59 results about "Meteorological reanalysis" patented technology

The invention discloses a high-resolution solar radiation data correction method and system based on deep learning, and the method comprises the steps: obtaining meteorological data and digital elevation model data, and carrying out the preprocessing; dynamically optimizing parameter configuration of the WRF model, and simulating and outputting high-temporal-spatial-resolution meteorological data as training data; fusing solar radiation flux values in meteorological reanalysis data and near-real-time product meteorological data as label data of a correction model; inputting the high-temporal-spatial-resolution meteorological data and the label data into a pre-constructed multi-scale feature fused deep residual coding and decoding network model for training, and obtaining an optimal parameter model based on multiple loop iterations; and correcting solar radiation flux data output by the WRF mode by using the optimal parameter model. According to the method, the problem that the radiation variable value output by the WRF in the traditional numerical weather mode is consistent with the real data trend but deviates is effectively solved, and the reliability of meteorological data simulation is enhanced.

The invention discloses a troposphere delay mapping model construction method and system based on machine learning. The method comprises the following steps: acquiring meteorological reanalysis data; calculating the ZTD at the global site or grid and the STD at different azimuth angles and elevation angles based on the meteorological reanalysis data; based on the ZTD and the STD, constructing an asymmetric mapping relation from the ZTD to the oblique path delay; constructing a data set based on meteorological reanalysis data, the ZTD and the asymmetric mapping relation, inputting the data set into a troposphere delay mapping model to train the troposphere delay mapping model, and generating prior troposphere delay for GNSS high-precision positioning through the trained troposphere delay mapping model; wherein the troposphere delay mapping model is a machine learning model. According to the method, the empirical model of the ZTD and the unified asymmetric mapping relation from the ZTD to the STD can be constructed, and the precision and the applicability of the model are remarkably improved.

The invention relates to an inshore water quality day-by-day kilometer-level inversion method, system, device and medium, and belongs to the technical field of water quality monitoring, the method comprises the following steps: obtaining remote sensing image data, meteorological reanalysis data and ground water quality site measured data, preprocessing, carrying out space-time matching fusion, using the fusion data as a sample to carry out model training, and obtaining a model training model; obtaining a space model; a spatial model offshore water quality spatial distribution diagram; and constructing a CNN-BiLSTM mixed time sequence model, and training the CNN-BiLSTM mixed time sequence model by taking historical water quality data of a ground water quality station and the offshore water quality spatial distribution map as samples to obtain a prediction model. The method is suitable for red tide early warning, pollution tracing and marine ecological management, breaks through monthly monitoring limitation, and realizes day-by-day kilometer-level water quality dynamic tracking; remote sensing large-range coverage and site high-precision data are fused, and the problem of inshore environment heterogeneity is solved; the time-space double-model cascade architecture significantly improves the generalization ability under complex hydrological conditions.

The invention provides a wide-area distributed photovoltaic power generation efficiency evaluation method, and relates to the technical field of photovoltaic power generation efficiency evaluation, and the method comprises the steps: obtaining the reanalysis time series data of the solar radiation of an area where a distributed photovoltaic power station is located, calculating a feature vector through employing an empirical orthogonal function decomposition method, and calculating the power generation efficiency of the distributed photovoltaic power station; a meteorological resource representative point set of final site selection is screened out in combination with site selection conditions; collecting the earth surface solar total radiation measurement data of the meteorological resource representative point set, and correcting the reanalysis time sequence data through a singular value decomposition method; matching a solar radiation reanalysis grid for the distributed photovoltaic power station, and evaluating the distributed photovoltaic power generation efficiency through a physical method and a statistical method based on the reanalysis time sequence data after correction processing to obtain an evaluation value. The method has wide applicability, and resource distribution analysis can be realized only by using a public weather reanalysis data set covering the whole world.

