Weather forecast method and system based on real-time data correction

Abstract

The invention relates to a weather forecast method and system based on real-time data correction, and the method comprises the following steps: collecting multi-source meteorological data, and carrying out the preprocessing of the multi-source meteorological data; constructing a multi-source data correction model based on the preprocessed multi-source meteorological data, obtaining weather forecast data predicted by the machine learning model, inputting the weather forecast data into the multi-source data correction model, and calculating to obtain a preliminarily corrected weather forecast result; constructing a short-time high-frequency prediction model, inputting the preprocessed multi-source meteorological data into the short-time high-frequency prediction model, and calculating to obtain a short-time high-frequency prediction result; constructing a regional optimization model, and inputting the preliminarily corrected weather forecast result into the regional optimization model to obtain a regional optimization result; and carrying out weighted summation on the preliminarily corrected weather forecast result, the short-time high-frequency prediction result and the regional optimization result to obtain a final weather forecast correction result.

Description

Technical Field

[0001] The invention relates to a weather forecast method and system based on real-time data correction, belonging to the technical field of meteorological forecasting. Background Art

[0002] Weather forecasting technology has evolved from empirical rules to numerical weather prediction (NWP). Modern weather forecasting relies on physical and mathematical models to predict weather by simulating physical processes such as the atmosphere, oceans, and land. However, these methods usually have the following shortcomings:

[0003] The numerical weather prediction model is currently the most common prediction method. It relies on the principles of atmospheric physics and predicts future weather by numerically solving atmospheric dynamics equations. However, the accuracy of the NWP model is limited by the accuracy of the initial and boundary conditions. Especially in short-term forecasts and local weather forecasts, the performance of the model is often unsatisfactory, especially in the prediction of extreme weather such as storms and heavy precipitation;

[0004] In recent years, artificial intelligence (AI) and machine learning (ML) have been introduced into the field of weather forecasting, especially in pattern recognition and feature extraction. Deep learning and ensemble learning have made some progress in improving forecast accuracy, but they still face problems such as overfitting, strong dependence on data quality, and poor interpretability, especially in the prediction of extreme weather events;

[0005] However, due to the diversity of geography and climate, local weather (such as urban heat island effect, mountain storms, etc.) is difficult to accurately predict through machine learning. In addition, the spatiotemporal mismatch and heterogeneity between different observation data sources (radar, satellites, ground meteorological stations, etc.) also affect the accuracy and reliability of the forecast. Summary of the invention

[0006] In order to solve the above problems existing in the prior art, the present invention proposes a weather forecast method and system based on real-time data correction.

[0007] The technical solution of the present invention is as follows:

[0008] In one aspect, the present invention provides a weather forecasting method based on real-time data correction, comprising the following steps:

[0009] Collect multi-source meteorological data and pre-process the multi-source meteorological data;

[0010] A multi-source data correction model is constructed based on the pre-processed multi-source meteorological data, the weather forecast data predicted by the machine learning model is obtained and input into the multi-source data correction model to calculate the preliminary corrected weather forecast results;

[0011] Construct a short-term high-frequency prediction model, input the pre-processed multi-source meteorological data into the short-term high-frequency prediction model to calculate the short-term high-frequency prediction results;

[0012] Construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results;

[0013] The preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results are weighted and summed to obtain the final weather forecast correction results.

[0014] As a preferred embodiment of the present invention, the multi-source meteorological data includes meteorological station data, remote sensing data and Internet of Things sensor data.

[0015] As a preferred embodiment of the present invention, the multi-source data correction model is specifically shown as follows:

[0016] P corr (x,t)=P nuM (x,t)+P real (x,t)

[0017] Where: P corr (x, t) represents the weather forecast result after preliminary correction at time t; P num (x, t) represents the weather forecast data at time t; P real (x, t) represents the correction value based on multi-source meteorological data at time t.

