244 results about "Numerical weather forecast" patented technology

A double-path ultra-short-term wind power prediction method based on numerical weather forecast and multi-order time sequence dynamic gating fusion comprises the following steps: acquiring wind power generation historical data and numerical weather forecast data of a wind power plant, and screening weather factors highly related to wind power by using an MIC; the CEEMDAN is adopted to decompose the power sequence into a plurality of intrinsic mode functions (IMF); a dual-path prediction architecture is constructed, one path adopts xLSTM to predict an intrinsic mode function (IMF), all subsequences are superposed, and a prediction result is obtained; in the other path, the XGBoost is combined with key meteorological characteristics of an intrinsic mode function (IMF) and a numerical weather forecast (NWP) for prediction, and all the subsequences are superposed to obtain a prediction result; the method comprises the following steps: designing an MT-DGFusion module through an enhanced attention and dynamic gating network; and fusing the dual-path prediction results through an MT-DGFusion module to obtain a final prediction result. According to the method, double breakthrough of prediction precision and stability is realized, and a new technical path is provided for a complex time sequence prediction task.

The invention discloses a risk scheduling method for a water-wind-solar complementary system, and the method comprises the steps: collecting historical data and power grid topological parameters, accessing global and regional numerical weather forecast data, employing a coupling model, fusing meteorological grid data with a historical power station output sequence, and generating hourly reservoir incoming water amount and wind-solar power probability prediction results. According to the predicted time sequence and the generated scene, calculating the scene probability based on the generated scene; according to the prediction time sequence, using a quantification method to obtain a peak regulation risk quantification value; calculating power grid power flow distribution and critical clearing time according to the generation scene and the power grid topological parameters, and calculating a system stability margin; a multi-target optimization model is constructed according to a peak regulation risk quantized value after splitting and a system stability margin, the system stability margin is introduced as an optimization target, the overall stability of the system is improved, a meteorological-hydrological-output three-mode feature mapping method is provided, and meteorological feature extraction of a key grid region is enhanced through an attention mechanism.

The invention discloses a photovoltaic power generation power prediction method and system based on a large language model. The method comprises the following steps: converting historical power data and numerical weather forecast data into time sequence embedded representation; through cross-modal semantic alignment, semantic embedding representation is generated; constructing a natural language prompt containing task context information, encoding the natural language prompt into prompt embedding, combining prompt embedding with semantic embedding representation to form a fusion input sequence, inputting the fusion input sequence into a pre-trained large language model, and outputting implicit features; synchronously generating an initial power prediction result and a weather prediction result obtained by correcting the numerical weather prediction data through a parallel collaborative prediction mechanism; and taking the meteorological prediction result as a correction signal, performing joint optimization on the preliminary power prediction result, and outputting a power generation power prediction value. According to the method, the problem of deep fusion of heterogeneous data is effectively solved, and the prediction accuracy is improved.

The invention relates to a photovoltaic power grid energy storage optimization regulation and control method based on multi-scale prediction. The method comprises the following steps: A1, obtaining historical power generation data, real-time meteorological data and numerical weather forecast of a photovoltaic power station; a2, generating a multi-time-scale photovoltaic output prediction sequence; a3, establishing an energy storage dynamic model of charge and discharge efficiency, capacity attenuation and operation constraint; a4, generating an energy storage charging and discharging demand curve under different time scales; a5, constructing a multi-time scale coupled optimization model by taking power grid operation cost minimization and renewable energy consumption maximization as targets; a6, updating an energy storage scheduling instruction in a rolling manner based on latest prediction data by adopting a model prediction control framework; a7, monitoring the deviation between the actual photovoltaic output and the power grid load, and dynamically adjusting the energy storage charging and discharging power; and A8, correcting a prediction error through Kalman filtering and a closed-loop feedback mechanism. According to the invention, high-efficiency operation can be realized, and power grid cost minimization and renewable energy consumption maximization can be realized.

