive weather index, compare the comprehensive weather index with a preset evaluation threshold, and divide the risk level according to the comparison result.

[0056] Compared with the prior art, the application has the beneficial effects that:

[0057] The application effectively solves the limitations of traditional prediction methods in dealing with complex nonlinear relationships and time series data by introducing a long short-term memory network model based on deep learning. The LSTM model can learn the deep relationship between wind speed, humidity, temperature and other factors and gale disaster events from historical meteorological data, improving the accuracy of the prediction. At the same time, terrain factors have a significant impact on wind speed and direction, which can change the path and impact range of gale, so they are key variables for accurate prediction. The terrain height and slope data of the area to be evaluated are collected by using the rasterization method, and the adjustment effect of terrain on gale disaster is fully considered through the correction of terrain data, making the prediction results more practical and closer to the actual disaster situation. This prediction method that comprehensively considers meteorological factors and terrain characteristics not only improves the accuracy and reliability of gale disaster prediction, but also provides more scientific and effective decision support for disaster prevention and emergency management, significantly improving the efficiency and effectiveness of disaster response. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 It is a schematic diagram of the overall system module of the application;

[0059] Figure 2 It is a schematic diagram of the overall method flow of the application. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical scheme and advantages of the application clearer, the application will be further described in detail below with specific examples.

[0061] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the application should be understood as the usual meaning understood by those skilled in the art to which the application belongs. The "first", "second" and similar words used in the application do not represent any order, quantity or importance, but are used to distinguish different components. "Include" or "contain" and similar words mean that the elements or objects before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connected" or "connected" and similar words are not limited to physical or mechanical connection, but can include electrical connection, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to represent the relative positional relationship, when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0062] Embodiment:

[0063] Please refer to Figure 1 The application provides a gale disaster weather index prediction system based on deep learning, which comprises:

[0064] The data collection module is configured to collect the disaster area and duration of a gale disaster event occurred in the past in the to-be-evaluated region, and collect historical weather data of the to-be-evaluated region one week before the gale disaster event, wherein the weather data includes wind speed, humidity, temperature, air pressure, precipitation and cloud cover.

[0065] In this embodiment, the specific logic for collecting the wind speed, humidity, temperature and air pressure of the to-be-evaluated region one week before the gale disaster event is as follows:

[0066] The wind speed, humidity, temperature and air pressure of the to-be-evaluated region one week before the gale disaster event are obtained through weather data recorded by the meteorological bureau.

[0067] Select n measuring points around the to-be-evaluated region, covering the four directions of east, west, south and north, and select the measuring points according to equal amount and equal distance in each direction.

[0068] The specific logic for collecting the weather data of the to-be-evaluated region in the past week is as follows:

[0069] Install an anemometer, a hygrometer, a thermometer and a barometer at the measuring point.

[0070] Record the wind speed data, humidity data, temperature data and air pressure data at different measuring points to obtain the average wind speed, average humidity, average temperature and average air pressure of the to-be-evaluated region.

[0071] The formula for obtaining the average wind speed is as follows:

[0072]

[0073] wherein, is the average wind speed, v i represents the wind speed measured at the i-th measuring point.

[0074] The formula for obtaining the average humidity is as follows:

[0075]

[0076] wherein, is the average humidity, RH i represents the humidity measured at the i-th measuring point.

[0077] The formula for obtaining the average temperature is as follows:

[0078]

[0079] wherein, is the average temperature, Temp i represents the temperature measured at the i-th measuring point.

[0080] The formula for obtaining the average air pressure is:

[0081]

[0082] where, is the average air pressure, PRES i represents the air pressure measured at the i-th measurement point;

[0083] The specific logic for collecting terrain height and slope data of the area to be evaluated using the gridding method is:

[0084] According to the size of the area, the area to be evaluated is divided into regular small grids, and then the average height and slope are calculated in each grid cell to generate a continuous terrain surface model, each cell represents an average value, and the height and slope of all grid cells are weighted again to obtain the final value as the terrain height H and slope C of the area to be evaluated.

