Meteorological drought early warning method, device, electronic device and storage medium

Abstract

The present invention discloses a meteorological drought early warning method, device, electronic device, and storage medium. The method comprises: determining target environmental data for a target area; determining key hazard factor values using a prediction model based on the target environmental data; determining drought assessment results from different assessment dimensions based on the key hazard factor values; determining a drought severity based on the drought assessment results; and issuing a drought early warning based on the drought severity. This method ensures the physical consistency of the key hazard factor values, assesses drought from different assessment dimensions based on the key hazard factor values, and comprehensively considers the drought severity, thereby improving the accuracy of drought predictions.

Description

Technical Field

[0001] The present invention relates to the field of meteorological technology, and in particular to a meteorological drought early warning method, device, electronic equipment and storage medium. Background Art

[0002] With global warming, the frequency and intensity of meteorological drought events are increasing, which has a profound impact on agriculture, water resources, ecosystems and human life. Once it occurs, it will cause significant social and economic losses.

[0003] With the rapid development of big data science in recent years, artificial intelligence, particularly deep learning technology, has achieved breakthroughs, with significant application results in fields such as large-scale meteorological models and disaster warning. However, existing drought warning methods based on deep learning suffer from inaccurate drought warnings. Summary of the Invention

[0004] The present invention provides a meteorological drought early warning method, device, electronic equipment and storage medium to solve the problem of inaccurate drought prediction.

[0005] According to one aspect of the present invention, a meteorological drought early warning system is provided, comprising:

[0006] Determining target environmental data for a target area, wherein the target environmental data is used to characterize the changing trends of the atmosphere and geology of the target area;

[0007] Determining a key disaster factor value through a prediction model based on the target environmental data, wherein the key disaster factor value is a corresponding range of variation when the key disaster factor changes under the target environmental data. The key disaster factor is a factor that causes drought in the target area. The prediction model is configured with a physical constraint function, which serves as a loss function to constrain the physical laws represented by the prediction results of the prediction model. The physical constraint function includes: a soil water balance equation and a high temperature tracking equation. The soil water balance equation is used to represent the relationship between precipitation and soil surface moisture and vegetation status. The high temperature tracking equation is used to track temperature changes.

[0008] Drought assessment results are determined from different assessment dimensions based on the values of the key disaster-causing factors, drought levels are determined based on the drought assessment results, and drought warnings are issued based on the drought levels.

[0009] According to another aspect of the present invention, a meteorological drought early warning device is provided, comprising:

[0010] A data determination module is used to determine target environmental data of a target area, wherein the target environmental data is used to characterize the changing trend of the atmosphere and geology of the target area;

[0011] A key disaster factor value determination module is configured to determine the key disaster factor value through a prediction model based on the target environmental data. The key disaster factor value is the corresponding range of variation when the key disaster factor changes under the target environmental data. The key disaster factor is a factor that causes drought in the target area. The prediction model is configured with a physical constraint function. The physical constraint function serves as a loss function to constrain the physical laws represented by the prediction results of the prediction model. The physical constraint function includes: a soil water balance equation and a high temperature tracking equation. The soil water balance equation is used to represent the relationship between precipitation and soil surface moisture and vegetation status. The high temperature tracking equation is used to track temperature changes.

[0012] The early warning module is used to determine drought assessment results from different assessment dimensions according to the values of the key disaster-causing factors, determine the drought level according to the drought assessment results, and issue drought early warning according to the drought level.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the meteorological drought early warning method described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the meteorological drought early warning method according to any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention determines target environmental data for a target area; determines key disaster factor values based on the target environmental data through a prediction model. The obtained key disaster factor values can meet the physical laws of meteorological changes, so that the obtained key disaster factors can better represent the changes in future drought influencing factors in the target area; determines drought assessment results from different assessment dimensions based on the key disaster factor values, determines the drought level based on the drought assessment results, and issues drought warnings based on the drought level. This multi-dimensional drought level assessment can ensure the accuracy of drought level assessment. This method can ensure the physical consistency of the key disaster factor values, assess drought from different assessment dimensions based on the key disaster factor values, and comprehensively consider the drought level to improve the accuracy of drought prediction.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flowchart of a meteorological drought early warning method provided by an embodiment of the present invention;

[0022] Figure 2 A flowchart of another meteorological drought early warning method provided by an embodiment of the present invention;

[0023] Figure 3 A schematic structural diagram of a meteorological drought early warning device provided by an embodiment of the present invention;

[0024] Figure 4 A schematic diagram of the structure of an electronic device for implementing the meteorological drought early warning method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Figure 1 This is a flow chart of a meteorological drought early warning method provided by an embodiment of the present invention. This embodiment is applicable to situations where drought in a target area is warned based on collected environmental data. The method can be executed by a meteorological drought early warning device. The meteorological drought early warning device can be implemented in the form of hardware and / or software. The meteorological drought early warning device can be configured in any electronic device with network communication capabilities. Figure 1 As shown, the method includes:

[0028] S110: Determine target environmental data of the target area.

