Weather clock forecasting system based on multi-source data fusion

Abstract

The invention relates to the technical field of weather forecasting, in particular to a weather clock forecasting system based on multi-source data fusion, which comprises an equipment terminal and a cloud end, the cloud end comprises a data acquisition module, a data processing module and a data transmission module, the data acquisition module is used for acquiring position information of the equipment terminal and multi-source data at the position, and the data processing module is used for processing the position information; the data processing module performs grid division on the target area range based on the position information of the equipment terminal to obtain grid units, and the data processing module is further used for mapping the position information of the equipment terminal into the grid units to obtain a target unit grid of the position where the equipment terminal is located. According to the method, the local feature data is introduced into the fine meteorological data of the target grid unit for correction, so that the finally obtained meteorological correction data has a finer space description capability, and the defect that prediction is inaccurate due to the fact that a traditional meteorological model ignores the local feature data under complex terrains or urban microclimate is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of weather forecasting, and in particular to a weather clock forecasting system based on multi-source data fusion. Background Art

[0002] Meteorological services are an important part of modern society. Their accuracy and timeliness directly affect people's production activities, social operations and daily life. With the development of modern technology, weather forecasts have changed from traditional empirical forecasts to numerical weather forecasts, and have gradually integrated advanced technologies such as big data, artificial intelligence and the Internet of Things. Weather trends within a day or a certain period of time are clearly and dynamically displayed to users through weather clocks, providing users with more convenient information services. However, forecasts from a single data source still have problems such as low spatial resolution, time delay and lack of accuracy.

[0003] After searching, Chinese patent number CN202211010895.8 discloses an emergency short-term forecast method based on multi-source data fusion, which packages multi-time multi-channel satellite cloud image sequences, multi-time radar echo image sequences and basic meteorological forecast factors under the same time sequence as data samples, performs preprocessing and data enhancement, and obtains training samples; constructs a network for fusing multi-source data, inputs the training samples into the network for training, and obtains a trained multi-source data fusion network; inputs the multi-time multi-channel satellite cloud image sequences, multi-time radar echo image sequences and basic meteorological forecast factors under the same time sequence before the prediction area that meet the short-term forecast effective time into the trained multi-source data fusion network, and performs weather forecasts for several future time periods in the region. The above scheme effectively integrates multiple data sources and makes up for the missing interactive information due to different data modes, which greatly improves the accuracy of forecasting emergency weather events and provides a strong early warning capability.

[0004] However, although the existing weather clock forecasting system has improved the forecast accuracy by correcting the historical deviation of grid data, it still has significant deficiencies in localization and short-term accurate forecasting; local weather is significantly affected by topography (such as mountains, river valleys, and plains). For example, precipitation in mountainous areas may be enhanced due to terrain uplift, but the existing correction methods are usually based on large-scale grid data and fail to fully combine high-resolution terrain data for optimization; in addition, urban microclimate also has a significant impact on regional temperature and wind field distribution, but the correction of grid data fails to fully capture these local characteristics.

[0005] Therefore, a weather clock forecasting system based on multi-source data fusion is proposed to solve the above problems. Summary of the invention

[0006] Technical issues solved

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a weather clock forecasting system based on multi-source data fusion, which can effectively solve the problem in the prior art that the historical deviation correction of large-scale grid data ignores the influence of local characteristics.

[0008] Technical Solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] The present invention provides a weather clock forecast system based on multi-source data fusion, including a device terminal and a cloud, the cloud including a data acquisition module, a data processing module and a data transmission module, the data acquisition module is used to collect the location information of the device terminal and the multi-source data at the location, the data processing module performs grid division on the target area range based on the location information of the device terminal to obtain grid units, the data processing module is also used to map the location information of the device terminal into the grid unit to obtain the target unit grid at the location of the device terminal; the target grid unit and the multi-source data are input into a meteorological matching model, and the fine meteorological data F of the target grid unit is calculated. qp ; and the detailed meteorological data F qp Input into the pre-trained meteorological correction model, and combine multi-source data to obtain the meteorological correction data F′ of the target grid unit qp ; The meteorological correction data F′ qp Input into the pre-trained weather prediction model to obtain the weather prediction result at the location of the device terminal; the data transmission module is used to transmit the weather prediction result to the device terminal.

[0011] Furthermore, the multi-source data includes:

[0012] Meteorological observation data: temperature, humidity, air pressure, wind speed and direction, precipitation, UV intensity and particle concentration;

[0013] Meteorological model numerical forecast data: temperature field, pressure field, wind field, humidity field and precipitation probability;

[0014] Historical meteorological data: daily temperature, daily precipitation, daily wind speed and direction, and periodic patterns;

[0015] Local characteristic data: microtopography changes, urban microclimate, water body details, vegetation types and distribution, soil conditions and local air pollution.

[0016] Furthermore, the method of dividing the target area into grids includes:

[0017] Define the center coordinates (lat device ,lon device) and the spatial radius R of the surrounding area region ; Set the space radius R region Converted into longitude and latitude value difference, the conversion formula is:

[0018] In the formula, Δlat is the distance in the latitude direction; Δlon is the distance in the longitude direction; cos(lat device ) indicates the current latitude of the device terminal;

[0019] Define the target area range A region ={(lat min ,lat max ), (lon min ,lon max )};

[0020] In the formula, (lat min ,lat max ) represents the boundary value of latitude in the north-south direction, (lon min ,lon max ) represents the boundary value of longitude in the east-west direction; lat min =lat device -Δlat, lat max =lat device +Δlat,lon min =lon device -Δlon,lon max =lon device +Δlon;

[0021] The target area is divided into regular grid cells based on the set grid resolution (Δx, Δy); Δx is the longitude difference, and Δy is the latitude difference Δy;

[0022] Based on target area range A region and the grid resolution (Δx, Δy) to generate the latitude index lat for each grid cell q and the longitude index lon p : The generating formula is:

[0023] lat q =lat min +(q-1)·Δy,q=1,2,...,n lat ; In the formula, lat q Indicates the minimum latitude of the qth grid cell; n lat Represents the total number of grid cells in the latitude direction, calculated as

[0024] lon p =lon min+(p-1)·Δx,p=1,2,...,m lon ;lon p Indicates the minimum longitude of the pth grid cell; m lon Represents the total number of grid cells in the latitude direction, and the calculation formula is;

[0025]

[0026] Based on the latitude index lat for each grid cell q and the longitude index lon p , calculate the center point coordinates of the grid cell (lat center ,lon center ) and generates a unique ID for each grid cell qp , where q and p correspond to the latitude index and longitude index of the grid cell respectively.