The invention belongs to the technical field of wind power prediction, and particularly relates to a wind power prediction method and system based on multi-source data fusion. According to the method, a cascade prediction architecture composed of a wind speed prediction model and a power prediction model is constructed, and the wind speed prediction module extracts cross-variable association features of multi-dimensional meteorological parameters from numerical weather forecast data by adopting a variable-level attention mechanism and a patch-level attention mechanism; capturing a long and short time dependence mode from the actually measured historical operation data; the power prediction model decomposes a predicted wind speed sequence through one-dimensional average pooling, and constructs a nonlinear power mapping model. Besides, an attention fusion module is provided, a learnable global variable is introduced as an information exchange bridge, information interaction between meteorological data and actually measured data is realized, and the collaborative utilization efficiency of multi-source data is effectively improved. Experiments show that the method is superior to an existing prediction model based on deep learning.

The invention discloses a high-precision distributed photovoltaic output prediction method and system based on micrometeorology and feature hierarchical clustering, and relates to the technical field of distributed photovoltaic output prediction.The method comprises the steps that numerical weather forecast data are collected, an initial micrometeorological field is generated through space-time alignment and self-adaptive KNN interpolation, and the initial micrometeorological field is subjected to feature clustering; a WRF-LES system and a bidirectional LSTM are combined to establish cross-scale mapping, a dynamic residual correction field is fused to generate hectometer-level high-resolution micrometeorological data, and the problem of insufficient resolution of traditional numerical forecasting is solved. MIC and PA-DTW are used for jointly analyzing the characteristics of the power station, and dynamic clustering is achieved through a sliding time window and incremental spectral clustering. According to the method, a physical information graph network and causal expansion convolution are coupled to extract features, federal learning cross-power-station cooperative training is combined, the distributed photovoltaic output prediction precision and robustness are improved, and privacy security is considered.

The invention discloses a drainage basin intelligent flood control scheduling method and system based on digital twinborn, and relates to the technical field of flood control and disaster mitigation, and the method comprises the steps: collecting static data and dynamic data of a drainage basin, building a hydrological and hydrodynamic coupling model based on the static data and the dynamic data, and forming a drainage basin digital twinborn body; inputting the received numerical weather forecast into the digital twin of the watershed for simulation, generating a plurality of flood routing scenes in a future time period, and calculating a dynamic flood risk probability graph; the method comprises the following steps: constructing a simulation training environment by using historical flood data and a high-precision drainage basin digital twinborn body, carrying out offline training on a scheduling strategy network in the simulation training environment based on a reinforcement learning algorithm, and outputting a scheduling instruction according to a real-time drainage basin state to complete training of the scheduling strategy network. According to the method, the core problem that the traditional method is insufficient in decision timeliness and weak in adaptive capacity in an uncertain environment is effectively solved.

The invention relates to the technical field of photovoltaic prediction, in particular to a distributed photovoltaic power prediction method and system based on high-dimensional gridding numerical weather forecast, and the method comprises the steps: carrying out the standardization of the numerical weather forecast data and photovoltaic power historical data of a target region, and achieving the time-space alignment based on a preset grid, generating a gridding data set; utilizing convolution processing to extract local space features, and converting and fusing the local space features into a feature sequence containing space and historical time sequence information at the same time; modeling is carried out through an encoder-decoder architecture, an encoder excavates historical power dependence, and a decoder dynamically couples future meteorological characteristics with historical power through an attention mechanism and outputs a grid-level predicted value; aggregating to obtain a system total power prediction result; by establishing a unified space-time grid, refined alignment of data is realized, cross-space-time dynamic fusion is performed in combination with convolution and an attention mechanism, and prediction precision and stability can be kept in complex weather.