[0029] Among them, target environmental data is used to characterize the changing trends of the atmosphere and geology in the target area.

[0030] Specifically, the original environmental data of the target area is cleaned, missing values are filled, the data is normalized or standardized, and the temporal and spatial resolutions are processed for consistency to obtain the target environmental data.

[0031] Furthermore, the consistency processing of spatiotemporal resolution is to unify the time scale and spatial scale of historical original environmental data, that is, the historical meteorological data, historical climate data, historical hydrological remote sensing data and historical disaster-prone environmental data collected at the same time and in the same period are unified in the time dimension and spatial dimensions such as districts and counties.

[0032] Raw environmental data include meteorological data, climate data, remote sensing data, and disaster-prone environmental data. Meteorological data include local meteorological variables such as precipitation, evaporation, and surface temperature, which serve as disaster-causing factors. Climate data include sea surface temperature, the Southern Oscillation Index (SOI), and the North Atlantic Oscillation (NAO). Remote sensing data, including the Normalized Difference Vegetation Index (NDVI), serves as a potential predictor of drought in areas with strong remote sensing correlations. Disaster-prone environmental data include soil moisture and land use types, which reflect the initial state of the land.

[0033] The above steps of preprocessing the data are to ensure the consistency of the collected data in space and time, so as to ensure that the data for subsequent drought prediction belong to the same time and space, thereby improving the accuracy of the prediction.

[0034] S120. Determine the values of key disaster-causing factors through a prediction model based on target environmental data.

[0035] Among them, the value of the key disaster factor is the corresponding range of change when the key disaster factor changes under the target environmental data. The key disaster factor is the factor that causes drought in the target area. The prediction model is configured with a physical constraint function. The physical constraint function serves as a loss function to constrain the physical laws represented by the prediction results of the prediction model. The physical constraint functions include: soil water balance equation and high temperature tracking equation. The soil water balance equation is used to characterize the relationship between precipitation and soil surface moisture and vegetation status. The high temperature tracking equation is used to track temperature changes.

[0036] Specifically, the target environmental data is input into the prediction model, and the values of key disaster-causing factors are calculated by the prediction model.

[0037] Furthermore, the key disaster factor value can be a single data value or a range of values. The key disaster factor is determined based on the historical environmental data of the target area. The key disaster factor can be data such as precipitation, soil moisture and temperature.

[0038] Furthermore, the prediction model consists of a neural network model, a physical constraint function, and boundary conditions. The physical constraint function, as the loss function of the neural network model, constrains the prediction results so that the values of the key disaster-causing factors conform to physical laws, while the boundary conditions are used to provide boundaries for the neural network model's predictions.

[0039] Among them, the neural network model can adopt the PINN network (Physics-informed Neural Network). The PINN network is a machine learning model that combines deep learning and physical laws. Unlike traditional neural networks, the PINN network uses physical laws to constrain the model during the learning process, thereby improving the generalization ability of the model.

[0040] Physical constraint functions include: soil water balance equation and high temperature tracking equation. The soil water balance equation is the physical law that controls precipitation and evapotranspiration, ensuring the water balance law on the soil surface; the high temperature tracking equation is a control equation involving continuous tracking of high temperatures. During drought, the duration of high temperatures plays a key role in the disaster losses caused by drought events.

[0041] The initial conditions are the disaster-prone environmental data of the target area, and the sea surface temperature, Southern Oscillation Index and North Atlantic Oscillation in the climate data are used as boundary conditions in the prediction model.

[0042] The above steps, using the PINN network and physical constraint functions, ensure that the values of the key disaster-causing factors conform to the physical laws of precipitation and soil changes. This not only maintains physical consistency but also improves the model's generalization ability, ensuring accurate predictions even with small data volumes. The determination of initial and boundary conditions effectively guides early warning information, such as whether the forecast year is prone to meteorological drought, improving forecast accuracy.