[0018] As a preferred embodiment of the present invention, the correction value calculation formula based on multi-source meteorological data is:

[0019]

[0020] Where: T t represents the temperature at time t; H t represents the humidity at time t; P t represents the air pressure at time t; W t represents the microclimate wind field at time t; T c,t represents the cloud top temperature at time t; R t represents the rainfall at time t; f TH represents the temperature and humidity correction function; f PW represents the pressure wind field correction function; represents the cloud top precipitation correction function; β1, β2, and β3 represent weight functions respectively.

[0021] As a preferred embodiment of the present invention, the short-term high-frequency prediction model is specifically shown in the following formula:

[0022] R short (t+Δt)=∫ω φ(F radar ,F sat ,T env )·W adv dω

[0023] Where: Δt represents the prediction time step; R short (t+Δt) represents the short-term high-frequency prediction result at time t+Δt; ω represents the prediction area; φ represents the data fusion function; F radar Indicates the radar reflectivity field characteristics; F sat Indicates the characteristics of satellite cloud images; T env Indicates environment variables; W adv Represents the advection velocity of the precipitation belt.

[0024] As a preferred embodiment of the present invention, the regional optimization model is specifically shown as follows:

[0025] P regional (P corr,z (x,t),t)=∫ ω K(z,z′)·P corr,z (x,t)dω

[0026] Where: P corr,z (x, t) represents the weather forecast result after preliminary correction at location z at time t; z′ represents the location of the meteorological station; K represents the terrain similarity function.

[0027] On the other hand, the present invention also provides a weather forecast system based on real-time data correction, including a data acquisition module, a preliminary correction module, a short-time high-frequency prediction module, a regional optimization module and a final correction module;

[0028] The data acquisition module is used to collect multi-source meteorological data and pre-process the multi-source meteorological data;

[0029] The preliminary correction module is used to construct a multi-source data correction model based on the pre-processed multi-source meteorological data, obtain the weather forecast data predicted by the machine learning model and input it into the multi-source data correction model to calculate and obtain the preliminary corrected weather forecast result;

[0030] The short-time high-frequency prediction module is used to construct a short-time high-frequency prediction model, and input the pre-processed multi-source meteorological data into the short-time high-frequency prediction model to calculate and obtain a short-time high-frequency prediction result;

[0031] The regional optimization module is used to construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results;

[0032] The final correction module is used to perform weighted summation on the preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results to obtain the final weather forecast correction result.

[0033] On the other hand, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any embodiment of the present invention is implemented.

[0034] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0035] The present invention has the following beneficial effects:

[0036] The present invention can continuously update the prediction results during the weather occurrence process, thereby improving the accuracy of short-term forecasts, and can also perform special local optimization to improve the accuracy of local weather forecasts. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0040] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0041] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0042] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0043] Embodiment 1:

[0044] See also Figure 1 , a weather forecast method based on real-time data correction, comprising the following steps:

[0045] Collect multi-source meteorological data and pre-process the multi-source meteorological data;

[0046] A multi-source data correction model is constructed based on the pre-processed multi-source meteorological data, the weather forecast data predicted by the machine learning model is obtained and input into the multi-source data correction model to calculate the preliminary corrected weather forecast results;

[0047] Construct a short-term high-frequency prediction model, input the pre-processed multi-source meteorological data into the short-term high-frequency prediction model to calculate the short-term high-frequency prediction results;

[0048] Construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results;

[0049] The preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results are weighted and summed to obtain the final weather forecast correction results.

[0050] As a preferred implementation scheme of this embodiment, the multi-source meteorological data includes meteorological station data (temperature, humidity, air pressure, etc.), remote sensing data (satellite images) and Internet of Things sensor data (sensors distributed in the city for monitoring temperature, humidity, air pressure, etc. in small areas).