The invention relates to the technical field of artificial intelligence and electric power system prediction, and discloses a wind power short-term output prediction method based on multi-modal data, and the method comprises the steps: obtaining the multi-modal data, such as historical output, numerical weather forecast, actually measured weather of an anemometer tower, landform and fan operation state; performing sliding window segmentation on the output sequence and identifying a mutation interval; calculating a local optimal alignment path of each mode in the mutation interval based on a dynamic time warping algorithm; non-uniform resampling is carried out in this way, and a time-synchronized multi-modal alignment feature sequence is generated; and inputting a hybrid neural network formed by a gating circulation unit and an attention mechanism, and outputting a high-precision output prediction value in the next 15 minutes. The system comprises corresponding function modules. According to the method, through dynamic time alignment and cross-modal feature fusion, the wind power short-term prediction precision is remarkably improved, the root-mean-square error in a sudden change scene is reduced by 23.7%, and reliable support is provided for power grid dispatching.

The invention discloses a multi-drive process multi-factor agricultural non-point source pollution prediction method based on coupling meteorological numerical forecasting. The method comprises the following steps: firstly, introducing numerical weather forecast data, and constructing a high-precision weather driving field with kilometer-level space and hour-level time resolution by combining WRF dynamic downscaling, DEM terrain correction and conservation resampling; secondly, establishing a multi-drive process coupling model system which comprises a meteorological drive layer, a hydrological response layer, a pollutant migration layer and a crop feedback layer and is used for simulating runoff production, sediment erosion, nitrogen and phosphorus migration and transformation and crop transpiration and root nutrient absorption processes; thirdly, performing precision verification on a simulation result by using observation data, and identifying key meteorological and hydrological factors through an error transfer matrix and a sensitivity analysis method; and finally, realizing parameter adaptive correction by adopting a long short-term memory network, finishing parameter optimization in combination with a multi-target genetic algorithm, packaging the model chain through a containerization technology, and realizing cross-platform deployment and visual output of a pollution load result. The method can be used for agricultural non-point source pollution prediction and management.

The invention provides a multi-scale photovoltaic power prediction method and system fusing numerical weather forecast and a physical information neural network, relates to the technical field of photovoltaic power prediction, and realizes rolling prediction of future meteorological parameters and horizontal radiation intensity through numerical weather forecast. The authenticity and the stability of the meteorological data are effectively improved by combining quantile mapping and cascade correction of the time sequence mode attention neural network; and through combined modeling of a physical information neural network and a time sequence characteristic network, meteorological and radiation data subjected to multi-level correction and conversion are deeply fused with a photovoltaic system physical law, and high-precision prediction of photovoltaic power is realized. The whole prediction process has automatic anomaly elimination and deletion complementation capabilities, the prediction robustness under complex meteorological conditions and extreme environments is enhanced, and the sequential response and physical consistency of photovoltaic power are optimized, so that the engineering applicability and intelligent level of the system are greatly improved.

The RIME-RF spatial downscaling-based photovoltaic output prediction method comprises the steps of collecting photovoltaic power data and local meteorological observation LMD data of a photovoltaic power station in a target area, extracting common data of numerical weather forecast NWP data and the local meteorological observation LMD data, and constructing an input feature set; the method comprises the following steps: optimizing hyper-parameters of a random forest (RF) algorithm based on a frost ice optimization (RIME) algorithm, constructing an RIME-RF model, and performing spatial downscaling on numerical weather forecast NWP data; a VMD-CNN-GRU-SE attention mechanism photovoltaic power prediction model optimized based on BKA is adopted, original numerical weather forecast NWP data is combined with photovoltaic power data to train the prediction model, and numerical weather forecast NWP data after spatial downscaling is combined with the photovoltaic power data to train the prediction model. According to the prediction method, changes of fine meteorological factors influencing the photovoltaic power can be more accurately captured, downscaling errors are remarkably reduced, and short-term power prediction precision is improved.

