29 results about "Numerical weather prediction" patented technology

This invention discloses a short-term intelligent numerical weather prediction method and system based on environmental perception, belonging to the field of meteorological management technology. The system includes a data acquisition and fusion module, an intelligent zoning and feature mining module, a short-term forecasting and risk assessment module, and a visualization and focusing module. The data acquisition and fusion module collects and fuses environmental data from meteorological radar and weather stations, as well as GIS maps. The intelligent zoning and feature mining module divides the data into grids based on the GIS map, analyzes the rainfall synergy between grids based on historical environmental data to construct analysis groups, and selects and constructs an impact set for each analysis group. The short-term forecasting and risk assessment module fits near-term precipitation intensity relationships and short-term precipitation intensity relationships based on the impact sets of each analysis group. The visualization and focusing module calculates and outputs the near-term and short-term predicted precipitation intensities for each analysis group based on the relationships, identifies key areas of concern using topographic and hydrological models, and enhances visualization.

The application discloses a power meteorological data fusion method and system based on multi-source meteorological prediction and a storage medium, and combines power meteorological monitoring and multi-source numerical weather prediction data. Through error analysis on the multi-source numerical weather prediction data, the numerical weather prediction is changed from single deterministic prediction to multi-source fusion prediction, which is beneficial to the real application of meteorological information to actual business work of the power grid. The application can provide precision evaluation of different prediction data sources, provide multi-source prediction fusion data, improve prediction precision, and provide technical support for power meteorological fine prediction.

The application discloses a numerical weather prediction optimization method and system based on meteorological resource coupling, belongs to the technical field of numerical weather prediction, focuses on the multi-source data in the target area, and constructs a double-dimension correlation between the power grid and the meteorology, integrates the meteorological resources, and is targeted to gather the power grid output, so that the subsequent numerical weather prediction is more in line with the power prediction requirements of the power grid, and through the correlation matrix and the comprehensive influence coefficient, the qualitative coupling relationship is converted into quantitative indexes, linear and nonlinear are considered, and therefore the prediction accuracy under complex weather and complex terrain is greatly improved, high-precision meteorological driving data is provided for photovoltaic short-term power prediction, and the problem that the existing numerical weather prediction is mostly based on single dimension for prediction, so that it is difficult to guarantee the numerical accuracy when facing environmental changes is solved.

The application discloses a global average sea clutter near real-time estimation method, system, device and storage medium, including the following steps: obtaining multi-source satellite-borne radar historical backscattering coefficient data, and performing time and space matching and binning on the ERA5 wind field; an average sea clutter estimation model is established according to a geophysical model function fitting; the satellite-borne microwave scatterometer wind field is corrected in deviation and fused with a numerical weather prediction wind field, so that a high-coverage, continuous near real-time global sea surface wind field is obtained; the relative azimuth is calculated in combination with radar observation parameters and fused wind direction, and global average sea clutter data is obtained by using the model; the application improves the precision and application range of sea clutter estimation under the condition of medium and small incident angles, provides a high-quality input wind field through a multi-source wind field fusion technology, realizes rapid, stable and near real-time estimation of global average sea clutter, has high timeliness and business feasibility, and can meet the needs of near real-time sea clutter monitoring and ocean remote sensing application.

The application provides a new energy power generation prediction method and system based on numerical weather prediction, and relates to the technical field of new energy power generation. First, the numerical weather prediction spatial grid multi-element data of a target area is obtained, feature construction is performed, and time domain features and space domain features are obtained. The time domain features and the space domain features are spliced to obtain space-time features. A preset point prediction model and a quantile regression model are trained to obtain trained point prediction models and quantile regression models. Finally, the space-time features are input into the trained point prediction models and quantile regression models, and new energy power generation prediction results are output. The application improves the accuracy of new energy power generation prediction by extracting the space-time features of the numerical weather prediction spatial grid multi-element data.