[0027] Furthermore, the method of mapping the location information of the device terminal to the grid unit includes:

[0028] Calculate the grid number q of the device terminal at latitude device and the grid number p in longitude device , get the target grid unit G where the device terminal is located qp and its center point coordinates The calculation formula is:

[0029]

[0030] Furthermore, the detailed meteorological data F qp The calculation method is:

[0031] The center point coordinates of the target grid cell The numerical forecast data of the meteorological model is input into the meteorological matching model to calculate the center point coordinates of the target grid cell and the center point coordinates of all meteorological model grids (lat k ,lon k ), the calculation formula is:

[0032]

[0033] Where k is the index of the meteorological model grid;

[0034] Find the meteorological model grid with the smallest distance D and use it as the object for target grid cell alignment; extract its meteorological data F k ={T k ,P k ,U k ,Rk ,V k ,D k ,C k} as the target grid cell G qp Initial meteorological data; where T k represents the temperature in the corresponding meteorological model grid; P k represents the probability of rainfall in the corresponding meteorological model grid; U k Represents the humidity in the corresponding meteorological model grid; R k Represents the ultraviolet intensity in the corresponding meteorological model grid; V k represents the average wind speed in the corresponding meteorological model grid; D k Indicates the direction in the corresponding meteorological model grid; C k represents the particle concentration in the corresponding meteorological model grid;

[0035] Compare the grid resolution (Δx, Δy) of the target grid cell with the grid resolution (Δx w ,Δy w ), if the grid resolution of the meteorological model grid is coarser than the grid resolution of the grid cell, the initial meteorological data is updated by the difference algorithm, and the update formula is:

[0036] In the formula, α k is the interpolation weight, and its calculation formula is: m=1, 2, 3, 4 represents the numbers of the four meteorological model grids adjacent to the target grid cell.

[0037] Furthermore, the weather correction data F′ qp The calculation formula is:

[0038] F′ qp =F qp +ΔF terrain +ΔF urban +ΔF water +ΔF vegetation +ΔF soil +ΔF pollution ;

[0039] Where, ΔF terrain is the correction value of micro-topography change; ΔF urban is the urban microclimate correction; ΔF water is the water detail correction; ΔF vegetation is the vegetation type and distribution correction; ΔF soil is the soil condition correction; ΔF pollution is the local air pollution correction;

[0040] And the detailed meteorological data Fqp Input into the pre-trained meteorological correction model, and combine multi-source data to obtain the meteorological correction data F′ of the target grid unit qp ; The meteorological correction data F′ qp Input into the pre-trained weather forecast model to obtain the weather forecast result for the location of the device terminal.

[0041] Furthermore, the micro-topography change correction amount ΔF terrain The calculation formula is:

[0042] ΔF terrain =δ H (H qp -H0)+δ S ·S qp +δ A ·f(A qp );where H qp is the altitude of the target grid cell; H0 is the reference altitude; S qp and A qp are the slope and aspect values ​​of the target grid cell respectively; δ H , δ S , δ A are the altitude weight coefficient, slope weight coefficient and aspect weight coefficient respectively, f(A qp ) is the slope function, f(A qp )=cos(A qp -A ref ), A ref is the reference slope aspect;

[0043] The urban microclimate correction ΔF urban The calculation formula is:

[0044] ΔF urban =γ UHI ·U qp +γ D ·D u,qp ; In the formula, U qp is the urban heat island effect intensity value of the target grid unit, D u,qp is the quantitative value of the building density of the target grid cell, γ UHI and γ D They are the urban heat island effect weight coefficient and the building density weight coefficient respectively;

[0045] Water detail correction ΔF water The calculation formula is:

[0046] ΔF water =λ W ·W qp +λ E ·Ew,qp Where W qp is the water area ratio of the target grid cell, E w,qp is the evaporation intensity of the target grid cell, λ W and λ E They are the water body area ratio weight coefficient and the water body evaporation intensity weight coefficient respectively;

[0047] Vegetation type and distribution correction ΔF vegetation The calculation formula is:

[0048] ΔF vegetation =ρ V ·V qp +ρ T ·f(T v,qp );Wherein, V qp is the vegetation coverage rate of the target grid cell, f(T v,qp ) is the vegetation type function, ρ V and ρ T They are respectively the vegetation coverage weight coefficient and the vegetation type weight coefficient;

[0049] Soil condition correction ΔF soil The calculation formula is:

[0050] ΔF soil =υ T ·f(T s,qp )+υ L ·L s,qp ; Among them, f(T s,qp ) is the soil type function, T s,qp is the soil type of the target grid cell, L s,qp is the soil utilization rate of the target grid unit, υ T and L They are soil type weight coefficient and soil utilization rate weight coefficient respectively;

[0051] Local air pollution correction ΔF pollution The calculation formula is:

[0052] ΔF pollution =ζ PM PM qp +ζ E ·E p,qp ; In the formula, PM qp is the particle concentration of the target grid cell, E p,qp is the pollution source intensity of the target grid unit, ζ PM and E They are the particulate matter concentration weight coefficient and the pollution source intensity weight coefficient respectively.