The invention provides a wind field inversion method and device based on multi-source prior data, equipment and a medium, and the method comprises the steps: building a target function of a three-dimensional fusion wind field based on the multi-source prior data and a fluid mechanics model simulation wind field, carrying out the iterative optimization of the target function, and determining the three-dimensional fusion wind field; constructing an initial deep learning neural network model, taking the three-dimensional fusion wind field as a truth value label, inputting the urban underlying surface features, the terrain elevation and the numerical weather forecast wind field into the deep learning neural network model, and training to obtain a target deep learning neural network model; and inputting the new numerical weather forecast wind field, the urban underlying surface features and the terrain elevation into the target deep learning neural network model, and outputting a refined wind field. According to the technical scheme provided by the embodiment of the invention, through the trained deep learning neural network model, high-precision rapid inversion of the low-altitude wind field is realized without depending on laser radar and ground observation data and depending on prior information such as numerical prediction and terrain.

The invention relates to the technical field of power systems and automation thereof, in particular to a generative adversarial network-based source-load joint scene generation method, which comprises the following steps of: constructing a multi-source time sequence database and extracting weather, time, space and historical state driving factors; establishing a joint probability distribution model based on a vine connection function; taking a numerical weather forecast path and a date type as conditional input, constructing a generative adversarial network embedded with a physical constraint microloss function of the power system, and forming a physical information generator; performing dependent structure fidelity verification on the generated scene by using the joint probability distribution model; generator parameters are fixed, potential space vectors are optimized through a gradient ascending method to maximize power grid risk indexes, and a high-risk source-load joint scene set is generated. According to the technical scheme, accurate generation of the source-load joint scene which is physically feasible and reasonable in statistics and focuses on the high-risk working condition is realized, and the safe operation toughness and the risk early warning capability of the novel power system are remarkably improved.

The invention discloses a numerical weather forecast deviation correction method and system based on physical conservation optimization. The method comprises the following steps: acquiring a specified prognostic variable in a numerical weather forecast system; inputting the prognostic variable into a prediction deviation correction model pre-trained by a loss function based on physical conservation optimization so as to obtain a deviation correction prediction result of numerical weather prediction; in the loss function of the physical conservation optimization, the physical conservation optimization refers to adding part or all of Coriolis force and barometric gradient force loss, static equation constraint derivation loss and vertical integral loss to the loss function of the value weather forecast deviation correction model. The invention aims to solve the problem of systematic deviation correction in mid-term numerical forecasting, better capture the complex dependency relationship in meteorological data, quantify the uncertainty of numerical weather forecasting and improve the forecasting precision.

The invention provides a micrometeorological prediction method and system, and belongs to the technical field of meteorological prediction. The method comprises the following steps: acquiring data of an unmanned aerial vehicle sensor, a ground meteorological station, satellite remote sensing and numerical weather forecast, and respectively constructing feature vectors of data sources; performing weighted average fusion on the multi-source feature vectors based on an attention mechanism at a target space-time position to obtain a fusion feature matrix; and training a target neural network by using the fusion feature matrix of the plurality of space-time positions, and taking the trained target neural network as a micro-meteorological prediction model to realize high-precision prediction of future micro-meteorological elements. According to the method, multi-source heterogeneous data and a deep learning technology are fused, the temporal-spatial resolution and accuracy of micrometeorological prediction are effectively improved, and the method is particularly suitable for unmanned aerial vehicle flight safety early warning and route dynamic optimization.

The invention relates to the technical field of data assimilation processing, in particular to a WRFDA assimilation method, system and device based on GIIRS data. The method comprises the following steps: acquiring GIIRS data, extracting data from the GIIRS data, and merging and checking the data to obtain checked GIIRS observation information; performing deviation correction on the background field data of the numerical weather forecasting mode by using an RTTOV rapid radiation transmission mode to obtain simulation data after stable deviation correction; performing cloud pollution screening on the GIIRS observation information after inspection and the simulation data after deviation correction to obtain GIIRS observation information without cloud pollution; wRFDA numerical assimilation is carried out on the GIIRS observation information without cloud pollution, an initial field file is obtained, and the initial field file is used for adjusting and optimizing an output weather forecast result. According to the method, a more accurate initialization file is obtained, and the simulation accuracy of the typhoon precipitation area is more accurately improved.