The invention discloses a CO concentration vertical profile inversion method and system based on multi-source remote sensing data fusion. The method comprises the steps that CO column concentration data of a TROPOMI satellite, CO vertical profile mixing ratio data of an MOPITT satellite and ERA5 meteorological reanalysis data are acquired; constructing a training data set through space-time matching and space resampling; an XGBoost machine learning algorithm is used for establishing an estimation model with the TROPOMI column concentration and the ERA5 meteorological parameters as input and the MOPITT ten-layer vertical profile mixing ratio as an output target; and carrying out concentration estimation and data filling on the missing region by adopting the trained model, and generating a CO vertical profile reconstruction data set with high space coverage. According to the method, through multi-source data fusion and intelligent algorithm modeling, the space-time continuity and precision of CO vertical distribution data are remarkably improved, the technical problems that traditional satellite data coverage is insufficient and vertical information is lost are solved, and high-reliability data support is provided for air pollution traceability, cross-regional transmission analysis and environmental governance.

The invention discloses a plateau mountain area avalanche susceptibility intelligent assessment method, and relates to the technical field of high mountain ice and snow disaster space early warning and risk mapping. According to the method, the influence of the microtopography is captured by adopting high resolution, the evaluation accuracy is improved, the control effect of the slope microtopography (the scale of 30-50 m) on the accumulated snow stress field can be accurately captured by adopting a factor map layer (based on data such as 12.5 mALOSPALSARDEM and Landsat-8OLI images) with the spatial resolution of 12.5 m, and compared with traditional 30mDEM and hectometer-level meteorological reanalysis data, the method has the advantages that the evaluation accuracy is greatly improved, and the evaluation efficiency is improved. Spatial differences of key factors such as terrains, earth surfaces and meteorology can be depicted more carefully, more accurate basic data support is provided for avalanche susceptibility assessment, and missing of key microtopography influence factors due to insufficient resolution is avoided.

The invention provides a drought monitoring method and device for a target area, electronic equipment and a storage medium, and relates to the technical field of drought monitoring, and the method comprises the steps: for each pixel corresponding to the target area, obtaining meteorological reanalysis data, a normalized vegetation index and a target land surface temperature corresponding to the pixel in a first preset time period; for each pixel, determining a hot boundary and a cold boundary corresponding to the pixel based on the meteorological reanalysis data and the normalized vegetation index corresponding to the pixel; and determining a drought monitoring result corresponding to the target area based on the hot boundaries, the cold boundaries and the target surface temperature corresponding to all the pixels. According to the technical scheme, for the target area, the hot boundary and the cold boundary are determined pixel by pixel through the meteorological reanalysis data, the situation that the weather conditions of the large area with the complex terrain are different is considered, accurate drought monitoring can also be carried out in the large area with the complex terrain, the applicability of drought monitoring is widened, and the drought monitoring accuracy is improved. Moreover, the method does not need to divide the region, and is simpler in steps.

The invention discloses a camellia oleifera planting area monitoring method and system considering forest age and growth suitability. Comprising the following steps: extracting multi-level features of newly-built and old camellia oleifera forests from a high-resolution image and a multispectral image by using a deep learning model and a machine learning model respectively, and outputting distribution results of the multi-level features; obtaining and integrating meteorological reanalysis data and topographic feature data, calculating and normalizing suitability scores of all dimensions in combination with camellia oleifera growth suitability evaluation, and generating a camellia oleifera growth suitability distribution map; new and old camellia oleifera forest distribution results are fused into a preliminary camellia oleifera planting area, constraint is applied by means of a suitability distribution diagram, and finally a corrected camellia oleifera planting area distribution diagram is obtained. The scheme is high in detection precision and suitable for large-scale oil tea resource monitoring and management.