This invention relates to the field of new energy power generation technology and discloses a multi-model fusion wind speed correction method and apparatus. The method includes: constructing a first wind speed correction model based on radar three-dimensional wind field data, numerical weather prediction data, and measured wind speed data at wind turbine hub height; constructing a second wind speed correction model using statistical algorithms based on wind speed forecast data and wind speed observation data; constructing a third wind speed correction model using data-driven algorithms based on wind speed forecast data, wind speed observation data, and terrain data; acquiring real-time wind condition data; and performing wind speed correction based on the real-time wind condition data using the first, second, and third wind speed correction models to obtain the wind speed correction result. This invention effectively solves the adaptability problem under complex meteorological conditions and improves the accuracy of wind speed correction.

This application provides a method, apparatus, terminal, and storage medium for generating a background error covariance matrix, relating to the field of numerical weather prediction technology. The method includes: constructing a short-term forecast sample set for a target numerical weather prediction; extracting control variables for each background error sample in the short-term forecast sample set; calculating the regression coefficients of each control variable to construct a balance operator; inputting the short-term forecast sample set into a pre-constructed characteristic length scale field generation model, outputting the characteristic length scale field of each background error sample, and constructing a horizontal correlation operator; calculating the vertical correlation scale of each control variable at each horizontal position to construct a vertical correlation operator; calculating the background error standard deviation of each control variable at each grid point to construct a standard deviation operator; and using the balance operator, horizontal correlation operator, vertical correlation operator, and standard deviation operator to obtain the background error covariance matrix. This application can reduce the computational complexity of the horizontal correlation operator and improve computational efficiency.

The present application belongs to the field of wind power prediction technology, and particularly relates to a method for improving the accuracy of wind power prediction. First, numerical weather prediction data, wind turbine operation data and measured power data in a short time period are selected, and abnormal data processing and data normalization processing are performed. By gradually reducing the input variables, the influence of variable loss on prediction accuracy is compared, and variables with greater influence on prediction accuracy are screened out. Then, longer time variables and measured power data are selected, and abnormal data processing and data normalization processing are completed. The normalized data is trained by a long short-term memory network to obtain a trained model for wind power prediction of an actual system. The method for improving the accuracy of wind power prediction can balance the calculation amount, calculation time and prediction accuracy of model training, has small calculation amount, short time consumption, high prediction accuracy and high application value.

The application provides a kind of distributed photovoltaic cluster power prediction method and system, and the application relates to the technical field of power system automation and new energy power generation prediction, the prediction method of the application specifically includes: obtaining the spatial hash index of the panoramic orthographic image and multi-view image covering the target area;Identify the target position of photovoltaic module on the panoramic orthographic image, solve the attitude parameter of photovoltaic array;Construct source and load characteristic standard library, determine the best matching standard user of the user to be tested;Decouple the data side photovoltaic output value of the user to be tested by using the source and load behavior mode of the best matching standard user, obtain the data side photovoltaic output value;Combine the attitude parameter to calculate the theoretical photovoltaic output value;Get the corrected photovoltaic power sequence and the corrected load sequence;Combine numerical weather prediction to carry out single-household photovoltaic and load prediction, and obtain the power prediction result of distributed photovoltaic cluster through topological aggregation.

The present application relates to the field of atmospheric science and numerical calculation technology, and discloses a kind of atmospheric numerical simulation method and application based on the conservation constraint of multiphase water substance, and the evolution calculation of each phase water substance under the condition of discrete grid and discrete time step is oriented to the cloud microphysical parameterization of atmospheric numerical model.The method obtains the density and velocity field of each component in the grid unit, constructs the mass proportion variable normalized with wet air density;Define the growth rate per unit volume per unit time and represent the phase change source and sink with the component continuity equation;According to the mass proportion evolution equation, the coupling discrete updating rule is established, the total water substance growth rate residual error and the proportion residual error corresponding to the equivalent conservation expression are calculated, and the corrected growth rate that satisfies the conservation constraint is obtained through consistent correction to rewrite and update the mass proportion field.The method can inhibit the drift of water substance balance and reduce the phase distribution error under the condition of mixed phase cloud, and is suitable for numerical weather prediction and regional numerical simulation.