The invention discloses a short-term photovoltaic power prediction device and method based on multi-mode collaborative attention. The method comprises the steps of obtaining historical photovoltaic power, numerical weather forecast and satellite image data and performing preprocessing; constructing a short-term photovoltaic power prediction model comprising an image feature extraction module, a time sequence data feature extraction module, a multi-modal fusion module and a time sequence prediction module, and training the model by using the preprocessed data; the method comprises the following steps: extracting image features from satellite image data through an image feature extraction module; a time sequence data feature extraction module captures time sequence features from the historical power generation power and the numerical weather forecast data; the image features and the time sequence features are fused through a multi-modal fusion module, and finally a final power prediction value is output through a time sequence prediction module; and the loss between the power prediction value and the real value is minimized. According to the method, key complementary information influencing photovoltaic output can be effectively captured, so that the precision and robustness of short-term power prediction are remarkably improved.

The invention discloses a transformer area distribution transformer reverse heavy overload prediction method and system, and the method comprises the steps: obtaining data which comprises the historical electrical quantity characteristics of a photovoltaic user, the numerical weather forecast characteristics of the geographic position of a transformer area, and a prior graph structure formed according to the topological graph of the transformer area; preprocessing the data to construct a data set; the method comprises the steps of establishing a space-time diagram prediction model, inputting historical electrical quantity characteristics and numerical weather forecast characteristics of N photovoltaic users and a prior diagram structure into the model to obtain a future K-step area distribution transformer load prediction result, performing heavy overload judgment based on a prediction value, and finally obtaining a future K-step area distribution transformer reverse heavy overload prediction result. According to the method, the topological structure of the power distribution network, the electrical characteristic data and the numerical weather forecast characteristics are comprehensively utilized, high-precision prediction of the multi-time-step power of the photovoltaic grid-connected system is achieved through the space-time diagram neural network model, and the method is suitable for application scenes such as operation optimization and early warning management of the power distribution network with distributed photovoltaic access.

The invention relates to the technical field of aerospace meteorology, and particularly discloses a domestic numerical weather forecast mode-based evaporation waveguide height determination method, which comprises the following steps of: obtaining meteorological coupling data output by a domestic numerical weather forecast mode CMA-MESO; determining parameters according to the meteorological coupling data and a turbulent flux algorithm COARE; constructing a temperature vertical profile and a specific humidity vertical profile according to the parameters, the Monin-Obchhoff similarity theory and an NPS model; constructing a modified refractive index profile according to the temperature vertical profile and the specific humidity vertical profile; and determining the height of the evaporation waveguide according to the height corresponding to the lowest point of the corrected refractive index profile. The method is suitable for determining the evaporation waveguide height in a domestic business meteorological mode.

The invention discloses a distributed ground-based radar atmospheric refraction error calculation method for celestial body observation. According to the method, a local numerical weather forecast model is corrected by using real-time surface meteorological parameters, atmospheric refractive index profile data containing elevation meteorological information is obtained, a geometric elevation angle of a celestial body target is taken as a standard, and a neutral atmospheric refractive index error is searched in combination with a corrected data source and a ray tracing method, so that the atmospheric refractive index is obtained. And finally, neutral atmospheric refraction error calculation and correction are realized. The effectiveness of the algorithm is verified through simulation.

The invention discloses a wind power prediction error decoupling evaluation method considering unit faults and real-time capacity changes. The method comprises the following steps: combining an actual wind power plant prediction process, and dividing wind power prediction into three links of numerical weather prediction, wind-electricity model conversion and power correction; for a power correction link, a unit fault prediction model is constructed, capacity reduction caused by faults is pre-judged in advance, a health index and a capacity attenuation coefficient are combined, a fault unit is removed in real time, the available capacity of the unit is dynamically corrected, and the equivalent actual capacity is calculated according to the health index in a weighted mode; finally, the equivalent actual capacity of the remaining unit is introduced into an error decoupling model, and quantitative evaluation is conducted on prediction errors caused by all links. And determining the proportion of the prediction error caused by each link, and determining the wind power error source after considering the unit fault and the capacity attenuation. The method can realize accurate decoupling of the wind power prediction error, and is suitable for wind power plant power prediction scenes with frequent meteorological sudden change and equipment aging.