[0085] The specific logic for collecting the precipitation M and cloud cover N of the area to be evaluated one week before the historical gale disaster event is:

[0086] The daily average precipitation M and daily average cloud cover N one week before the occurrence of the gale disaster event in the area to be evaluated are obtained through the historical weather data recorded by the meteorological bureau;

[0087] The collected cloud cover N is the proportion of the sky covered by clouds in the disaster area when the gale disaster occurs, expressed in percentage, ranging from 0% to 100%;

[0088] The specific logic for collecting the precipitation and cloud cover data of the area to be evaluated in the last week is:

[0089] The precipitation and cloud cover of the area to be evaluated in the last week are collected through the historical weather data recorded by the meteorological bureau and the prediction results of weather forecasts;

[0090] The specific method for collecting the disaster area S is:

[0091] Satellite images are used to identify and measure the disaster area, and image processing software is used to analyze the images, identify the disaster area and calculate its area;

[0092] The specific method for collecting the disaster duration T is:

[0093] Remote sensing technology is used to monitor the development process of the disaster event, and by analyzing continuous remote sensing images, the duration T of the disaster event can be determined;

[0094] Historical weather data is a key input for training prediction models, by analyzing patterns and trends in historical data, the model can learn the rules of weather changes, thereby improving the accuracy of predicting future weather.

[0095] a model construction module, which establishes an index prediction model based on a long short-term memory network, inputs historical meteorological data as a training set into the index prediction model, takes the affected area and duration of a corresponding gale disaster as labels, trains the model, and obtains a trained index prediction model;

[0096] The process of establishing the index prediction model based on the long short-term memory network specifically includes:

[0097] The historical meteorological data is formed into multiple groups of time series data according to types and taken as a training set, the affected area and duration of a gale disaster that occurs are taken as labels, a deep learning model based on a long short-term memory network is constructed, the training set and the labels are input into the deep learning model, the model is trained, and a trained index prediction model is obtained, and the historical meteorological data includes wind speed, humidity, temperature, air pressure, precipitation, and cloud cover of an area to be evaluated;

[0098] LSTM is a special recurrent neural network (RNN) that can effectively handle and predict long-term dependencies in time series. In meteorological data, weather patterns often have complex temporal dependencies, and LSTM can capture these patterns to improve prediction accuracy; the LSTM model can directly map from the original input data (wind speed, temperature, humidity, etc.) to the output labels (affected area and duration), realizing end-to-end learning. This direct mapping simplifies the model training process and may improve prediction accuracy.

[0099] a prediction module, which collects meteorological data of an area to be evaluated in the past week, inputs the data into the trained index prediction model, obtains a predicted affected area and a predicted duration of a gale disaster weather, and collects terrain height and slope data of the area to be evaluated using a gridding method;

[0100] In this embodiment, the specific logic for obtaining the predicted affected area S and the predicted duration T of the gale disaster weather is as follows:

[0101] The prediction module collects meteorological data of an area to be evaluated in the past week, including wind speed, humidity, temperature, air pressure, precipitation, and cloud cover of the area to be evaluated, inputs the data into the trained index prediction model, and outputs a predicted affected area S and a predicted duration T of a gale disaster weather;

[0102] Real-time acquisition of meteorological data can ensure that the model inputs the latest information, which helps to improve the real-time and accuracy of the prediction. Compared with techniques that rely on historical data, real-time data can better reflect the current weather conditions. By combining terrain height and slope data, the impact of gale disasters on different terrain areas can be more comprehensively evaluated. This multi-dimensional data fusion helps to improve the predictive ability of the model, as it takes into account the impact of terrain on wind speed and direction, which can help the model better adapt to prediction needs in different geographical environments, thereby improving the generalization ability of the model.

[0103] The comprehensive evaluation module is used to correct the predicted disaster area and the predicted duration of the gale disaster weather using the collected terrain height and slope data of the area to be evaluated, and obtain a comprehensive weather index. The comprehensive weather index is compared with a preset evaluation threshold, and a risk level is divided according to the comparison result.