[0043] S130. Determine drought assessment results from different assessment dimensions based on the values of key disaster-causing factors, determine the drought level based on the drought assessment results, and issue drought warnings based on the drought level.

[0044] Specifically, drought assessment results are calculated from different dimensions based on the values of key disaster-causing factors, the drought level is determined based on the proportion of the impact of the drought assessment results on drought, and an early warning is issued for the degree of drought in the target area based on the drought level.

[0045] Furthermore, drought assessment results can be calculated from different dimensions, such as the Standard Precipitation Index (SPI), the Palmer Drought Severity Index (PDSI), and the Comprehensive Meteorological Drought Index (CI index).

[0046] The Standardized Precipitation Index (SPI) is one of the indicators that characterizes the probability of precipitation occurring during a given period. The Palmer Drought Index (PDI) indicates a water deficit, defined as a period of time when the actual water supply in a region is consistently less than the climatically suitable water supply. The meteorological drought index reflects both short-term (monthly) and long-term (seasonal) precipitation anomalies, as well as short-term water deficits that affect crops.

[0047] The above steps determine the drought assessment results of the target area from different dimensions, which can determine the disaster situation from multiple dimensions and ensure the accuracy of drought level assessment.

[0048] For example, Figure 2 As shown, the original environmental data and historical environmental data of the target area are obtained, and the original environmental data are preprocessed to obtain the target environmental data. The key disaster factors of the target area are determined based on the historical environmental data, and the key disaster factors are fed back to the prediction model. The prediction model predicts the key disaster factor values based on the input target environmental data, and determines the drought assessment results based on the key disaster factor values. The meteorological drought level is determined based on the multi-dimensional drought assessment results, and drought warnings are issued based on the meteorological drought level.

[0049] Optionally, determine the key hazard factors, including steps A1-A2:

[0050] Step A1: Determine the historical environmental data of the target area.

[0051] Among them, historical environmental data include: historical meteorological data, historical climate data, historical hydrological remote sensing data and historical disaster-prone environmental data.

[0052] Obtain the historical original environmental data of the target area, clean the acquired historical original environmental data, fill in missing values, normalize or standardize the completed historical original environmental data, and perform consistency processing of spatiotemporal resolution to obtain historical environmental data.

[0053] Step A2: Screen key disaster-causing factors based on historical environmental data.

[0054] Specifically, historical environmental data are analyzed, and the environmental data that cause drought in the target area are identified as key disaster factors.

[0055] For example, if the target area is the northern region, the key disaster-causing factors may be precipitation, temperature, river runoff, etc.; if the target area is the southern region, the key disaster-causing factors may be monsoon changes, high pressure, etc.

[0056] Optionally, the prediction model building process includes steps B1-B3:

[0057] Step B1: Determine initial conditions and boundary conditions based on historical environmental data.

[0058] Specifically, the disaster-prone environmental data in the historical environmental data are used as initial conditions, and the sea surface temperature, Southern Oscillation Index and North Atlantic Oscillation in the historical climate data are used as boundary conditions.

[0059] Step B2: Embed the physical constraint function into the neural network model, and construct the model to be trained based on the initial conditions and boundary conditions.

[0060] Furthermore, the physical constraint function is embedded into the neural network model as a loss function, the initial conditions and boundary conditions are introduced into the neural network model, and the Adam optimizer (Adaptive Moment Estimation) is used as the loss function to construct the model to be trained.

[0061] Step B3: Use historical meteorological data, historical climate data, and historical hydrological remote sensing data as inputs of the model to be trained, and historical key disaster factor values as outputs of the model to be trained. Train the model to be trained, determine deviation values based on the key disaster factor values obtained through training and the historical key disaster factor values, and optimize the weight parameters of the model to be trained based on the deviation values through an optimizer until the deviation value is less than or equal to a preset deviation value. Then, use the model to be trained corresponding to the deviation value that meets the requirements as the prediction model.

[0062] Optionally, determine the soil water balance equation, including steps C1-C2:

[0063] Step C1: Obtain historical precipitation, historical soil surface moisture, and historical normalized difference vegetation index of the target area from historical environmental data.

[0064] Specifically, historical precipitation, historical soil surface moisture and historical normalized difference vegetation index of the target area are extracted from historical environmental data through data screening.