The invention belongs to the technical field of power generation power prediction, and particularly relates to a photovoltaic power prediction method based on multi-source data fusion and deep learning, and the method comprises the steps: firstly obtaining the historical power, satellite irradiance and numerical weather forecast data of a photovoltaic power station; performing preprocessing such as space-time alignment, missing value processing and abnormal value elimination, and constructing periodic time features; training a deep learning model taking a long short-term memory network as a core by using the processed data so as to capture a complex nonlinear time sequence relationship between the weather and the power; inputting the satellite and forecast data in a to-be-predicted time period into the model, and outputting a future multi-step power predicted value; and finally, carrying out online deviation correction and physical amplitude limiting post-processing to obtain a final prediction result. By effectively fusing multi-source data and deep learning, the precision and practicability of short-term photovoltaic power prediction are remarkably improved.

The invention provides a wind power prediction method and system based on a multivariate combination prediction model and a medium, and relates to the technical field of machine learning, and the method comprises the steps: obtaining a historical wind power data time sequence and a historical numerical weather forecast data time sequence of a wind power plant; analyzing a nonlinear dependency relationship by using correlation, and optimally training a multi-modal parallel time sequence prediction model by using a variational mode decomposition algorithm in combination with a sliding window length; and based on the final sliding window length, constructing an input sample, inputting the input sample into a multi-mode parallel time sequence prediction model, outputting future prediction values of a plurality of intrinsic mode function components, and carrying out summation to obtain a wind power prediction result. According to the method and the device, the technical problem of insufficient wind power prediction precision caused by strong volatility and nonlinearity of a wind power sequence and limited learning ability of a single prediction model in the prior art can be solved, and the wind power prediction precision is improved by combining variational mode decomposition with a machine learning model for prediction.

The invention relates to the technical field of numerical weather forecast, in particular to a meteorological large model prediction method based on a data correction model, which comprises the following steps of: firstly acquiring multi-source atmospheric observation, screening observation by topology-optimal transmission quantum annealing, and constructing a weighted error covariance; applying mass, energy and earth rotation gradient, and generating a conservation assimilation field through diffusion implicit sampling; calculating a mutual information mask and coupling a cloud top optical flow fine tuning phase; cloud motion consistent field pulse codes are sent to the symplectic decomposition pulse neural network for neural form hardware reasoning, a pulse threshold is adjusted in a closed loop to control energy drift, and an uncertainty field is output through parallel disturbance reasoning. The method has the advantages of high resolution, low power consumption and probability prediction capability, and the extreme weather path and intensity prediction precision is obviously improved.

The present application relates to the technical field of aviation meteorological support, and provides a meteorological display and alarm method, system, device and medium for aircraft navigation, which comprises receiving an aircraft route plan and collecting numerical weather forecasts of each place at the takeoff time of the aircraft; extracting wind speed, temperature, relative humidity, liquid water content and cloud droplet particle number density, and calculating the horizontal visibility, aircraft icing index, total cloud cover and aircraft turbulence index at the forecast time through adaptive attenuation interpolation based on feature perception; constructing an information box and a floating box of the route points on the aircraft route map; and performing meteorological display and alarm for aircraft navigation through the information box and the floating box according to the aircraft route plan. The present application uses regional meteorological forecast results to form meteorological forecast results for flight routes, can intuitively display the comprehensive meteorology on the route and form an alarm prompt, reduces the risk of information omission, increases the safety margin, and reduces the meteorological information processing workload of flight support personnel.