[0053] Furthermore, the training method of the weather correction model is:

[0054] Collect historical meteorological data of different target grid cells The historical fine meteorological data of different target grid cells are processed and used as the training input of the meteorological correction model;

[0055] Initialize the learnable parameters of the meteorological correction model, including the weight coefficient matrix of each influencing factor in the local characteristic data;

[0056] Define the loss function of the weather correction model

[0057] In the formula, is the revised predicted value; is the corresponding true observation value; n is the number of samples;

[0058] Define the regularized loss function for the weather correction model

[0059] In the formula, ω is the regularization coefficient, θ u is the learnable parameter in the meteorological correction model, u is the index label of the weight coefficient, is the regularization term of the sum of squares of all weight coefficients;

[0060] Update the learnable parameters θ using gradient descent u , to reduce the loss function L reg The value of is set; the historical detailed meteorological data is trained repeatedly until the meteorological correction model converges or reaches the preset number of iterations, that is, the training of the meteorological correction model is completed.

[0061] Furthermore, the training method of the weather prediction model includes:

[0062] Define the long short-term memory network model as the basic structure of the weather prediction model, which includes an input layer, a hidden layer, and an output layer;

[0063] Introduce the parameter vector time step d, and transform the vector time step d and the meteorological correction data F′ qp and historical weather data Combined to get the fusion feature vector Where s is the number of vector time steps d;

[0064] Collect historical meteorological data of the target grid unit within a fixed period of time in the past And the corresponding fusion feature vector is constructed, recorded as the historical fusion feature vector Fusion of history into feature vector Expand according to the vector time step to form sequence data as the training input of the weather prediction model. Correspondingly, the sequence data is labeled according to the vector time step d, and the label is the historical weather result

[0065] Initialize the network parameters of the weather prediction model; the network parameters include the weight matrix W from the input layer to the hidden layer x , the recurrent weight matrix W of the hidden layer h , the bias vector b of the hidden layer h , the weight matrix W from the hidden layer to the output layer o and the bias vector b of the output layer o ;

[0066] Define the loss function for the weather prediction model

[0067] In the formula, S is the length of the sequence data; Y true,d The label for the dth vector time step, i.e. the historical weather results Y pred,d is the label of the dth vector time step, that is, the predicted weather forecast result

[0068] The sequence data is passed to the input layer, and then passes through the hidden layer and the output layer in turn to output the predicted weather forecast results And calculate the corresponding weather forecast results The value of the loss function is obtained, and the error gradient is back-propagated from the output layer to the input layer; according to the error gradient, the network parameters are updated using the optimization algorithm to reduce the value of the loss function; the sequence data is trained repeatedly until the weather forecast model converges or reaches the preset number of iterations, which means that the weather forecast model training is completed.

[0069] Furthermore, the input layer is used to receive the fused feature vector X d As input; the hidden layer consists of y state vectors, and the calculation formula for each state vector is: d =f(W h ·h d-1 +W x ·X d +b h );where h d is the hidden state of the current vector at time step d, f(...) is the activation function, W x is the weight matrix from the input layer to the hidden layer, W h is the recurrent weight matrix of the hidden layer, b h is the bias vector of the hidden layer;

[0070] The output layer calculates the output of the weather forecast model based on the output of the hidden layer, that is, the weather forecast result. The calculation formula is: Among them, η(...) is the activation function of the output layer, W o is the weight matrix from the hidden layer to the output layer, b o is the bias vector of the output layer.

[0071] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0072] In this solution, the grid subdivision module takes the longitude and latitude coordinates of the device terminal as the center, divides the target area of ​​the device terminal into grids, obtains standard grid units, and maps the location information of the device terminal to the corresponding grid units; compared with traditional weather, which usually uses large-scale areas (such as cities or national geographical areas) divided by regular grids, this solution can refine and analyze multi-source data in a smaller spatial range by dynamically generating local grids with the device terminal as the center, and can better reflect small-scale weather differences in the region, such as local rainfall and wind speed changes; and compares the weather conditions around the target area (such as precipitation, wind force, temperature, etc.) with a larger range, so as to more accurately identify abnormal weather data related to the location of the device terminal; and by dynamically generating grid units with the device terminal location as the center, it can reduce meteorological calculations for non-essential areas, optimize computing performance, and improve the response speed of local weather forecasts;

[0073] In this scheme, local characteristic data are introduced into the fine meteorological data of the target grid unit for correction, so that the final meteorological correction data has a more refined spatial characterization capability, overcoming the defect of inaccurate prediction caused by traditional meteorological models in complex terrain or urban microclimates due to ignoring local characteristic data; and responding to the challenges of environmental diversity, being able to adapt to diverse landform environments such as mountainous areas, plains, and cities, and enhancing the sensitivity of data to extreme meteorological phenomena by dynamically adjusting the meteorological correction model, while making up for the shortcomings of traditional numerical weather forecast models in responding slowly to small-scale and large-gradient change characteristics, thereby improving the early warning capability of extreme weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 The schematic diagram of the structure of the weather clock forecast system in the embodiment of the present invention is shown in FIG. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0076] Embodiment 1:

[0077] Reference Figure 1 , this case proposes a weather clock forecast system based on multi-source data fusion, including a device terminal and a cloud, the device terminal and the cloud are connected by wireless signals; the cloud includes a data acquisition module, a data processing module and a data transmission module, the data acquisition module is used to obtain the location information of the device terminal, and obtain the multi-source data of the location based on the location information of the device terminal; the data processing module divides the target area into grids based on the location information of the device terminal to obtain grid units; then the location information of the device terminal is mapped into the grid units to obtain the target unit grid at the location of the device terminal; the target grid unit and the multi-source data are input into the meteorological matching model to calculate the fine meteorological data F of the target grid unit qp ; The fine meteorological data F of the target grid cell qp Input into the pre-trained meteorological correction model, and combine multi-source data to obtain the meteorological correction data F of the target grid unit q ' p ; Finally, the weather correction data F q ' p The data is input into the pre-trained weather forecast model to obtain the weather forecast result at the location of the device terminal; the data transmission module is used to send the weather forecast result to the device terminal for dynamic display.