The invention relates to the technical field of intelligent data identification, in particular to a wetland ecological drought condition identification method based on multi-source data. According to the technical scheme, the method comprises the following steps: data acquisition: acquiring a multi-source data set of a target wetland region in a preset time sequence; the multi-source data set at least comprises a remote sensing reflectivity data set, a surface temperature data set, a meteorological reanalysis data set and an on-site monitoring hydrological data set; and data processing: sequentially carrying out space-time registration, denoising and missing value interpolation processing on the multi-source data set to generate a standardized space-time data cube. According to the method, multi-source data can be integrated to generate a standardized spatio-temporal data cube, each ecological drought index threshold value is dynamically determined based on the background of the historical wet season of the wetland, the comprehensive drought intensity is calculated in a weighted mode according to the wetland type, and accurate wetland ecological drought spatio-temporal distribution, grade sequence and early warning information are output after multi-scale verification. And a basis is provided for wetland ecological protection and water resource regulation and control.

The invention provides a remote sensing estimation method for high spatial-temporal resolution crop evapotranspiration in an arid and semi-arid region, which comprises the following steps of: acquiring cloudless or less-cloudless high spatial resolution optical remote sensing data of a crop growing season in a research region, and calculating or inverting remote sensing earth surface parameters required by high spatial resolution evapotranspiration; comprising a vegetation index NDVI, a leaf area index LAI, a vegetation coverage FVC, a surface water index LSWI and a surface albedo; the method comprises the following steps: acquiring meteorological reanalysis data of daily scales ERA5-Land and GLADAS of a research area, wherein the meteorological reanalysis data comprises relative humidity RH, average temperature Ta, air pressure Pa, solar short-wave radiation # imgabs0 #, downward long-wave radiation # imgabs1 # and upward long-wave radiation # imgabs2 #; according to the technical scheme, multi-source remote sensing data, an evapotranspiration physical model and a space-time fusion technology based on evapotranspiration characteristics are utilized, and estimation of high space-time resolution evapotranspiration of the arid and semi-arid agricultural areas is achieved.

The invention relates to the technical field of power grid icing disaster prediction, and particularly provides a non-monitoring area power grid icing data set construction method and device, and the method comprises the steps: obtaining the topographic data of a target area; taking the topographic data of the target area as input of a pre-trained physical information neural network model to obtain a monitoring data set which is output by the pre-trained physical information neural network model and is consistent with the topographic data in time and space; wherein the monitoring data comprises meteorological reanalysis data and power grid real-time monitoring data. According to the technical scheme provided by the invention, a physical rule and data driven deep fusion mechanism is formed, and micro-terrain self-adaptive correction of meteorological elements can be realized.

The invention discloses a sand storm prediction method based on multi-source data fusion and conditional diffusion generation, and the method comprises the steps: definitely defining a sand storm prediction task as a time-space sequence prediction problem, inputting meteorological data at a plurality of moments, and outputting sand storm change information at a plurality of moments; performing noise addition on the sand storm data according to a noise addition formula to obtain sand storm data with noise; jointly training a multi-source data noise prediction network through a satellite cloud picture at a historical moment, meteorological reanalysis data, sand storm data and a time step t; inputting the complete Gaussian noise map, the current time step tx, and the satellite cloud map, the meteorological reanalysis data and the sand storm data at the historical moment into a multi-source data noise prediction network, predicting the noise of the current noise map, and removing the noise at the current moment through a denoising formula to obtain the sand storm data with noise at the tx-1 moment; and repeating the operation until the t0 moment to obtain a prediction result. According to the invention, sand storm prediction can be accurately carried out.

The present invention discloses a method and system for monitoring oil-tea camellia plantations that takes into account forest age and growth suitability. The system comprises the following steps: extracting multi-level features of newly planted and old oil-tea camellia forests from high-resolution and multispectral images using a deep learning model and a machine learning model, respectively, and outputting their distribution results; acquiring and integrating meteorological reanalysis data and terrain feature data, combining them with an oil-tea camellia growth suitability assessment, calculating and normalizing the suitability scores for each dimension, and generating an oil-tea camellia growth suitability distribution map; integrating the distribution results of new and old oil-tea camellia forests into preliminary oil-tea camellia planting areas, and then applying constraints using the suitability distribution map to ultimately obtain a revised oil-tea camellia planting area distribution map. This solution has high detection accuracy and is suitable for monitoring and managing large-scale oil-tea camellia resources.