The invention discloses an end-to-end scheduling method and system for a multi-source weather-driven urban integrated energy system. The method comprises the following steps: constructing an integrated energy system model; constructing a source load prediction model based on multi-source numerical weather forecast data in combination with historical photovoltaic output data and power load and thermal load data; the method comprises the following steps: establishing a comprehensive energy system optimization scheduling model with minimization of system operation cost as an optimization target, designing a differentiable optimization layer, and reversely transmitting the gradient of the optimization target in the scheduling model to a source load prediction model parameter to the source load prediction model through a back propagation algorithm by the differentiable optimization layer, the parameters of the driving source load prediction model are updated, and end-to-end linkage optimization from prediction to scheduling is achieved; and periodically obtaining updated multi-source numerical weather forecast data and source load data, readjusting prediction model parameters, and executing optimization solution of the integrated energy system optimization scheduling model. According to the invention, cooperative training and iterative optimization of the prediction model and the scheduling decision are realized.

The invention discloses a coastal wind turbine group generation power multi-scale space-time prediction method, which mainly comprises the following steps: carrying out space interpolation and error correction on a target wind turbine position by adopting a numerical weather forecast statistical downscaling technology, generating a high-resolution wind speed prediction sequence covering a short term and a long term, aligning and splicing the predicted wind speed, the field actually-measured wind speed, the environment and the unit operation variables into node dynamic input characteristics; the method comprises the following steps: encoding longitude and latitude and time sequence monitoring data of N fans of a coastal fan group into graph nodes, determining an edge weight according to geographic distance and wake flow coupling, and forming a fan graph network containing static and dynamic characteristics; and inputting the static and dynamic feature sequences into a graph neural network comprising a space attention layer, a time recursion layer and a physical constraint regular term, completing model training, and outputting the generated power of each fan in a plurality of time steps in the future and the total power predicted value of the fan group. The method can provide powerful support for wind power plant operation scheduling, power grid-connected management and new energy consumption.

The invention relates to the technical field of meteorological disaster monitoring and early warning and numerical forecasting fusion, in particular to an extreme gale weather forecasting method based on a weather forecasting large model, which comprises the following steps: acquiring a numerical weather forecasting three-dimensional physical field, a weather radar, a weather satellite and ground meteorological observation and assimilating the numerical weather forecasting three-dimensional physical field into a consensus field; time advances, sinking potential energy and cold pool diagnosis are obtained through a micro-downburst calculation program, coarse-resolution gust is formed through boundary layer similarity mapping and a hysteresis kernel, and a fine-scale base map is generated under cold pool frontal surface limitation through optimal transmission of advection constraint; the conditional diffusion probability generation model outputs pixel gust distribution and a threshold exceeding probability and extracts a wind damage polygon; radar, satellites, lightning and ground gust are fused, monotonous normalized flow calibration is used, parameters are updated through intersection-to-parallel ratio and shape-preserving coverage inspection and online amplitude limitation, and the credibility of spatial positioning, occurrence time and probability description is improved.

The invention provides an atmospheric humidity profile inversion method based on a Unet3 + neural network. The atmospheric humidity profile inversion method comprises the following steps: constructing an atmospheric humidity profile inversion model based on Unet3 + MHS microwave data; performing radiometric calibration, projection conversion and quality control on the multi-channel brightness temperature observed by the original satellite; the method comprises the following steps: selecting long-time scale radiation brightness temperature data in a Chinese area range and specific humidity data of ERA-5 corresponding to time and space to make a data set; and performing humidity profile inversion according to the humidity profile inversion model of the brightness temperature data constructed in the step S1. The method has the advantages that the Unet3 + model in deep learning and microwave sensor data which are not affected by time and weather are combined, and the problems that the space distribution of the atmospheric humidity profile is small, and the time resolution is low can be further solved; the method provides atmospheric condition data with higher temporal and spatial resolution for numerical weather forecast research, and has good application value.