[0065] Furthermore, data screening can be achieved through data classification algorithms or clustering algorithms.

[0066] Step C2: Construct a soil water balance equation based on the historical precipitation, historical soil surface moisture, and historical normalized difference vegetation index of the target area.

[0067] Specifically, a soil water balance equation is constructed based on the changing patterns among historical precipitation, historical soil surface moisture and historical normalized difference vegetation index in the target area.

[0068] Furthermore, the soil water balance equation can be expressed as follows:

[0069]

[0070] Where t is time; p is precipitation; s is soil surface moisture; NDVI is the normalized difference vegetation index; Z, a, b, c, and k are parameters.

[0071] In the above steps, the soil water balance equation expresses the physical laws between the changes in soil surface moisture and vegetation status as precipitation changes during drought, and has advantages in reflecting crop losses.

[0072] Optionally, determine the high temperature tracking equation, including steps D1-D2:

[0073] Step D1: Obtain the historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature of the target area from historical environmental data.

[0074] Specifically, the historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature of the target area are obtained from historical environmental data through data screening.

[0075] Step D2: construct a high temperature tracking equation based on the historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature.

[0076] Specifically, a high temperature tracking equation is constructed based on the changing patterns of historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature, combined with the gradient operator and near-surface air pressure.

[0077] Furthermore, the high temperature tracking equation can be expressed as follows:

[0078]

[0079] Where x is the location of the target area; t x is time; t g is the high temperature starting time; T' is the temperature anomaly; T is the temperature; θ is the potential temperature; v is the horizontal wind speed; is the gradient operator; is the climatological mean temperature; p is the near-surface air pressure; ω is the vertical wind speed; K and p0 are parameters.

[0080] The above steps and high temperature tracking equation can predict the start, peak and end of the high temperature process, and can be used to predict the time and degree of threat to population health and water demand in meteorological drought forecasting.

[0081] Optionally, determining drought assessment results from different assessment dimensions based on the values of key disaster factors, and determining the drought level based on the drought assessment results, includes steps E1-E3:

[0082] Step E1: Determine drought assessment results using different drought assessment models based on the values of key disaster factors.

[0083] Specifically, key disaster factors are input into different drought assessment models, and drought assessment results are obtained through calculation.

[0084] Furthermore, drought assessment models include: Standardized Precipitation Index model, Palmer Drought Index model and Meteorological Drought Comprehensive Index model.

[0085] Among them, the standardized precipitation index model is:

[0086] Assuming that the precipitation in a certain period is x, the probability density function of the Γ distribution is:

[0087]

[0088] Where β is the scale parameter, γ is the shape parameter, and x is the precipitation. β and γ can be obtained using the maximum likelihood estimation method:

[0089]

[0090]

[0091]

[0092] Among them, x i is the precipitation data sample; is the climate mean of precipitation; n is the length of the calculation sequence. The cumulative probability of a given time scale can be calculated as follows:

[0093]

[0094] Since the above formula does not include the case of x=0, and the actual precipitation can be 0, the probability of the event when the precipitation is 0 is:

[0095] F(x=0)=m / n,

[0096] Where m is the number of samples with zero precipitation, and n is the climatological mean of precipitation.

[0097] By performing normal standardization on the probability of Γ distribution and performing an approximate solution, we can obtain:

[0098]

[0099] Where, t = ln1 / F 2 When F>0.5, S=1; when F≤0.5, S=-1. c0=2.515517; c1=0.802853; c2=0.010328; d1=1.432788; d2=0.189269; d3=0.001308.

[0100] The Palmer Drought Index model is:

[0101] x i =0.805x i-1 +z i / 57.136

[0102]

[0103]

[0104] Among them, X i is the Palmer Drought Index for the i-th month; K is the weight factor; is the multi-year average possible evapotranspiration for month i; is the multi-year average soil water replenishment in month i; is the multi-year average monthly runoff in month i; is the multi-year average precipitation in month i; is the multi-year average soil water loss in month i; is the average absolute value of monthly moisture deviation.

[0105] Among them, the meteorological drought comprehensive index model is:

[0106] CI=aZ 30 +bZ 90 +cM 30 ,

[0107] Among them, Z 30 , Z 90 The standardized precipitation index values for the past 30 days and the past 90 days respectively; M 30 is the relative humidity index in the past 30 days; a, b, c are parameters.

[0108] Step E2: Determine a comprehensive drought assessment result based on the drought assessment result.