The application discloses a kind of distributed spaceborne D-InSAR troposphere atmospheric three-dimensional water vapor and cloud retrieval method.The application utilizes the multiple sets of differential interference observation data obtained by distributed spaceborne synthetic aperture radar (SAR) at different times, combined with external numerical weather prediction (NWP) data, to realize the inversion of troposphere absolute refractive index;On this basis, based on the physical coupling relationship between absolute refractive index and water vapor, the three-dimensional water vapor structure is obtained by inversion.Further, by comparing the water vapor obtained by inversion with saturated water vapor pressure to calculate relative humidity distribution, and combining the relationship model between cloud amount and relative humidity to set the relative humidity threshold, a binary cloud mask is constructed, so that the inversion of high-resolution three-dimensional cloud distribution is realized.

This invention discloses a photovoltaic power prediction method and system, belonging to the field of photovoltaic power prediction technology. It acquires the solar position parameters of a photovoltaic power station at each acquisition time, uses interpolation to determine the corresponding sunrise and sunset times, and then determines the time offset characteristics of each acquisition time relative to sunrise and sunset. The solar position parameters and time offset characteristics are encoded as physical enhancement features. Dynamic bias correction is applied to the numerical weather prediction (NWP) data through convolution operations, outputting the corrected meteorological features. The physical enhancement features, historical power data, NWP data, measured meteorological data, and the corrected meteorological features are used as inputs to a prediction network, which then outputs the photovoltaic power prediction result. The prediction model constructed by this method significantly improves prediction accuracy by introducing multi-dimensional physical features and establishing a dynamic NWP correction mechanism.

This application relates to a distributed storage method for large-scale GRIB grid data. The method includes: a GRIB objectification technique based on encoding features, which treats the entire GRIB message as an object data body and self-describing information as metadata, achieving native support for grid fields of arbitrary sizes; a core metadata dataset definition technique based on virtual keys, which maps each GRIB message in the GRIB file to a single storage object and maps the self-describing information in the GRIB message to the metadata of the object storage; and fully utilizing the independent and distributed storage characteristics of GRIB objects, employing distributed parallel pipeline processing to leverage the advantages of distributed computing, thereby significantly improving the distributed parallel processing efficiency of large-scale grid data. This application solves the problems of coarse granularity, poor consistency, and low access efficiency in traditional retrieval methods, and can be widely applied to Earth observation, numerical weather prediction, and big data analysis scenarios.

This invention discloses a short-term ZWD prediction method based on deep learning and attention mechanisms. To address the problem of insufficient accuracy and stability in ZWD short-term prediction under conditions of multiple influencing factors coupled with significant seasonal variations and sudden non-stationary fluctuations in GNSS positioning and numerical weather prediction applications, this invention obtains ZWD observations, water vapor pressure, and surface temperature at L consecutive time points from the site to be predicted and concatenates them with latitude and longitude static features to form a real-time historical feature sequence. Further, it calculates the first-order difference, second-order difference, and relative moving mean deviation of ZWD to form an event feature vector. Using an event-triggered attention-gated network, it outputs seasonal, sudden, and transitional gating weights, driving the seasonal, sudden, and transitional experts in the hybrid expert Transformer model to generate ZWD prediction sequences with a prediction step size of H. Finally, the output is obtained by weighted fusion according to the gating weights. This method effectively captures the abrupt changes at key moments while taking into account seasonal patterns, thus improving the accuracy and stability of ZWD short-term prediction.

The present application relates to a kind of adaptive short-term-numerical precipitation forecast fusion method, specific steps are: obtaining the grid hour precipitation forecast data of short-term forecast precipitation field and numerical forecast precipitation field of target area;According to the size of precipitation, the precipitation is divided into six precipitation grades and setting hierarchical protection coefficient;Calculate basic weight function;Dynamically adjust the end time of exclusive period and the end time of transition period of short-term precipitation forecast;For the precipitation of more than 3 hours continuously dominated by the same forecast source, extend the trust time of the forecast source;Calculate the fusion weight of short-term precipitation forecast and numerical precipitation forecast;The fusion weight of short-term precipitation forecast is carried out spatial consistency correction;Short-term precipitation forecast and numerical precipitation forecast are preliminarily fused to obtain fusion precipitation field;First-order lag filter is carried out to fusion precipitation field, and abnormal value is removed, to obtain final fusion result.