[0104] In this embodiment, the predicted disaster area and the predicted duration of the gale weather are corrected using the collected terrain height and slope data of the area to be evaluated, and the formula for obtaining the comprehensive weather index is as follows:

[0105]

[0106] wherein QS is the comprehensive weather index, H is the terrain height of the area to be evaluated, a is a preset proportion coefficient thereof, C is the terrain slope of the area to be evaluated, β is a preset proportion coefficient thereof, S is the predicted disaster area, T is the predicted duration, ρ is a preset proportion coefficient of (S 2 +T 3 ), C1 is a constant correction index, and a, β, ρ are all greater than zero. When the terrain height H of the area to be evaluated increases, the comprehensive weather index QS increases, because it will cause a change in wind direction, making the distribution of wind in different terrain areas uneven. This change in wind direction may cause some areas to be subjected to stronger wind forces, thereby increasing the risk of disaster. When the terrain slope C increases, the comprehensive weather index QS increases. Steep slopes can cause stronger local wind effects, such as slope wind or downwind, which can produce extreme wind speeds in specific areas, increasing the risk of disaster. When the predicted disaster area S increases, QS increases. When the predicted duration T increases, QS also increases accordingly. That is, H, C, S, T and the comprehensive weather index QS are positively correlated.

[0107] The process of comparing the comprehensive weather index QS with the preset evaluation threshold and dividing the risk level according to the comparison result specifically includes: setting a series of threshold values corresponding to different gale disaster risk levels, comparing the obtained comprehensive weather index with the preset threshold values, and dividing the risk level according to the comparison result.

[0108] When QS < 30, it is predicted to be low risk, and the public can normally carry out outdoor activities, but need to pay attention to weather changes;

[0109] When 30 ≤ QS < 60, it is predicted to be medium risk, and it is suggested that the public reduce unnecessary outdoor activities and pay attention to fixing loose objects to prevent them from being blown away by the wind;

[0110] When 60 ≤ QS < 80, it is predicted to be high risk, and it is suggested that the public avoid outdoor activities, and relevant departments should strengthen patrol to ensure the stability of public safety facilities, and residents should ensure that doors and windows are closed and outdoor objects are fixed;

[0111] When QS ≥ 80, it is predicted to be extremely high risk, and it is suggested that the public stay indoors and avoid all outdoor activities, and relevant departments should start emergency response measures such as evacuating residents in high-risk areas and closing public facilities that may be affected.

[0112] Among them, QS is a comprehensive weather index.

[0113] Please refer to Figure 2 , the application further provides a gale disaster weather index prediction method based on deep learning, which is executed by the gale disaster weather index prediction system described above, and the specific steps include:

[0114] Step 1: Collect the disaster area and duration of the gale disaster event that occurred in the past in the evaluation area, and collect the historical meteorological data one week before the gale disaster event occurred in the evaluation area, including wind speed, humidity, temperature, air pressure, precipitation and cloud cover;

[0115] Step 2: Establish an index prediction model based on a long short-term memory network, input the historical meteorological data as a training set into the index prediction model, and input the corresponding disaster area and duration of the gale disaster as a label, train the model, and obtain a trained index prediction model;

[0116] Step 3: Collect the meteorological data of the evaluation area in the past week, input it into the trained index prediction model, obtain the predicted disaster area and duration of the gale disaster weather, and collect the terrain height and slope data of the evaluation area using the gridding method;

[0117] Step 4: Correct the predicted disaster area and duration of the gale disaster weather using the collected terrain height and slope data of the evaluation area, obtain a comprehensive weather index, compare the comprehensive weather index with a preset evaluation threshold, and divide the risk level according to the comparison result.