[0078] Specifically, in this solution, the data acquisition module obtains the location information of the device terminal based on the location service LBS technology: LBS(t)={ID i ,(lat i ,lon i ),R accuracy}; Where, ID i It is a unique identifier of a device terminal (such as a device number or user ID) used to distinguish different device terminals and provide end-to-end services. i ,lon i ) is the latitude and longitude coordinates of the current location of the device terminal; R accuracy is the error radius of the device terminal location information; i is the unique identifier index of the device terminal; t is the timestamp, indicating the time when the device terminal location information is obtained, that is, the time dimension of the location information data.

[0079] The data acquisition module then uses the location information LBS(t) of the device terminal to obtain multi-source data at the location, including: meteorological observation data: temperature, humidity, air pressure, wind speed and direction, precipitation, ultraviolet intensity and particle concentration; meteorological model numerical forecast data: temperature field, air pressure field, wind field, humidity field and precipitation probability; historical meteorological data: daily temperature, daily precipitation, daily wind speed and direction and periodic laws; local characteristic data: micro-topography changes, urban microclimate, water body details, vegetation type and distribution, soil conditions and local air pollution.

[0080] More specifically, in this case, the collected location information LBS(t) and multi-source data are input into the data processing module; the data processing module cleans and calibrates the collected multi-source data, removes outliers in the data and fills in missing values, and then performs feature normalization on the data to map the value domain to a range; the location information LBS(t) is then input into the grid segmentation module, which uses the latitude and longitude coordinates (lat i ,lon i ) as the center, the surrounding area of ​​the device terminal is divided into grids (i.e. the target area range), and the division method is:

[0081] Define the center coordinates (lat device ,lon device ), the spatial radius R of the surrounding area region ; Set the space radius R region Converted into longitude and latitude value difference, the conversion formula is: and In the formula, Δlat is the distance in the latitude direction, which means the change of latitude in the target area; Δlon is the distance in the longitude direction, which means the change of longitude in the target area; the denominator of the conversion formula is 111, which means that 1° latitude or 1° longitude is approximately equal to 111 kilometers (a fixed constant on the earth's surface); cos(lat device ) represents the current latitude of the device terminal, that is, the latitude lat with the device terminal as the center point device The cosine value of adjusts the longitude distance;

[0082] Define the target area range A region ={(lat min ,lat max ), (lon min ,lon max )}; where (lat min ,lat max ) represents the boundary value of latitude in the north-south direction, (lon min ,lon max ) represents the boundary value of longitude in the east-west direction; latmin =lat device -Δlat, lat max =lat device +Δlat,lon min =lon device -Δlon,lon max =lon device +Δlon;

[0083] The target area is divided into regular grid cells. The division of the grid cells is based on the set grid resolution (Δx, Δy), including the longitude difference Δx and the latitude difference Δy. The longitude difference Δx represents the width of each grid cell in the longitude direction, and the latitude difference Δy represents the width of each grid cell in the latitude direction.

[0084] Based on the definition of target area range A region and the grid resolution (Δx, Δy) to generate the latitude index lat for each grid cell q and the longitude index lon p : The generating formula is:

[0085] lat q =lat min +(q-1)·Δy,q=1,2,...,n lat ; In the formula, lat q Indicates the minimum latitude of the qth grid cell, that is, the latitude value of the southern boundary of the grid cell; (q-1) means starting from the first grid cell, the current grid cell is from lat min Starting from the qth grid; n lat Represents the total number of grid cells in the latitude direction, calculated as

[0086] lon p =lon min +(p-1)·Δx,p=1,2,...,m lon ;lon p Indicates the minimum longitude of the pth grid cell, that is, the longitude value of the western boundary of the grid cell; (p-1) means starting from the first grid cell, the current grid cell is from lon min Starting from the pth grid; m lon Represents the total number of grid cells in the latitude direction, calculated as

[0087] Based on the latitude index lat for each grid cell q and the longitude index lon p , calculate the center point coordinates of the grid cell (lat center ,loncenter ) and generates a unique ID for each grid cell qp , where q and p correspond to the latitude index and longitude index of the grid cell respectively.

[0088] The data processing module maps the location information LBS(t) of the device terminal to the corresponding grid unit and obtains the grid number q of the device terminal in latitude. device and the grid number p in longitude device , the mapping formula is: In the formula, q device Indicates the row number corresponding to the device terminal location in the grid division, p device Indicates the column number corresponding to the device terminal position in the grid division, Indicates that the calculation result is rounded down, and the grid number starts from 1, so 1 needs to be added to the rounded value; then the grid number q of the device terminal at latitude is device and the grid number p in longitude device The latitude index of the grid cell lat q and the longitude index lon p Correspondingly, q device =lat q And p device =lon p , get the target grid unit G where the device terminal is located qp and its center point coordinates

[0089] Based on the above content, the grid subdivision module in this solution uses the latitude and longitude coordinates (lat i ,lon i ) is used as the center to grid the target area of ​​the device terminal to obtain standard grid units, and the location information LBS(t) of the device terminal is mapped to the corresponding grid unit; compared with traditional weather, which usually uses large areas (such as cities or national geographical areas) divided by regular grids, this solution dynamically generates local grids with the device terminal as the center, which can refine and analyze multi-source data in a smaller spatial range, and can better reflect small-scale weather differences in the region, such as local rainfall, wind speed changes, etc.; and compare the weather conditions around the target area (such as precipitation, wind force, temperature, etc.) with a larger range, which can more accurately identify weather anomaly data related to the location of the device terminal; and by dynamically generating grid units with the device terminal location as the center, it can reduce meteorological calculations for non-essential areas, optimize computing performance, and improve the response speed of local weather forecasts.