This invention discloses a method for monitoring and early warning of icing on power transmission lines that integrates BeiDou and PolSAR. The method includes: synchronous acquisition and preprocessing of multi-source data; inversion of high spatiotemporal resolution atmospheric precipitable water using BeiDou CORS station data, combined with temperature and pressure data to calculate the tropospheric delay phase; simultaneous correction of orbital errors in PolSAR images using precise BeiDou ephemeris data; extraction of a set of characteristic parameters closely related to icing conditions; establishment of an icing thickness inversion model based on phase change and polarization characteristics after fusion correction; construction of an icing growth prediction model based on time-series icing thickness inversion results, combined with near-real-time atmospheric temperature and humidity data inverted from BeiDou and numerical weather prediction data; and initiation of icing early warning based on real-time inverted thickness, predicted thickness, and line design icing thickness. This method addresses the problems of existing technologies that primarily focus on icing condition identification but lack accurate prediction and early warning capabilities for icing growth trends.

The application discloses a numerical model prediction deviation self-adaptive correction method fusing satellite and ground observation data, and belongs to the technical field of meteorological data processing and numerical weather prediction. The application obtains numerical model prediction grid data, ground station observation data and satellite remote sensing observation data and carries out space-time matching; calculates self-adaptive confidence according to station density and satellite quality index, and generates comprehensive observation reference weight; adopts an adaptive sliding window Kalman filtering method to recursively estimate the model prediction deviation; constructs an improved U-Net convolutional neural network residual correction model to nonlinearly correct the prediction field; and is deployed to a business prediction process and continuously updated through an online learning strategy. The application effectively solves the problems of insufficient utilization of observation data, insufficient self-adaptive capacity and low correction accuracy in sparse observation areas, and significantly improves the accuracy and reliability of numerical model prediction under complex terrain and long-time conditions.

This invention discloses a method, system, equipment, and medium for forecasting ultra-local aquaculture operation windows, belonging to the field of smart aquaculture management technology. It collects and fuses meteorological and oceanographic data from offshore wind farms with regional numerical weather prediction data to obtain fused data. Based on the fused data, the regional numerical weather prediction data is processed through downscaling or correction models to generate ultra-local environmental parameter forecasts for target aquaculture sites. According to preset environmental parameter suitability rules corresponding to specific aquaculture operations, the ultra-local environmental parameter forecasts are judged, and suitable time windows for specific aquaculture operations are identified and output. This effectively solves the pain points of insufficient targeting and low accuracy in traditional aquaculture environmental forecasts.

This application belongs to the field of artificial intelligence and numerical weather prediction fusion technology. Addressing the technical problems in existing technologies where there is insufficient physical consistency, poor multivariate synergy, inadequate observation fusion, and difficulty in effectively suppressing systematic biases between the output of large-scale AI weather prediction models and the input of regional numerical models, this application proposes a physical perception fusion bridging model, a weather prediction method, and related devices. A multi-scale encoder extracts multi-scale meteorological field features from the output of the large-scale AI weather prediction model; a physical constraint decoder decodes the meteorological field features to generate a high-resolution meteorological field that conforms to physical conservation laws, with a physical conservation penalty term introduced during training; and an observation fusion layer fuses multi-source observation data and the high-resolution meteorological field to obtain a high-resolution initial field. This application achieves seamless integration between the large-scale AI weather prediction model and the regional numerical model, ensuring physical consistency and fusing multi-source observation data.

The application provides a typhoon forecast-based target observation sensitive area identification method and system, and belongs to the technical field of numerical weather prediction and ensemble prediction, and comprises the following steps: a target area is delimited with a typhoon message center position; based on a business prediction model, Lanczos iteration algorithm and multi-scale singular vector algorithm are used to obtain tropical cyclone wet singular vectors and mesoscale and small-scale singular vectors by coupling large-scale condensation linear physical processes; humidity variables and the contribution of multi-scale singular vectors to the thermal structure of the typhoon are introduced to calculate the total energy of each grid point; the total energy of the grid points on different isobaric surfaces is accumulated along the vertical direction to obtain a multi-scale energy field; weighted summation normalization processing is performed according to singular value weight distribution; and the target observation sensitive area is obtained based on a set threshold value. The application solves the problems of incomplete physical process description, limited sensitive area identification precision and insufficient business applicability in the existing typhoon target observation technology.