[0051] As a preferred implementation of this embodiment, the multi-source data correction model is specifically shown as follows:

[0052] P corr (x,t)=P num (x,t)+P real (x,t)

[0053] Where: P corr (x, t) represents the weather forecast result after preliminary correction at time t; P num (x, t) represents the weather forecast data at time t; P real (x, t) represents the correction value based on multi-source meteorological data at time t.

[0054] As a preferred implementation of this embodiment, the correction value calculation formula based on multi-source meteorological data is:

[0055]

[0056] Where: T t represents the temperature at time t; Ht represents the humidity at time t; P t represents the air pressure at time t; W t represents the microclimate wind field at time t; T c,t represents the cloud top temperature at time t; R t represents the rainfall at time t; f TH represents the temperature and humidity correction function; f PW represents the pressure wind field correction function; represents the cloud top precipitation correction function; β1, β2, β3 represent weight functions respectively; γ represents the weight parameter;

[0057] The W t It means that the wind field affects the movement of the precipitation belt through horizontal advection and turbulent exchange, as shown in the following formula:

[0058]

[0059] Where: W x Represents the component of wind speed in the x direction; W y represents the component of wind speed in the y direction;

[0060] As a preferred implementation of this embodiment, the short-term high-frequency prediction model is mainly used to predict sudden weather (such as thunderstorms and short-term heavy rainfall). The short-term high-frequency prediction model is specifically shown in the following formula:

[0061] R short (t+Δt)=∫ ω φ(F radar ,F sat ,T env )·W adv dω

[0062] Where: Δt represents the prediction time step; R short (t+Δt) represents the short-term high-frequency prediction result at time t+Δt; ω represents the prediction area; φ represents the data fusion function; F radar Represents the radar reflectivity field characteristics, which is usually used to detect the spatial distribution of precipitation, thunderstorms, cloud structures, etc. Radar reflectivity (unit: dBZ) indicates the ability of a target object to reflect radar signals, which is usually proportional to the amount of precipitation. Changes in radar reflectivity can reveal the intensity and type of precipitation (such as heavy rain, light rain, or thunderstorms) in real time; F sat Indicates the characteristics of satellite cloud images (such as cloud top brightness temperature); T env Indicates environmental variables (temperature, humidity, etc.); W adv It represents the advection velocity of the precipitation belt, as shown in the following formula:

[0063]

[0064] As a preferred implementation of this embodiment, the regional optimization model is specifically shown as follows:

[0065] P regional (P corr,z (x,t),t)=∫ ω K(z,z′)·P corr,z (x,t)dω

[0066] Where: P corr,z (x, t) represents the weather forecast result after preliminary correction at location z at time t; z′ represents the location of the meteorological station; K represents the terrain similarity function, which is shown in the following formula:

[0067]

[0068] Where: ||zz′|| represents the spatial distance between the two locations; h(z) and h(z′) represent the altitudes of the two locations respectively; σ d , σ h Respectively represent the influence range of controlling spatial distance and terrain similarity.

[0069] Embodiment 2:

[0070] A weather forecast system based on real-time data correction, comprising a data acquisition module, a preliminary correction module, a short-term high-frequency prediction module, a regional optimization module and a final correction module;

[0071] The data acquisition module is used to collect multi-source meteorological data and pre-process the multi-source meteorological data;

[0072] The preliminary correction module is used to construct a multi-source data correction model based on the pre-processed multi-source meteorological data, obtain the weather forecast data predicted by the machine learning model and input it into the multi-source data correction model to calculate and obtain the preliminary corrected weather forecast result;

[0073] The short-time high-frequency prediction module is used to construct a short-time high-frequency prediction model, and input the pre-processed multi-source meteorological data into the short-time high-frequency prediction model to calculate and obtain a short-time high-frequency prediction result;

[0074] The regional optimization module is used to construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results;

[0075] The final correction module is used to perform weighted summation on the preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results to obtain the final weather forecast correction result.

[0076] The system is used to implement the method in Example 1, which will not be described in detail here.