[0090] Furthermore, in this case, the data processing module calculates the center point coordinates of the target grid unit where the device terminal is located. The numerical forecast data of the meteorological model are input into the meteorological matching model, and the center point of the target grid cell is Corresponding to the center point coordinates of the meteorological model grid (lat k ,lon k ), find the nearest weather model grid;

[0091] The way to correspond the target grid cell to the meteorological model grid is to calculate the distance between the center point coordinates of the target grid cell and the center point coordinates of all meteorological model grids. The calculation formula is: Where k is the index of the meteorological model grid; then find the meteorological model grid with the smallest distance D and use it as the object of target grid unit alignment; extract its meteorological data as the target grid unit G qp Extract meteorological data F from the meteorological model grid k ={T k ,P k ,U k ,R k ,V k ,D k ,C k}; where T k represents the temperature in the corresponding meteorological model grid; P k represents the probability of rainfall in the corresponding meteorological model grid; U k Represents the humidity in the corresponding meteorological model grid; R k Represents the ultraviolet intensity in the corresponding meteorological model grid; V k represents the average wind speed in the corresponding meteorological model grid; D k Indicates the direction in the corresponding meteorological model grid; C k represents the particle concentration in the corresponding meteorological model grid;

[0092] Compare the grid resolution (Δx, Δy) of the target grid cell with the grid resolution (Δx w ,Δy w ), if the grid resolution of the meteorological model grid is coarser than the grid resolution of the grid cell, the initial meteorological data is updated by the difference algorithm to generate finer meteorological data of the target grid cell and expressed as a vector variable, that is, the fine meteorological data F of the target grid cell qp , and its calculation formula is: In the formula, α kis the interpolation weight, which represents the relative distance between the target grid cell and each neighboring point (i.e., the adjacent meteorological model grid), and is used to determine the fine meteorological data F of each neighboring point to the target grid cell. qp The contribution degree is calculated as follows: m=1, 2, 3, 4 represents the numbers of the four meteorological model grids adjacent to the target grid cell; when the meteorological model grid has a lower resolution, the target grid cell requires more refined meteorological data, and the meteorological information of the target grid cell is uniformly adjusted by interpolation.

[0093] Furthermore, in this case, the data processing module is based on the local characteristic data and the fine meteorological data F of the target grid unit. qp Construct and train the meteorological correction model to obtain the meteorological correction data F′ of the target grid cell qp , improve the accuracy of subsequent weather forecast results;

[0094] Specifically, the construction method of the meteorological correction model includes:

[0095] Based on the linear superposition method, the contribution of each influencing factor to the meteorological variables in the local characteristic data is corrected to define the meteorological correction data F′ qp The calculation formula is:

[0096] F′ qp =F qp +ΔF terrain +ΔF urban +ΔF water +ΔF vegetation +ΔF soil +ΔF pollution ;

[0097] Where, ΔF terrain It is the correction value for micro-topography changes, including the influence of changes in altitude, slope, and aspect parameters. Usually, the temperature decreases with increasing altitude, generally about 0.65℃ / 100m. The slope has a significant impact on precipitation collection and surface runoff, indirectly regulating humidity and precipitation. The aspect affects the distribution of solar radiation, which in turn affects the temperature (e.g., the sunny slope in the northern hemisphere is hotter). The calculation formula is: ΔF terrain =δ H (H qp -H0)+δ S ·S qp +δ A ·f(A qp ), where H qp is the altitude of the target grid cell; H0 is the reference altitude; S qp and A qp are the slope and aspect values ​​of the target grid cell respectively; δ H , δS , δ A are the altitude weight coefficient, slope weight coefficient and aspect weight coefficient, which are determined based on the actual experience of the region; f(A qp ) is the aspect function, which describes the degree of correction of the specific aspect to the meteorological variables. qp )=cos(A qp -A ref ), A ref is the reference slope aspect;

[0098] ΔF urban ΔF is the urban microclimate correction, including the urban heat island effect and the influence of building height and density. The temperature in urbanized areas is generally higher than that in surrounding areas. The specific intensity is related to the urban form and area. High-density buildings change the wind field distribution, affecting the temperature and wind speed. The calculation formula is: urban =γ UHI ·U qp +γ D ·D u,qp Among them, U qp is the urban heat island effect intensity value of the target grid unit, D u,qp is the quantitative value of the building density of the target grid cell, γ UHI and γ D They are the urban heat island effect weight coefficient and the building density weight coefficient respectively;

[0099] ΔF water ΔF is the water body detail correction, including the influence of water body area ratio and water body evaporation intensity. Water bodies have a regulating effect on the surrounding climate and significantly affect the temperature distribution (cooling in summer and warming in winter). The evaporation intensity of water bodies affects the local humidity and precipitation distribution. The calculation formula is: ΔF water =λ W ·W qp +λ E ·E w,qp ; Among them, W qp is the water area ratio of the target grid cell, E w,qp is the evaporation intensity of the target grid cell, λ W and λ E They are the water body area ratio weight coefficient and the water body evaporation intensity weight coefficient respectively;

[0100] ΔF vegetation It is the correction amount for vegetation type and distribution, including the influence of vegetation coverage and vegetation type. The temperature fluctuation is small and the humidity is high in the area with high vegetation coverage. The shading and transpiration of different types of vegetation have different regulating abilities on the regional climate. The calculation formula is: ΔF vegetation =ρ V ·V qp +ρ T ·f(Tv,qp ), where V qp is the vegetation coverage rate of the target grid cell, f(T v,qp ) is the vegetation type function, which is used to describe the vegetation type T of the target grid cell. v,qp The degree of influence on meteorological variables is obtained through empirical values, ρ V and ρ T They are respectively the vegetation coverage weight coefficient and the vegetation type weight coefficient;