[0118] The specific values of a, β and p in the formula are generally determined by a person skilled in the art according to actual conditions, a plurality of sets of sample data are collected by a person skilled in the art, and a corresponding preset proportion coefficient is set for each set of sample data, the set preset proportion coefficient and the collected sample data are substituted into the formula, the accuracy of the model output and the rationality of the result are observed through repeated experiments and parameter adjustment, the factor coefficients are gradually adjusted, the performance and effect of the model under different parameter settings are compared, the optimal combination of the coefficients is found, the calculated factor coefficients are screened and averaged to obtain the values of a, β and p.

[0119] In addition, the size of the preset factor coefficient is a specific value obtained by quantizing each parameter, which is for subsequent comparison. The size of the coefficient depends on the number of sample data and the corresponding preset proportion coefficient initially set by a person skilled in the art for each set of sample data, and is not unique, as long as the proportion relationship between the parameter and the quantized value is not affected.

[0120] The above formulas are dimensionless values calculated, the formula is obtained by collecting a large amount of data to simulate the most real situation, and the preset parameters in the formula are set by a person skilled in the art according to actual conditions.

[0121] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product. A person skilled in the art can realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized by hardware or software methods depends on the specific application and design constraints of the technical solutions.

[0122] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, which can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0123] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A gale disaster weather index prediction system based on a long short-term memory network model, characterized in that, Specifically comprising: The data acquisition module, the model construction module, the prediction module and the comprehensive evaluation module; The gale disaster weather index prediction system executes the following gale disaster weather index prediction method, comprising: Step 1: Collect the disaster area and duration of the gale disaster event occurred in the history of the to-be-evaluated region, collect the historical meteorological data of the to-be-evaluated region one week before the gale disaster event occurred, the meteorological data including wind speed, humidity, temperature, air pressure, precipitation and cloud cover; Step 2: Based on the long short-term memory network, an index prediction model is established, the historical meteorological data is input into the index prediction model as a training set, the disaster area and duration of the corresponding gale disaster are taken as labels, the model is trained, and a trained index prediction model is obtained; Step 3: Collect the meteorological data of the to-be-evaluated region in the recent one week, input into the trained index prediction model, obtain the predicted disaster area and duration of the gale disaster weather, and collect the terrain height and slope data of the to-be-evaluated region by using the gridding method; The specific logic of collecting the terrain height and slope data of the to-be-evaluated region by using the gridding method is: According to the area size, the to-be-evaluated region is divided into regular small grids, then the average height and slope in each grid unit are calculated, a continuous terrain surface model is generated, each unit represents an average value, and the height and slope of all grid units are weighted and averaged again, and the finally obtained value is taken as the terrain height and slope of the to-be-evaluated region; Step 4: The predicted disaster area and duration of the gale disaster weather are corrected by using the collected terrain height and slope data of the to-be-evaluated region, a comprehensive weather index is obtained, the comprehensive weather index is compared with a preset evaluation threshold, and according to the comparison result, a risk level is divided; The specific method for collecting the disaster area is: satellite images are used to identify and measure the disaster area, image processing software is used to analyze the images, the disaster area is identified and the area thereof is calculated; The specific method for collecting the disaster duration is: remote sensing technology is used to monitor the development process of the disaster event, and the duration of the disaster event is determined by analyzing continuous remote sensing images.

2. The long short-term memory network model-based gale disaster weather index prediction system according to claim 1, characterized in that: The specific logic for collecting the wind speed, humidity, temperature and air pressure of the to-be-evaluated region one week before the gale disaster event occurred in the history is: The wind speed, humidity, temperature and air pressure of the to-be-evaluated region one week before the gale disaster event occurred in the history are obtained through the weather data recorded by the meteorological bureau; Selecting measurement points around the area to be evaluated covering the four cardinal directions, and selecting measurement points in each direction according to equal amounts and equal distances; The specific logic for collecting the meteorological data of the to-be-evaluated region in the recent one week is: Install anemometers, hygrometers, thermometers and barometers at measurement points; Record wind speed data, humidity data, temperature data and air pressure data at different measurement points to obtain the average wind speed, average humidity, average temperature and average air pressure of the to-be-evaluated region; The formula for obtaining the average wind speed is: wherein is the average wind speed, denotes the wind speed measured at the th measurement point; The formula for obtaining the average humidity is: wherein is the average humidity, denotes the humidity measured at the nth measurement point; The formula for obtaining the average temperature is: wherein is the average temperature, denotes the temperature measured at the th measurement point; The formula for obtaining the average air pressure is: wherein is the average air pressure, denotes the air pressure measured at the th measurement point.