The invention discloses an industrial park weather typing method and device integrating a self-organizing mapping neural network and a k-means algorithm, and the method comprises the steps: firstly, constructing an initial data set through air quality monitoring data and ERA5 meteorological reanalysis data, converting the data into a numerical value type, and carrying out the standardization, so as to guarantee that the feature data has the distribution with the mean value being 0 and the standard deviation being 1; thirdly, initializing a self-organizing mapping (SOM) model, and performing dimension reduction on the data by setting parameters such as grid size, sigma and learning rate so as to obtain two-dimensional feature representation; then, initializing a k-means clustering model, setting a clustering number and a random seed, and performing clustering analysis on the feature data after dimension reduction by using a fitpredict () method so as to obtain a clustering label of each sample; in the evaluation and visualization stage, evaluation indexes such as SSE, MSE, RMSE, a contour coefficient and a Clinski-Harabasz index are calculated so as to evaluate the clustering effect and the separation degree. And finally, adding a clustering label into the original data, and visually analyzing the distribution of each feature under different clusters by using a violin chart, thereby providing a visual basis for further analysis. According to the method, through integration of the SOM and the k-means model, the precision of ozone concentration weather typing of the small-scale industrial park is remarkably improved, and the high-concentration ozone weather is divided more finely and more accurately.

The invention discloses a method and a system for evaluating distribution and intensity of freezing rain disasters in hilly areas. The evaluation method adopted by the invention comprises the following steps: collecting meteorological reanalysis grid point data with fine vertical layering in a freezing rain event influence period, calculating freezing rain occurrence altitudes and potential freezing rain amounts of different grid points in the hilly area hour by hour, and downscaling to fine grid points; the method comprises the following steps: taking the potential freezing rain amount of a fine grid point at a certain moment when the altitude is lower than the ground elevation as the freezing rain amount of the grid point at the moment, and then accumulating the freezing rain amounts at all moments by fine grids to obtain the accumulated freezing rain amount of all the fine grids in an influence time period, thereby determining the freezing rain disaster distribution of the hilly and mountainous areas; and determining a freezing rain risk level by combining the obtained freezing rain disaster information of the partial region. The problem that freezing rain disaster assessment is difficult due to serious shortage of glaze data and incomplete freezing rain disaster information in existing hilly areas is solved.

A sand storm prediction method based on multi-source data fusion and static sand source prior belongs to the technical field of sand storm prediction, and comprises the following steps: modeling a sand storm prediction task into a multi-source driven sequence generation problem, and deducing by using historical satellite cloud pictures, grouped meteorological reanalysis data and static sand source prior; through a double-branch architecture, satellite space-time features are extracted by using a three-dimensional U-Net, meteorological physical features are extracted by using a SimVP-Mama network, static prior is fused, multi-source feature alignment and fusion are realized in a bottleneck layer, and space-time evolution information of a future sand storm is output through a decoder. In addition, a level mapping loss function is introduced to optimize intensity level prediction. According to the invention, deep fusion of a physical mechanism and observation data is realized, and the accuracy of sand storm prediction is effectively improved.

The invention provides a water vapor chromatography method, system and device based on machine learning, and a medium, and the method comprises the following steps: constructing a historical training sample set, and employing the training gradient of the historical training sample set to promote a decision tree model, and generating an inclined path wet delay prediction model; establishing a virtual site coordinate in a blank area which does not cover a global navigation satellite system site, constructing a virtual physical feature vector in combination with a digital elevation model and meteorological reanalysis data, and obtaining a virtual oblique path wet delay observation value through the prediction model; and combining a real observation value of a global navigation satellite system station with the virtual oblique path wet delay observation value, establishing an enhanced chromatography equation set, solving by adopting an algebraic reconstruction algorithm, and performing inversion to generate a three-dimensional water vapor density distribution field of the target monitoring area. By implementing the technical scheme provided by the invention, high-precision virtual observation data can be generated in an area with sparse observation stations or uneven signal path distribution, and the stability and spatial resolution of a three-dimensional water vapor inversion result are improved.