The application provides a bus-level ultra-short-term net load interval prediction method, device, equipment and storage medium, the method comprises the following steps: obtaining electric load historical data, photovoltaic power historical data and wind power historical data, and preprocessing the data; obtaining numerical weather prediction data, and performing downscaling processing on the time resolution thereof; constructing a prediction feature set based on the preprocessed historical data and the downscaling-processed numerical weather prediction data; constructing a deep neural network prediction model for electric load, photovoltaic power and wind power respectively; and calculating bus-level net load ultra-short-term interval prediction results according to obtained load ultra-short-term interval prediction results, photovoltaic power ultra-short-term interval prediction results and wind power ultra-short-term interval prediction results. The application can accurately predict the probability interval of bus-level net load, and can provide data support and important reference basis for real-time risk early warning and auxiliary decision-making of the power system.

This invention belongs to the field of power prediction technology, specifically relating to a wind farm power prediction method based on improved numerical weather prediction and artificial intelligence algorithms. The steps include: acquiring multi-source data from the wind farm, performing time alignment, anomaly detection, and standardization, and then inputting the data into a generative adversarial network (GAN) with fused physical constraints for data completion; extracting physical features, temporal features, and spatial features through a multi-branch feature extraction network and fusing them to obtain a comprehensive fused feature; identifying error-sensitive areas in numerical weather prediction using a multi-scale attention module and correcting the numerical weather prediction wind speed; based on the corrected wind speed data, outputting ultra-short-term power prediction results and short-term power prediction results through a multi-task learning framework including trend consistency constraints; and performing uncertainty quantification analysis on the ultra-short-term and short-term power prediction results. This invention achieves high-precision, highly robust, interpretable, and easily deployable wind power prediction.

This invention discloses a method for optimizing the scattering properties of condensed objects based on physically constrained deep learning. The method includes constructing a pre-trained dataset and a fine-tuning dataset; pre-training a deep neural network model to obtain initial values ​​for the scattering properties of condensed objects; inputting the background field of the corresponding numerical model into a physical radiative transfer model for forward computation to obtain simulated brightness temperature; using the adjoint operator of the physical radiative transfer model to calculate the gradient of the brightness temperature data, establishing a complete differentiable gradient chain from observed brightness temperature to network parameters; using the consistency of the probability distribution between observed and simulated brightness temperature as the optimization objective, performing gradient descent iterative training; fine-tuning the network parameters of the deep neural network model; and embedding the trained deep neural network model into a fast radiative transfer model to generate optimized scattering properties of condensed objects in real time. This method can provide higher-quality observation operators for the assimilation of satellite data in cloud and rain areas, thereby improving the accuracy of numerical weather prediction.

The application discloses a hybrid particle variational assimilation method based on Gauss relaxation and incremental weighting, which comprises the following steps: adding random disturbance noise in a mode initial field to generate a priori set containing a plurality of particles; calculating the non-Gaussian degree of the deviation distribution of the a priori set; serially processing observations to calculate the marginal weight of each particle at each grid point and the corresponding adaptive inflation parameter of the observation error; selecting the hybrid particle variational assimilation method based on Gauss relaxation or incremental weighting to output a posteriori particle set according to the non-Gaussian degree of the distribution of the a priori set; using the mean field of the posteriori set for deterministic prediction, using the set prediction field as the a priori set at the next moment, and repeating the above steps. The application effectively combines the advantages of the particle filter and the variational assimilation method, significantly enhances the adaptability of the data assimilation system to nonlinear and non-Gaussian observations, and can effectively improve the accuracy of numerical weather prediction.