3. The gale disaster weather index prediction system based on a long short-term memory network model according to claim 2, characterized in that: The specific logic for collecting the precipitation and cloud cover of the to-be-evaluated region one week before the gale disaster event occurred in the history is: The daily average precipitation and the daily average cloud cover of the week before the gale disaster event in the area to be evaluated are obtained from the historical weather data recorded by the meteorological bureau; The collected cloud cover is the proportion of the sky covered by the cloud layer in the disaster area when the gale disaster occurs, expressed in percentage, ranging from 0% to 100%; The specific logic for collecting the precipitation and cloud cover data of the area to be evaluated in the past week is: The precipitation and cloud cover data of the area to be evaluated in the past week are collected from the historical weather data recorded by the meteorological bureau and the prediction results of weather forecast.

4. The long short-term memory network model-based gale disaster weather index prediction system according to claim 1, characterized in that: The process of establishing an index prediction model based on a long short-term memory network specifically includes: The historical meteorological data is divided into multiple groups of time series data according to type and used as a training set, and the affected area and duration of the gale disaster that occurred are used as labels to build a deep learning model based on a long short-term memory network. The training set and labels are input into the deep learning model to train the model, and a trained index prediction model is obtained. The historical meteorological data includes wind speed, humidity, temperature, air pressure, precipitation, and cloud cover in the area to be evaluated. 5.The system for gale disaster weather index prediction based on long short-term memory network model according to claim 1, wherein: The specific logic for obtaining the predicted affected area and predicted duration of the gale weather is: The meteorological data of the area to be evaluated in the past week, including wind speed, humidity, temperature, air pressure, precipitation, and cloud cover, are input into the trained index prediction model, and the predicted affected area and predicted duration of the gale weather are obtained. 6.The system according to claim 5, wherein the system is configured to: The formula for obtaining the comprehensive weather index based on the collected topographic height and slope data of the area to be evaluated to correct the predicted affected area and predicted duration of the gale weather is as follows: wherein, is a comprehensive weather index, is a terrain height of the area to be evaluated, is a preset proportionality coefficient thereof, is a terrain slope of the area to be evaluated, is a preset proportionality coefficient thereof, is a predicted disaster area, is a predicted duration, is a preset proportionality coefficient of a constant correction index, , , all of which are greater than zero.

7. The wind disaster weather index prediction system based on a long short-term memory network model according to claim 6, characterized in that: Comprehensive Weather Index The process of classifying risk levels based on the comparison with preset assessment thresholds specifically includes: setting a series of thresholds corresponding to different levels of gale hazard; comparing the obtained comprehensive weather index with the preset thresholds; and classifying the hazard level based on the comparison results. When the prediction is low risk, the public can carry on outdoor activities normally, but need to pay attention to weather changes; When the risk is medium, it is suggested that the public reduce unnecessary outdoor activities and pay attention to securing loose objects to prevent them from being blown away by the wind; When the prediction is high risk, the public is advised to avoid outdoor activities, the relevant departments are advised to strengthen the patrol, to ensure the stability of public safety facilities, residents should ensure that the doors and windows are closed, and the outdoor items are fixed; When the prediction is extremely high risk, it is recommended that the public stay indoors and avoid all outdoor activities, and relevant departments should start emergency response measures, such as evacuating residents in high-risk areas and closing affected public facilities.

8. The application of the gale disaster weather index prediction system based on a long short-term memory network model according to any one of claims 1-7 in weather monitoring.

Images