The present invention provides a method for evaluating the efficiency of wide-area distributed photovoltaic power generation, which relates to the technical field of photovoltaic power generation efficiency evaluation, and includes: obtaining the reanalysis time-series data of solar radiation in the area where the distributed photovoltaic power station is located, calculating the eigenvectors by using the empirical orthogonal function decomposition method, and screening out the set of meteorological resource representative points for the final site selection in combination with the site selection conditions; collecting the measured data of the total solar radiation on the surface of the set of meteorological resource representative points, and performing correction processing on the reanalysis time-series data by using the singular value decomposition method; matching the solar radiation reanalysis grid for the distributed photovoltaic power station, and evaluating the efficiency of distributed photovoltaic power generation by using the physical method and the statistical method based on the corrected reanalysis time-series data to obtain the evaluation value. The present invention has wide applicability, and only needs to use the publicly available global meteorological reanalysis data set to realize the resource distribution analysis.

The invention discloses a typhoon disaster assessment method, device and equipment and a storage medium, and relates to the field of typhoon disaster assessment, and the method comprises the steps: analyzing a historical typhoon path data set, determining a historical typhoon record of a target region, and determining a historical typhoon center position according to satellite data and meteorological reanalysis data; determining historical polar coordinate data of the historical typhoon based on the historical typhoon center position, and extracting a typhoon structure according to the historical polar coordinate data; the typhoon structure comprises typhoon eyes, typhoon eye walls and a peripheral wind field. Utilizing a preset power law formula to fit the historical wind speed data to determine a historical wind shear index set of the typhoon eye, the typhoon eye wall and the peripheral wind field, and establishing a target wind shear index model based on the historical wind shear index set; and determining historical wind field data in combination with a preset typhoon wind field determination model and the target wind shear index model, and inputting the historical wind field data into a preset typhoon disaster assessment model to obtain a typhoon disaster assessment result. And the typhoon disaster assessment requirement of an unconventional height layer is met.

This disclosure relates to a method, apparatus, equipment, and medium for processing data at wind farm sites. The method includes: acquiring multi-source data, including but not limited to wind power equipment data, ground station meteorological data, and meteorological reanalysis and numerical model data; performing comprehensive checks on the multi-source data and processing the data based on the check results to obtain data to be identified; wherein the comprehensive checks include but not limited to boundary value checks, internal consistency checks, temporal consistency checks, spatial consistency checks, and missing value checks; performing interpolation processing on the data to be identified to obtain target data; wherein the interpolation processing includes but not limited to inverse distance weighted interpolation, spline interpolation, and spatiotemporal bilinear interpolation; and performing anomaly identification on the target data to obtain anomaly identification results. Using the above technical solution can improve data processing efficiency and wind power prediction accuracy, providing technical support for the safe and efficient operation of wind farms.

The application discloses a runoff reconstruction method for a glacier-covered area without actual measurement data, relates to the technical field of geographic information and hydrology, and comprises the following steps: acquiring multi-source heterogeneous data of a target basin; extracting an area time sequence of an ice lake based on remote sensing images, constructing an area-water level-storage capacity relationship curve based on a digital elevation model, and determining an ice lake storage capacity change time sequence; constructing an ice hydrology physical model and integrating an ice lake reservoir module based on ice distribution data and meteorological reanalysis forcing data; determining a prior distribution through a cross-basin parameter migration technology, constructing a likelihood function based on the ice lake storage capacity change time sequence, determining an optimal parameter set based on Bayes theorem, and running the ice hydrology physical model to obtain a primary simulated runoff sequence; and performing deviation correction through a correction model based on the primary simulated runoff sequence, the ice lake storage capacity simulation time sequence and an environmental characteristic sequence to generate a reconstructed runoff sequence. The application can improve the accuracy and reliability of runoff reconstruction.