[0109] Specifically, the comprehensive drought assessment result is determined according to the impact of the drought assessment result on drought.

[0110] Step E3: Match the corresponding drought level according to the comprehensive drought assessment results.

[0111] Specifically, the corresponding drought level is matched from a preset corresponding relationship according to the comprehensive drought assessment result.

[0112] Furthermore, the preset corresponding relationship is a pre-established corresponding relationship between the comprehensive drought rating result and the drought level.

[0113] For example, it is assumed that the drought level corresponding to the comprehensive drought assessment result between 0%-A% is low risk; the drought level corresponding to the comprehensive drought assessment result between A%-B% is lower risk; the drought level corresponding to the comprehensive drought assessment result between B%-C% is higher risk; and the drought level corresponding to the comprehensive drought assessment result between C%-D% is high risk.

[0114] The above steps can comprehensively consider the impact of different drought assessment results on drought in determining the drought level, thereby improving the accuracy of drought level determination.

[0115] Optionally, determining the drought assessment result according to the drought assessment result includes steps F1-F2:

[0116] Step F1: Match the corresponding grade weight according to the drought assessment result.

[0117] Among them, the grade weight is used to represent the proportion of drought assessment results to drought impact.

[0118] According to the drought assessment results and the target area, the corresponding level weights are matched from the preset level relationships.

[0119] Furthermore, the preset level relationship is constructed by the drought assessment results, the target area and the level weight. The construction process is: the level weight is determined according to the degree of impact of the drought assessment results on the drought in the target area.

[0120] For example, assuming that for target area A, the Standardized Precipitation Index is the primary metric, then its proportion can be 50%, and the Palmer Drought Index and the Meteorological Drought Composite Index are secondary metrics, each accounting for 25%. For target area B, the Standardized Precipitation Index is the secondary metric, then its proportion can be 30%, the Palmer Drought Index is the secondary metric, then its proportion can be 20%, and the Meteorological Drought Composite Index is the primary metric, then its proportion can be 50%. Therefore, the preset hierarchical relationship can be expressed as: dict = {target area: A, SPI: 0.5, PDSI: 0.25, CI: 0.25; target area: B, SPI: 0.3, PDSI: 0.2, CI: 0.5; ...}.

[0121] Step F2: Determine a comprehensive drought assessment result based on the drought assessment result and the level weight.

[0122] Specifically, the drought assessment results and the corresponding grade weights are weighted and summed to obtain a comprehensive drought assessment grade.

[0123] The above steps, by introducing grade weights, can ensure that the drought grade is comprehensively evaluated according to the impact of different indices on drought.

[0124] The technical solution of this embodiment determines target environmental data for a target area; based on the target environmental data, determines the values of key hazard factors through a prediction model. The obtained key hazard factor values can meet the physical laws of meteorological changes, making the obtained key hazard factors more representative of the changes in future drought influencing factors in the target area; based on the key hazard factor values, a drought assessment result is determined from different assessment dimensions, the drought level is determined based on the drought assessment results, and a drought warning is issued based on the drought level. This multi-dimensional drought level assessment can ensure the accuracy of the drought level assessment. This method can ensure the physical consistency of the key hazard factor values, assess drought from different assessment dimensions based on the key hazard factor values, and comprehensively consider the drought level to improve the accuracy of drought prediction.

[0125] Figure 3 This is a schematic diagram of the structure of a meteorological drought early warning device provided by an embodiment of the present invention. This embodiment is applicable to the situation where a drought in a target area is warned based on collected environmental data. The meteorological drought early warning device can be implemented in the form of hardware and / or software. The meteorological drought early warning device can be configured in any electronic device with network communication capabilities. Figure 3 As shown, the device includes: a data determination module 210, a key disaster factor value module 220 and an early warning module 230, wherein:

[0126] Data determination module 210: used to determine target environmental data of a target area, where the target environmental data is used to characterize the changing trends of the atmosphere and geology of the target area;

[0127] Key disaster factor value module 220: used to determine the key disaster factor value based on the target environmental data through the prediction model. The key disaster factor value is the corresponding range of change when the key disaster factor changes under the target environmental data. The key disaster factor is the factor that causes drought in the target area. The prediction model is configured with a physical constraint function. The physical constraint function serves as a loss function to constrain the physical laws represented by the prediction results of the prediction model. The physical constraint function includes: a soil water balance equation and a high temperature tracking equation. The soil water balance equation is used to represent the relationship between precipitation and soil surface moisture and vegetation status. The high temperature tracking equation is used to track temperature changes.