This invention discloses a 100-meter-level terrain processing method that preserves the refined terrain features of a model, relating to the field of numerical weather prediction technology. The method includes: analyzing the changes in terrain power spectrum under the action of a terrain smoothing function based on a terrain set constrained by the terrain spectral density, and adjusting the terrain smoothing function to generate improved smoothing function configuration data; smoothing the 100-meter-level basic terrain set based on the improved smoothing function configuration data to generate spectrally faithful smoothed terrain data; performing terrain power spectrum consistency verification and terrain height deviation verification on the spectrally faithful smoothed terrain data to generate a spectrally faithful model terrain field; and writing the spectrally faithful model terrain field into the numerical weather prediction model preprocessing interface to generate a spectrally faithful preprocessed terrain set. This invention achieves the effects of improving the model's refined terrain representation capability, spectral consistency, and adaptability to numerical weather prediction model preprocessing.

The present application belongs to the field of numerical weather prediction data correction, and relates to a numerical weather prediction wind speed correction method and system based on wind profile information. The method comprises the following steps: obtaining numerical weather prediction data of a target position point in a target period, fitting a target numerical weather prediction wind profile function based on the numerical weather prediction data, extracting a target wind profile feature parameter from the target numerical weather prediction wind profile function, inputting numerical weather prediction wind speed values of the target position point in each preset standard height layer and the target wind profile feature parameter into a trained layered mapping model to obtain corrected wind speed values of the standard height layer. The present application improves the wind speed correction accuracy and spatial representativeness of different spatial positions in a wind farm, and obtains more accurate target height wind speed.

This invention relates to the field of new energy power generation prediction technology, specifically, to a method for constructing new energy power prediction samples and training models. By simultaneously collecting historical actual power data, measured meteorological data, and historical numerical weather prediction data, measured meteorological sample sets and NWP sample sets are constructed respectively. The two sample sets are merged and source identification features are added to train a hybrid prediction model. A hierarchical model is established based on a Bayesian framework to analyze the systematic bias and random error distribution of NWP data, generating diverse virtual NWP samples and constructing a data augmentation training set. The augmented prediction model is trained using the augmented training set and then weighted and integrated with the baseline model, NWP model, and hybrid model to output the final prediction model. This invention, through innovative sample construction and training mechanisms, effectively solves the problem of inconsistent distribution between training and inference data, significantly improving the accuracy and robustness of power prediction.

This invention discloses a method for correcting incoming wind speed in offshore wind farms based on SAR satellite observations. First, SAR satellite images are preprocessed and masked to remove interference from bright targets such as wind turbines. Then, a high-resolution near-sea surface wind speed is retrieved from the images using an improved ConvNeXtv2 deep learning network. This is then matched with numerical weather prediction (NMR) wind fields, and the amplitude and spatial structure of the NMR wind fields are corrected using the LightGBM model. Finally, the wind shear index is corrected by combining atmospheric elements at the wind height level with the wind speed extrapolation formula, thereby correcting the extrapolation ratio and obtaining high-precision wind speed extrapolation results at the wind turbine hub height. This invention integrates the high spatial resolution of SAR with the continuity of numerical weather prediction, effectively eliminating wind turbine target interference and numerical weather prediction system bias, significantly improving the accuracy of incoming wind speed at the wind turbine hub height, and providing reliable wind field data for offshore wind farm power prediction and safe operation.

This invention belongs to the field of weather forecasting technology and relates to an ultra-fast numerical weather prediction method. It includes regional numerical model compilation and setup, acquisition of global numerical weather prediction data products, processing and preprocessing of multi-source meteorological observation data, fusion and assimilation analysis of multi-source meteorological observation data, numerical model integral calculation, post-processing of numerical weather prediction products, rapid updating of numerical weather predictions, comparative verification of the assimilation and forecasting effects of different data, selection of observation data with shorter latency for assimilation application and numerical integral calculation, and establishment of an operational ultra-fast numerical weather prediction system. This invention establishes an ultra-fast numerical weather prediction system that, while ensuring forecast accuracy, further shortens the generation time of numerical weather predictions, thereby improving the timeliness of regional model forecast products and the ability to forecast and warn of severe weather.