[0077] Embodiment three:

[0078] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present invention when executing the program.

[0079] Embodiment 4:

[0080] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present invention is implemented.

[0081] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0082] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0083] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0084] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0085] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A weather forecast method based on real-time data correction, characterized in that: The following steps are involved: Collect multi-source meteorological data and pre-process the multi-source meteorological data; A multi-source data correction model is constructed based on the pre-processed multi-source meteorological data, the weather forecast data predicted by the machine learning model is obtained and input into the multi-source data correction model to calculate the preliminary corrected weather forecast results; Construct a short-term high-frequency prediction model, input the pre-processed multi-source meteorological data into the short-term high-frequency prediction model to calculate the short-term high-frequency prediction results; Construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results; The preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results are weighted and summed to obtain the final weather forecast correction results.

2. A weather forecast method based on real-time data correction according to claim 1, characterized in that: The multi-source meteorological data includes weather station data, remote sensing data and Internet of Things sensor data.

3. A weather forecast method based on real-time data correction according to claim 2, characterized in that: The multi-source data correction model is specifically shown as follows: P corr (x,t)=P num (x,t)+P real (x,t) Where: P corr (x, t) represents the weather forecast result after preliminary correction at time t; P num (x, t) represents the weather forecast data at time t; P real (x, t) represents the correction value based on multi-source meteorological data at time t.

4. A weather forecast method based on real-time data correction according to claim 3, characterized in that: The correction value calculation formula based on multi-source meteorological data is: Where: T t represents the temperature at time t; H t represents the humidity at time t; P t represents the air pressure at time t; W t represents the microclimate wind field at time t; T c,t represents the cloud top temperature at time t; R t represents the rainfall at time t; f TH represents the temperature and humidity correction function; f PW represents the pressure wind field correction function; represents the cloud top precipitation correction function; β1, β2, β3 represent weight functions respectively; γ represents the weight parameter.

5. A weather forecast method based on real-time data correction according to claim 4, characterized in that: The short-term high-frequency prediction model is specifically shown as follows: R short (t+Δt)=∫ ω φ(F radar ,F sat ,T env )·W adv dω Where: Δt represents the prediction time step; R short (t+Δt) represents the short-term high-frequency prediction result at time t+Δt; ω represents the prediction area; φ represents the data fusion function; F radar Indicates the radar reflectivity field characteristics; F sat Indicates the characteristics of satellite cloud images; T env Indicates environment variables; W adv Represents the advection velocity of the precipitation belt.

6. A weather forecast method based on real-time data correction according to claim 5, characterized in that: The regional optimization model is shown in the following formula: P regional (P corr,z (x,t),t)=∫ ω K(z,z ′ )·P corr,z (x,t)dω Where: P corr,z (x, t) represents the weather forecast result after preliminary correction at location z at time t; ′ represents the location of the weather station; K represents the terrain similarity function.

7. A weather forecast system based on real-time data correction, characterized in that: It includes data acquisition module, preliminary correction module, short-term high-frequency prediction module, regional optimization module and final correction module; The data acquisition module is used to collect multi-source meteorological data and pre-process the multi-source meteorological data; The preliminary correction module is used to construct a multi-source data correction model based on the pre-processed multi-source meteorological data, obtain the weather forecast data predicted by the machine learning model and input it into the multi-source data correction model to calculate and obtain the preliminary corrected weather forecast result; The short-time high-frequency prediction module is used to construct a short-time high-frequency prediction model, and input the pre-processed multi-source meteorological data into the short-time high-frequency prediction model to calculate and obtain a short-time high-frequency prediction result; The regional optimization module is used to construct a regional optimization model, and input the preliminary corrected weather forecast results into the regional optimization model to obtain the regional optimization results; The final correction module is used to perform weighted summation on the preliminary corrected weather forecast results, short-term high-frequency prediction results and regional optimization results to obtain the final weather forecast correction result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images