[0128] Early warning module 230: used to determine drought assessment results from different assessment dimensions based on the values of key disaster factors, determine the drought level based on the drought assessment results, and issue drought early warning based on the drought level.

[0129] Optionally, the key disaster factor value module 220 includes:

[0130] Historical environmental data determination unit: used to determine the historical environmental data of the target area, including historical meteorological data, historical climate data, historical hydrological remote sensing data and historical disaster-prone environmental data;

[0131] Key disaster factor determination unit: used to screen key disaster factors based on historical environmental data.

[0132] Optionally, the key disaster factor value module 220 includes:

[0133] Condition determination unit: used to determine initial conditions and boundary conditions based on historical environmental data;

[0134] The model to be trained is determined by the unit: it is used to embed the physical constraint function into the neural network model and construct the model to be trained according to the initial conditions and boundary conditions;

[0135] Model training unit: used to take historical meteorological data, historical climate data and historical hydrological remote sensing data as the input of the model to be trained, and the historical key disaster factor values as the output of the model to be trained, to train the model to be trained, and to determine the deviation value based on the key disaster factor values obtained through training and the historical key disaster factor values. The optimizer is used to optimize the weight parameters of the model to be trained based on the deviation value until the deviation value is less than or equal to the preset deviation value, and the model to be trained corresponding to the deviation value that meets the requirements is used as the prediction model.

[0136] Optionally, the key disaster factor value module 220 includes:

[0137] Precipitation data acquisition unit: used to obtain historical precipitation, historical soil surface moisture and historical normalized difference vegetation index of the target area from historical environmental data;

[0138] Soil water balance equation construction unit: used to construct the soil water balance equation based on the historical precipitation, historical soil surface moisture and historical normalized difference vegetation index of the target area.

[0139] Optionally, the key disaster factor value module 220 includes:

[0140] Temperature data acquisition unit: used to obtain the historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature of the target area from historical environmental data;

[0141] High temperature tracking equation construction unit: used to construct a high temperature tracking equation based on historical climate average temperature, historical potential temperature, historical horizontal wind speed, historical vertical wind speed and historical abnormal temperature.

[0142] Optionally, the early warning module 230 includes:

[0143] Drought assessment result determination unit: used to determine drought assessment results through different drought assessment models based on the values of key disaster factors;

[0144] Comprehensive drought assessment result determination unit: used for determining the comprehensive drought assessment result according to the drought assessment result;

[0145] Drought level determination unit: used to match the corresponding drought level according to the comprehensive drought assessment results.

[0146] Optional, drought assessment results determination unit, including:

[0147] Level weight determination subunit: used to match the corresponding level weight according to the drought assessment results. The level weight is used to represent the proportion of the drought assessment results to the drought impact;

[0148] Drought assessment result determination subunit: used to determine the comprehensive drought assessment result based on the drought assessment results and level weights.

[0149] The meteorological drought warning device provided in the embodiment of the present invention can execute the meteorological drought warning method provided in any embodiment of the present invention, and has the corresponding functions and beneficial effects of executing the meteorological drought warning method. For detailed process, please refer to the relevant operations of the meteorological drought warning method in the above embodiment.

[0150] Figure 4Schematic diagram of the structure of an electronic device for implementing the meteorological drought early warning method of an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0151] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0152] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0153] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the meteorological drought early warning method.

[0154] In some embodiments, the meteorological drought early warning method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the meteorological drought early warning method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the meteorological drought early warning method via any other suitable means (e.g., via firmware).