[0101] ΔF soil It is the soil condition correction amount, including the influence of soil type and soil utilization rate. The thermodynamic properties of soil determine the surface heat exchange efficiency. The utilization methods such as farming and construction change the soil albedo and drainage capacity. The calculation formula is: ΔF soil =υ T ·f(T s,qp )+υ L ·L s,qp ; Among them, f(T s,qp ) is the soil type function, which is used to describe the soil type T of the target grid cell. s,qp The degree of influence on meteorological variables, L s,qp is the soil utilization rate of the target grid unit, υ T and L They are soil type weight coefficient and soil utilization rate weight coefficient respectively;

[0102] ΔF pollution is the local air pollution correction, including the influence of atmospheric particulate matter concentration and pollution source intensity, the temperature drop in high PM concentration areas, industrial emissions, traffic pollution, etc. affecting the local microclimate; the calculation formula is: ΔF pollution =ζ PM PM qp +ζ E ·E p,qp Among them, PM qp is the particle concentration of the target grid cell, E p,qp is the pollution source intensity of the target grid unit, ζ PM and E They are the particulate matter concentration weight coefficient and the pollution source intensity weight coefficient respectively.

[0103] The training method of the meteorological correction model is to collect historical meteorological data of different target grid cells The historical fine meteorological data of different target grid cells are processed and used as the training input of the meteorological correction model;

[0104] Initialize the learnable parameters of the meteorological correction model, including the weight coefficient matrix of each influencing factor in the local characteristic data;

[0105] Define the loss function of the weather correction model

[0106] In the formula, is the corrected forecast value, that is, the output of the meteorological correction model after correcting the historical fine meteorological data of the target grid cell with latitude index q and longitude index p; is the corresponding true observation value; n is the number of samples;

[0107] Define the regularized loss function for the weather correction model

[0108] In the formula, ω is the regularization coefficient, θ u is the learnable parameter in the meteorological correction model (i.e., different weight coefficients, such as the water area ratio weight coefficient λ UHI , urban heat island effect weight coefficient γ UHI etc.), u is the index label of the weight coefficient, is the regularization term of the sum of squares of all weight coefficients, which is used to constrain the weight value of the model;

[0109] Update the learnable parameters θ using gradient descent u , to reduce the loss function L reg The historical detailed meteorological data are repeatedly trained until the meteorological correction model converges (that is, the value of the loss function no longer changes) or reaches the preset number of iterations, that is, the training of the meteorological correction model is completed.

[0110] In the target grid unit's fine meteorological data F qp The local characteristic data is introduced into the final meteorological correction data F′ qp It has a more detailed spatial characterization capability, overcoming the defect of traditional meteorological models that ignores local characteristic data in complex terrain or urban microclimates and causing inaccurate predictions; it can also cope with the challenges of environmental diversity and adapt to diverse landform environments such as mountainous areas, plains, and cities. It can enhance the sensitivity of data to extreme meteorological phenomena by dynamically adjusting the meteorological correction model, while making up for the shortcomings of traditional numerical weather forecast models in responding slowly to small-scale and large-gradient change characteristics, thereby improving the early warning capability of extreme weather.

[0111] It is worth noting that in this case, the weather forecast model is trained in the following ways:

[0112] Define the long short-term memory network model as the basic structure of the weather forecast model, which includes input layer, hidden layer and output layer;

[0113] Introduce the parameter vector time step d, and transform the vector time step d and the meteorological correction data F′ qp and historical weather data Combined to get the fusion feature vector Where s is the number of vector time steps d;

[0114] The input layer is used to receive the fused feature vector X d As input; the hidden layer consists of y state vectors, and the calculation formula for each state vector is: d =f(W h ·h d-1 +W x ·X d +b h );where h d is the hidden state of the current vector at time step d, f(...) is the activation function, W x is the weight matrix from the input layer to the hidden layer, W h is the recurrent weight matrix of the hidden layer, b h is the bias vector of the hidden layer;

[0115] The output layer is based on the output of the hidden layer, i.e., the hidden state h at the current vector time step d. d , calculate the output of the weather forecast model, that is, the weather forecast result The calculation formula is: Among them, η(...) is the activation function of the output layer, W o is the weight matrix from the hidden layer to the output layer, b o is the bias vector of the output layer;

[0116] Collect historical meteorological data of the target grid unit within a fixed period of time in the past And the corresponding fusion feature vector is constructed, recorded as the historical fusion feature vector Fusion of history into feature vector Expand according to the vector time step to form sequence data as the training input of the weather prediction model. Correspondingly, the sequence data is labeled according to the vector time step d, and the label is the historical weather result

[0117] Initialize the network parameters of the weather prediction model; the network parameters include the weight matrix W from the input layer to the hidden layer x , the recurrent weight matrix W of the hidden layer h , the bias vector b of the hidden layer h , the weight matrix W from the hidden layer to the output layer o and the bias vector b of the output layer o ;

[0118] Define the loss function for the weather prediction model

[0119] In the formula, S is the length of the sequence data; Ytrue,d The label for the dth vector time step, i.e. the historical weather results Y pred,d is the label of the dth vector time step, that is, the predicted weather forecast result

[0120] The sequence data is passed to the input layer, and then passes through the hidden layer and the output layer in turn to output the predicted weather forecast results And calculate the corresponding weather forecast results The value of the loss function is obtained, and the error gradient is back-propagated from the output layer to the input layer; according to the error gradient, the network parameters are updated using an optimization algorithm (such as gradient descent, Adam, etc.) to reduce the value of the loss function; the sequence data is trained repeatedly until the weather forecast model converges (that is, the value of the loss function no longer changes) or reaches the preset number of iterations, which means that the weather forecast model training is completed.

[0121] Based on historical meteorological data, we can capture the long-term trends and periodic changes in the weather conditions of the target grid cells. Combining these characteristics can effectively improve the accuracy of short-term and medium-term weather forecasts and provide a basis for continuous optimization of the system. The model can enhance its adaptability to different meteorological conditions by continuously introducing new historical data for online learning.