The invention discloses a human settlement suitability evaluation method based on remote sensing data and an urban heat island effect, and relates to the field of human settlement suitability monitoring, and the method comprises the steps: extracting the urban heat island intensity and the grade thereof in a spatial region range based on meteorological satellite remote sensing data; satellite surface temperature data and relative humidity meteorological reanalysis data are fused, numerical values of surface temperature and relative humidity in a spatial region range are extracted, a temperature and humidity index and a grade thereof are calculated, and a wind efficiency index and a grade thereof are also calculated; and after the obtained urban heat island intensity, the temperature and humidity index and the wind efficiency index are subjected to standardization processing according to the grades, the human settlement environment condition evaluation index is calculated according to the weight of each index. The method focuses on providing an evaluation system based on meteorological satellite product data and the urban heat island effect, and is used for evaluating the influence of climatic elements on the human settlement suitability of the whole country and the city under the climatic change background.

The invention belongs to the technical field of atmospheric pollutant concentration estimation, and particularly relates to an earth surface PM2.5 concentration estimation method and system suitable for high spatial resolution, and the method comprises the steps: collecting PM2.5 concentration data, meteorological reanalysis data and AHI AOD data of a ground pollutant monitoring station in a preset spatial range, and carrying out the space-time matching, and obtaining a matched data set; on the basis of the matched data set, space missing value complementation is carried out on the AHI AOD data, and all-day AOD data is produced; constructing a GSTDNN model based on the graph convolutional neural network, the LSTM model and the CNN model; based on all-day AOD data and a GSTDNN model, PM2.5 concentration estimation is achieved. According to the method, a novel deep neural network GSTDNN model is constructed, and more accurate estimation of the surface PM2.5 concentration is realized.

The invention relates to the technical field of climate prediction, in particular to a high-resolution refined underlying surface sub-seasonal climate prediction method and a related device, and the method comprises the steps: respectively carrying out the correlation analysis according to monthly average analysis data, first monthly average reanalysis data, a weather reanalysis data set and a weather forecast center reanalysis data set, obtaining a plurality of key regions and a plurality of buffer regions, screening to obtain a target simulation region, a target time step size and a target spatial resolution, and obtaining a first region climate mode; processing according to an output result of the third-generation climate prediction system and the first region climate mode to obtain a target initial boundary field; obtaining a target initial field, an initial configuration file and a target region climate mode according to the first region ground data and an output result of a preset third-generation climate prediction system; and performing simulation analysis processing according to the target instruction data to obtain target prediction data. According to the method, the problem of regional high-precision and high-accuracy sub-seasonal climate prediction is solved.

The present application relates to the technical field of aircraft operation state monitoring, in particular to an aircraft fuel consumption calculation method and system based on ADS-B track data; the method comprises the following steps: acquiring ADS-B track data and performing multi-level quality detection and hierarchical repair; acquiring aircraft model information and performance level identifier; fusing meteorological reanalysis data and performing space-time interpolation, and calculating derived flight parameters; constructing a multi-dimensional flight profile and identifying at least eight flight stages; extracting pilot operation habit characteristic parameters and controller command style characteristic parameters from the flight profile, and generating corresponding correction factors; injecting human factor correction factors into a multi-level fuel consumption model; calculating instantaneous fuel consumption rate and performing integration; standardizing output fuel consumption and emission data according to requirements; the present application incorporates two types of core human factors, i.e., pilot operation habit and controller command style, into the fuel consumption calculation framework based on ADS-B data, eliminates inherent calculation bias caused by human factors, and has higher calculation accuracy.

The invention discloses an InSAR troposphere delay correction method and system based on multi-source model adaptive fusion, and relates to the technical field of InSAR monitoring. The InSAR troposphere delay correction method comprises the following steps: acquiring SAR, DEM, meteorological reanalysis data and optical remote sensing water vapor data; preprocessing the SAR data to obtain a standard differential interferogram and a phase; constructing a phase-elevation empirical model based on a differential interferometric phase and a DEM, calculating an initial estimated value of troposphere delay corresponding to each model in combination with a weather model and optical remote sensing data, and performing magnitude optimization on the troposphere delay estimated by the weather model; and multi-source troposphere delay adaptive fusion is carried out through a window-pixel level weighting strategy, and troposphere delay correction is carried out on the differential interference phase according to the fused troposphere delay phase. The advantages of a multi-source troposphere delay correction technology can be integrated, the troposphere delay space change characteristics are adapted, and the correction accuracy and the deformation monitoring precision are effectively improved.