[0155] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0156] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0157] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0158] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0159] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0160] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0161] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0162] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A meteorological drought early warning method, characterized in that: include: Determining target environmental data for a target area, wherein the target environmental data is used to characterize the changing trends of the atmosphere and geology of the target area; The prediction model determines the value of the key disaster factor according to the target environmental data, wherein the value of the key disaster factor is the range of change corresponding to the change of the key disaster factor under the target environmental data. The key disaster factor is a factor that causes drought in the target area. The prediction model uses the disaster-prone environmental data in the historical environmental data as the initial conditions and the sea surface temperature, the Southern Oscillation Index and the North Atlantic Oscillation in the historical climate data as the boundary conditions; embeds the physical constraint function into the PINN neural network model as the loss function, introduces the initial conditions and the boundary conditions into the PINN neural network model, and constructs the model to be trained by the optimizer as the minimization loss function; uses the historical meteorological data, the historical climate data and the historical hydrological remote sensing data as the input of the model to be trained, and uses the historical key disaster factor values as the output of the model to be trained, trains the model to be trained, determines the deviation value based on the key disaster factor values obtained by training and the historical key disaster factor values, and optimizes the weight parameters of the model to be trained according to the deviation value by the optimizer until the deviation value is less than or equal to the preset deviation value; the physical constraint function includes: a soil water balance equation and a high temperature tracking equation; The soil water balance equation is expressed as: ， Where, t is time; p is precipitation; s is soil surface moisture; NDVI is the normalized difference vegetation index; Z, a, b, c, k are parameters; The high temperature tracking equation is expressed as: ； Where x is the location of the target area; t x is time; t g is the high temperature starting time; T' is the temperature anomaly; T is the temperature; is the potential temperature; is the horizontal wind speed; is the gradient operator; is the climatological mean temperature; p is the surface pressure; is the vertical wind speed; K, p0 are parameters; Drought assessment results are determined from different assessment dimensions based on the values of the key disaster-causing factors, drought levels are determined based on the drought assessment results, and drought warnings are issued based on the drought levels.

2. The method according to claim 1, characterized in that Determine the key hazard factors, including: Determine the historical environmental data of the target area, including historical meteorological data, historical climate data, historical hydrological remote sensing data, and historical disaster-prone environmental data; Screen key disaster-causing factors based on the historical environmental data.

3. The method according to claim 1, characterized in that Determining drought assessment results from different assessment dimensions based on the values of the key disaster factors, and determining the drought level based on the drought assessment results, includes: Determine the drought assessment result according to the values of the key disaster factors through different drought assessment models; determining a comprehensive drought assessment result based on the drought assessment result; The corresponding drought level is matched according to the comprehensive drought assessment result.

4. The method according to claim 3, characterized in that Determining a comprehensive drought assessment result based on the drought assessment result includes: Matching corresponding grade weights according to the drought assessment results, wherein the grade weights are used to represent the proportion of the drought assessment results to the drought impact; A comprehensive drought assessment result is determined according to the drought assessment result and the grade weight.

5. A meteorological drought early warning device, characterized in that: include: A data determination module is used to determine target environmental data of a target area, wherein the target environmental data is used to characterize the changing trend of the atmosphere and geology of the target area; The key disaster factor value module is used to determine the key disaster factor value through the prediction model according to the target environmental data. The key disaster factor value is the corresponding range of change when the key disaster factor changes under the target environmental data. The key disaster factor is the factor that causes drought in the target area. The prediction model uses the disaster-prone environmental data in the historical environmental data as the initial condition, and the sea surface temperature, Southern Oscillation Index and North Atlantic Oscillation in the historical climate data as the boundary condition; the physical constraint function is embedded in the PINN neural network model as the loss function, and the initial condition and the boundary condition are introduced into the PINN neural network model. A neural network model is constructed by using an optimizer as a loss function to minimize the loss function; historical meteorological data, the historical climate data, and historical hydrological remote sensing data are used as inputs of the model to be trained, and historical key disaster factor values are used as outputs of the model to be trained; the model to be trained is trained, and a deviation value is determined based on the key disaster factor values obtained through training and the historical key disaster factor values; the weight parameters of the model to be trained are optimized by the optimizer based on the deviation value until the deviation value is less than or equal to a preset deviation value; the physical constraint function includes: a soil water balance equation and a high temperature tracking equation; The soil water balance equation is expressed as: ， Where, t is time; p is precipitation; s is soil surface moisture; NDVI is the normalized difference vegetation index; Z, a, b, c, k are parameters; The high temperature tracking equation is expressed as: ； Where x is the location of the target area; t x is time; t g is the high temperature starting time; T' is the temperature anomaly; T is the temperature; is the potential temperature; is the horizontal wind speed; is the gradient operator; is the climatological mean temperature; p is the surface pressure; is the vertical wind speed; K, p0 are parameters; The early warning module is used to determine drought assessment results from different assessment dimensions according to the values of the key disaster-causing factors, determine the drought level according to the drought assessment results, and issue drought early warning according to the drought level.

6. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the meteorological drought early warning method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the meteorological drought early warning method according to any one of claims 1 to 4 when executed.

Images