[0122] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A weather clock forecast system based on multi-source data fusion, comprising a device terminal and a cloud, wherein the cloud comprises a data acquisition module, a data processing module and a data transmission module, wherein the data acquisition module is used to collect the location information of the device terminal and the multi-source data at the location, and wherein: The data processing module divides the target area into grids based on the location information of the device terminal to obtain grid units. The data processing module is also used to map the location information of the device terminal into the grid units to obtain the target unit grid where the device terminal is located; The target grid cell and multi-source data are input into the meteorological matching model to calculate the fine meteorological data F of the target grid cell. qp ; and the detailed meteorological data F qp Input into the pre-trained meteorological correction model, and combine multi-source data to obtain the meteorological correction data F of the target grid unit q ' p ; The weather correction data F q ' p Input into the pre-trained weather forecast model to obtain the weather forecast result at the location of the device terminal; The data transmission module is used to transmit weather forecast results to the device terminal.

2. The weather clock forecast system based on multi-source data fusion according to claim 1 is characterized in that: The multi-source data includes: Meteorological observation data: temperature, humidity, air pressure, wind speed, wind direction, precipitation, UV intensity and particle concentration; Meteorological model numerical forecast data: temperature field, pressure field, wind field, humidity field and precipitation probability; Historical meteorological data: daily temperature, daily precipitation, daily wind speed and direction, and periodic patterns; Local characteristic data: microtopography changes, urban microclimate, water body details, vegetation types and distribution, soil conditions and local air pollution.

3. The weather clock forecast system based on multi-source data fusion according to claim 2 is characterized in that: The method of gridding the target area range includes: Define the center coordinates (lat device ,lon device ) and the spatial radius R of the surrounding area region ; Set the space radius R region Converted into longitude and latitude value difference, the conversion formula is: and In the formula, Δlat is the distance in the latitude direction; Δlon is the distance in the longitude direction; cos(lat device ) indicates the current latitude of the device terminal; Define the target area range A region ={(lat min ,lat max ), (lon min ,lon max )}; In the formula, (lat min ,lat max ) represents the boundary value of latitude in the north-south direction, (lon min ,lon max ) represents the boundary value of longitude in the east-west direction; lat min =lat device -Δlat, lat max =lat device +Δlat,lon min =lon device -Δlon,lon max =lon device +Δlon; The target area is divided into regular grid cells based on the set grid resolution (Δx, Δy); Δx is the longitude difference, and Δy is the latitude difference Δy; Based on the target area range A region and the grid resolution (Δx, Δy) to generate the latitude index lat for each grid cell q and the longitude index lon p : The generating formula is: lat q =lat min +(q-1)·Δy,q=1,2,...,n lat ; In the formula, lat q Indicates the minimum latitude of the qth grid cell; n lat Represents the total number of grid cells in the latitude direction, calculated as lon p =lon min +(p-1)·Δx,p=1,2,...,m lon ;lon p Indicates the minimum longitude of the pth grid cell; m lon Represents the total number of grid cells in the latitude direction, calculated as Based on the latitude index lat for each grid cell q and the longitude index lon p , calculate the center point coordinates of the grid cell (lat center ,lon center ) and generates a unique ID for each grid cell qp , where q and p correspond to the latitude index and longitude index of the grid cell respectively.

4. The weather clock forecast system based on multi-source data fusion according to claim 3 is characterized in that: The method of mapping the location information of the device terminal to the grid unit includes: Calculate the grid number q of the device terminal at latitude device and the grid number p in longitude device , get the target grid unit G where the device terminal is located qp and its center point coordinates The calculation formula is:

5. The weather clock forecast system based on multi-source data fusion according to claim 1 is characterized in that: The detailed meteorological data F qp The calculation method is: The center point coordinates of the target grid cell The numerical forecast data of the meteorological model is input into the meteorological matching model to calculate the center point coordinates of the target grid cell and the center point coordinates of all meteorological model grids (lat k ,lon k ), the calculation formula is: Where k is the index of the meteorological model grid; Find the meteorological model grid with the smallest distance D and use it as the object for target grid cell alignment; extract its meteorological data F k ={T k ,P k ,U k ,R k ,V k ,D k ,C k } as the target grid cell G qp Initial meteorological data of k represents the temperature in the corresponding meteorological model grid; P k represents the probability of rainfall in the corresponding meteorological model grid; U k Represents the humidity in the corresponding meteorological model grid; R k Represents the ultraviolet intensity in the corresponding meteorological model grid; V k represents the average wind speed in the corresponding meteorological model grid; D k Indicates the direction in the corresponding meteorological model grid; C k represents the particle concentration in the corresponding meteorological model grid; Compare the grid resolution (Δx, Δy) of the target grid cell with the grid resolution (Δx w ,Δy w ), if the grid resolution of the meteorological model grid is coarser than the grid resolution of the grid cell, the initial meteorological data is updated by the difference algorithm, and the update formula is: In the formula, α k is the interpolation weight, and its calculation formula is: m=1, 2, 3, 4 represents the numbers of the four meteorological model grids adjacent to the target grid cell.

6. The weather clock forecast system based on multi-source data fusion according to claim 1 is characterized in that: The weather correction data F q ' p The calculation formula is: F′ qp =F qp +ΔF terrain +ΔF urban +ΔF water +ΔF vegetation +ΔF soil +ΔF pollution ; Where, ΔF terrain is the correction value of micro-topography change; ΔF urban is the urban microclimate correction; ΔF water is the water detail correction; ΔF vegetation is the vegetation type and distribution correction; ΔF soil is the soil condition correction; ΔF pollution is the local air pollution correction; And the detailed meteorological data F qp Input into the pre-trained meteorological correction model, and combine multi-source data to obtain the meteorological correction data F′ of the target grid unit qp ; The meteorological correction data F′ qp Input into the pre-trained weather forecast model to obtain the weather forecast result for the location of the device terminal.