This invention discloses a high-resolution 2-meter surface temperature estimation method based on remote sensing imagery. The method utilizes both high-resolution and low-resolution remote sensing imagery data to extract feature information at different spatial scales, and then fuses and models them within a unified neural network framework. The model employs a feature interaction structure incorporating an attention mechanism to achieve the association and optimization of multi-scale features. The output generates a gridded 2-meter surface temperature distribution result for the corresponding spatial region. During training, a multi-objective hybrid supervision strategy is designed, combining meteorological reanalysis data and surface temperature reference data to simultaneously ensure prediction accuracy and spatial consistency. This invention enables city-level high-resolution temperature estimation in areas with sparse meteorological observation stations, effectively improving the spatial accuracy and stability of 2-meter surface temperature estimation.

This invention relates to the field of aviation big data analysis technology, and particularly to a method, device, electronic device, medium, and product for predicting aircraft fuel consumption. The method includes: acquiring aircraft flight access records and meteorological reanalysis data; processing the data to obtain standardized historical flight data, standardized reference trajectory data, and future meteorological field data; processing the standardized reference trajectory data and future meteorological field data respectively to obtain a reconstructed trajectory for the future flight segment and a four-dimensional meteorological analysis result; combining the two to extract environmental parameters of the trajectory points; generating a physical enhancement feature set based on the standardized historical flight data and trajectory point environmental parameters; and obtaining the aircraft fuel consumption prediction result by using a preset gradient boosting decision tree. This invention solves the problems of poor physical interpretability, inaccurate prediction in complex stages, and weak future prediction ability in related technologies, improving the physical interpretability and prediction accuracy of aircraft fuel consumption prediction, and enhancing the precision and robustness of fuel consumption prediction.

The present application discloses a regional wind field prediction method, device, equipment and storage medium, which relates to the field of weather forecast technology, including: using a mesoscale weather model to simulate the terrain wind field of a target area based on a preset meteorological reanalysis data set and preset static high-resolution geographic data to obtain mesoscale wind field data and performing block interpolation processing on the mesoscale wind field data to obtain a target interpolation result, inputting the target interpolation result into a preset LBM microscale model to obtain a stable wind field prediction result and determine the corresponding current result acquisition time; using the mesoscale weather model to obtain the mesoscale simulation result corresponding to the target area at the current result acquisition time, and jumping to the step of obtaining mesoscale wind field data until the current result acquisition time is the target prediction stop time to obtain the target wind field prediction result of the target area. In this way, a high-precision wind field prediction result can be obtained to enhance the accuracy of weather forecasts under complex terrain.

The invention discloses a bridge site area wind speed long-term prediction method fusing meteorological data and actually measured data, and the method comprises the steps: firstly obtaining meteorological reanalysis data and corresponding bridge site actually measured wind speed data; then, uniformly dividing the data into a training set, a verification set and a test set for model training, hyper-parameter optimization and prediction performance evaluation; afterwards, meteorological reanalysis data is used as input, the low-resolution actually-measured average wind speed is used as output, a space-time diagram neural network is applied to carry out multi-step advanced wind speed prediction, and the low-resolution predicted average wind speed is obtained; and finally, carrying out up-sampling processing on the prediction result to generate a high-resolution interpolation average wind speed, taking the high-resolution interpolation average wind speed as input and the high-resolution actually measured average wind speed as output, and carrying out wind speed time sequence probability regression by applying a U-shaped convolutional neural network to obtain a high-resolution predicted average wind speed and a corresponding prediction interval. The method can effectively improve the accuracy of long-term prediction of the wind speed of the bridge site area, and synchronously considers the uncertainty of the prediction result.