7. The weather clock forecast system based on multi-source data fusion according to claim 6 is characterized in that: The micro-topography change correction amount ΔF terrain The calculation formula is: ΔF terrain =δ H (H qp -H0)+δ S ·S qp +δ A ·f(A qp );where H qp is the altitude of the target grid cell; H0 is the reference altitude; S qp and A qp are the slope and aspect values ​​of the target grid cell respectively; δ H , δ S , δ A are the altitude weight coefficient, slope weight coefficient and aspect weight coefficient respectively, f(A qp ) is the slope function, f(A qp )=cos(A qp -A ref ), A ref is the reference slope aspect; The urban microclimate correction ΔF urban The calculation formula is: ΔF urban =γ UHI ·U qp +γ D ·D u,qp ; In the formula, U qp is the urban heat island effect intensity value of the target grid unit, D u,qp is the quantitative value of the building density of the target grid cell, γ UHI and γ D They are the urban heat island effect weight coefficient and the building density weight coefficient respectively; Water detail correction ΔF water The calculation formula is: ΔF water =λ W ·W qp +λ E ·E w,qp Where W qp is the water area ratio of the target grid cell, E w,qp is the evaporation intensity of the target grid cell, λ W and λ E They are the water body area ratio weight coefficient and the water body evaporation intensity weight coefficient respectively; Vegetation type and distribution correction ΔF vegetation The calculation formula is: ΔF vegetation =ρ V ·V qp +ρ T ·f(T v,qp );Wherein, V qp is the vegetation coverage rate of the target grid cell, f(T v,qp ) is the vegetation type function, ρ V and ρ T They are respectively the vegetation coverage weight coefficient and the vegetation type weight coefficient; Soil condition correction ΔF soil The calculation formula is: ΔF soil =υ T ·f(T s,qp )+υ L ·L s,qp ; Among them, f(T s,qp ) is the soil type function, T s,qp is the soil type of the target grid cell, L s,qp is the soil utilization rate of the target grid unit, υ T and L They are soil type weight coefficient and soil utilization weight coefficient respectively; Local air pollution correction ΔF pollution The calculation formula is: ΔF pollution =ζ PM PM qp +ζ E ·E p,qp ; In the formula, PM qp is the particle concentration of the target grid cell, E p,qp is the pollution source intensity of the target grid unit, ζ PM and E They are the particulate matter concentration weight coefficient and the pollution source intensity weight coefficient respectively.

8. The weather clock forecast system based on multi-source data fusion according to claim 6 is characterized in that: The training method of the weather correction model is: Collect historical meteorological data of different target grid cells The historical fine meteorological data of different target grid cells are processed and used as the training input of the meteorological correction model; Initialize the learnable parameters of the meteorological correction model, including the weight coefficient matrix of each influencing factor in the local characteristic data; Define the loss function of the weather correction model In the formula, is the revised predicted value; is the corresponding true observation value; n is the number of samples; Define the regularized loss function for the weather correction model In the formula, ω is the regularization coefficient, θ u is the learnable parameter in the meteorological correction model, u is the index label of the weight coefficient, is the regularization term of the sum of squares of all weight coefficients; Update the learnable parameters θ using gradient descent u , to reduce the loss function L reg The value of is set; the historical detailed meteorological data is trained repeatedly until the meteorological correction model converges or reaches the preset number of iterations, that is, the training of the meteorological correction model is completed.

9. The weather clock forecast system based on multi-source data fusion according to claim 8, characterized in that: The training method of the weather prediction model includes: Define the long short-term memory network model as the basic structure of the weather prediction model, which includes an input layer, a hidden layer, and an output layer; Introduce the parameter vector time step d, and transform the vector time step d and the meteorological correction data F q ' p and historical weather data Combined to get the fusion feature vector Where s is the number of vector time steps d; Collect historical meteorological data of the target grid unit within a fixed period of time in the past And the corresponding fusion feature vector is constructed, recorded as the historical fusion feature vector Fusion of history into feature vector Expand according to the vector time step to form sequence data as the training input of the weather prediction model. Correspondingly, the sequence data is labeled according to the vector time step d, and the label is the historical weather result Initialize the network parameters of the weather prediction model; the network parameters include the weight matrix W from the input layer to the hidden layer x , the recurrent weight matrix W of the hidden layer h , the bias vector b of the hidden layer h , the weight matrix W from the hidden layer to the output layer o and the bias vector b of the output layer o ; Define the loss function for the weather prediction model In the formula, S is the length of the sequence data; Y true,d The label for the dth vector time step, i.e. the historical weather results Y pred,d is the label of the dth vector time step, that is, the predicted weather forecast result The sequence data is passed to the input layer, and then passes through the hidden layer and the output layer in turn to output the predicted weather forecast results And calculate the corresponding weather forecast results The value of the loss function is obtained, and the error gradient is back-propagated from the output layer to the input layer; according to the error gradient, the network parameters are updated using the optimization algorithm to reduce the value of the loss function; the sequence data is trained repeatedly until the weather forecast model converges or reaches the preset number of iterations, which means that the weather forecast model training is completed.

10. The weather clock forecast system based on multi-source data fusion according to claim 9, characterized in that: The input layer is used to receive the fused feature vector X d As input; the hidden layer consists of y state vectors, and the calculation formula for each state vector is: d =f(W h ·h d-1 +W x ·X d +b h );where h d is the hidden state of the current vector at time step d, f(...) is the activation function, W x is the weight matrix from the input layer to the hidden layer, W h is the recurrent weight matrix of the hidden layer, b h is the bias vector of the hidden layer; The output layer calculates the output of the weather forecast model based on the output of the hidden layer, that is, the weather forecast result. The calculation formula is: Among them, η(...) is the activation function of the output layer, W o is the weight matrix from the hidden layer to the output layer, b o is the bias vector of the output layer.

Images