342 results about "Weather forecasting" patented technology

The invention provides a new energy electric power meteorological prediction method and device based on a multi-modal large model, and belongs to the field of meteorological prediction. The method provided by the invention comprises the steps of obtaining multi-source meteorological observation data and new energy station operation data, and generating a multi-modal fusion feature; a large meteorological prediction model is constructed, the large meteorological prediction model adopts a dynamic graph neural network structure, nodes represent geographic space positions, edges represent spatial adjacent relations, and node features comprise numerical values of meteorological elements; using the multi-modal fusion features to train the meteorological prediction large model, and minimizing a meteorological element prediction error; and inputting real-time multi-source data of a to-be-predicted area into the trained meteorological prediction large model, and outputting meteorological types and meteorological element values of the to-be-predicted area within a preset duration. According to the method and the device provided by the invention, the problems of insufficient accuracy of new energy electric power weather prediction, weak combination of geographical and physical rules and difficulty in adaptation to different weather types can be solved.

The invention discloses a radar echo extrapolation method and system based on frequency domain enhancement, and the method mainly comprises the following steps: obtaining and preprocessing a historical radar echo grayscale image sequence, generating a sequence sample through a sliding window, and dividing the sequence sample into a training set, a verification set and a test set; the method comprises the following steps: constructing a frequency domain enhanced U-Net network comprising an encoder-decoder structure, introducing a multi-scale deep convolution structure into an encoder and a decoder, and enhancing frequency domain features by using a frequency domain dynamic attention mechanism in jump connection; inputting the training set into the model for training by adopting a composite loss function comprising intensity weighted loss, frequency domain consistency loss and structural similarity loss; and inputting the test set into the trained model, and outputting a radar echo prediction result at a future moment. The method can be effectively applied to the fields of short-term and temporary weather forecast, severe convection monitoring and the like, and provides more accurate and reliable radar echo prediction support for meteorological disaster early warning.

The invention provides an HY-2A ocean water vapor inversion method based on machine learning, and the method comprises the steps: constructing a high-quality HY-2A scanning microwave radiometer data set through the data preprocessing, matching, feature extraction and normalization processing of HY-2A satellite scanning microwave radiometer data and ERA5 reanalysis data. A plurality of machine learning models are used for training and testing, Bayesian optimization is used for realizing hyper-parameter automatic adjustment, an optimal solution in finite time is obtained, an SHAP method is used for explaining the models, contribution of each characteristic variable is clear, interpretability of the used machine learning models is improved, and finally high-precision inversion of the water vapor over the sea is realized. The method has efficient calculation performance, can realize rapid processing of large-scale marine meteorological data, and improves the interpretability of a machine learning model. The method has important application value in numerical weather forecast, ocean extreme weather forecast, climate monitoring and other geophysical and meteorological fields.

The invention discloses a photovoltaic cleaning robot operation scheduling method and system based on environment prediction, and belongs to the technical field of photovoltaic power station intelligent operation and maintenance and robots, and the method comprises the steps: obtaining weather forecast data, photovoltaic panel real-time dust accumulation data and historical power generation efficiency attenuation data, and generating environment prediction parameters; generating a dynamic weight value based on the environment prediction parameters and a preset power generation loss threshold value, and judging the sweeping emergency degree of each photovoltaic panel area; distributing an operation priority according to the cleaning emergency degree, and generating an initial cleaning task sequence; and a multi-robot cooperative path planning result is generated by combining the photovoltaic power station topological structure and the real-time position data of the robots. According to the method, the environment prediction technology and the dynamic scheduling strategy are combined, the cleaning task can be adaptively adjusted according to weather changes and dust accumulation conditions, the robot operation path is optimized, and the power generation efficiency of a photovoltaic power station and the utilization rate of robot resources are remarkably improved.

The invention discloses a multi-source data and physical information fused tornado space-time intelligent prediction method, which comprises the steps of collecting observation data of a Doppler weather radar and reanalysis data of a mid-term weather forecast center, performing projection conversion, space alignment, time interpolation and the like of space-time dimensions, and fusing key physical characteristics of the tornado; establishing a space-time convolution model network suitable for tornado multi-scale feature extraction, inputting fusion features and corresponding coordinate information, and outputting a prediction control quantity of tornado space-time evolution; a physical control equation of the atmospheric motion and thermodynamic process is converted into a penalty term, a space-time convolution model loss function is embedded, and physical information constraint is carried out on network parameters; and obtaining tornado prediction control quantities of different time windows in the future, and generating a probability graph of the potential development area of the tornado. Through fusion of multi-source data and a physical mechanism, future tornado event probability is quantified, and reliable data support and scientific basis are provided for meteorological prediction and emergency response.

The invention relates to the technical field of weather forecast, discloses a method for classifying severe convection weather forecast, and aims to solve the problem that complexity of severe convection weather requires multi-dimensional data support, so that a three-dimensional data acquisition network covering'ground-air-sky 'needs to be constructed. Ground observation data need to include minute-level rainfall, hourly air temperature and humidity (emphatically paying attention to humidity difference between 850hPa and 500hPa and reflecting unstable stratification) and 10-minute average wind speed of a meteorological station, and high-altitude detection data need to extract temperature vertical profiles (calculating convective condensation height LCL) and wind speed vertical shear (shear values of 0-3km and 0-6km) at 08 o'clock and 20 o'clock every day. According to the method for classifying severe convection weather forecast, new signals (such as sudden cloud top brightness temperature drop) observed in real time are rapidly absorbed, and meanwhile, the method is adaptive to severe convection characteristic differences of different areas (such as mountainous areas and plains) and different seasons, so that the forecast precision is improved, and the requirements of refined disaster prevention for high-accuracy and high-timeliness forecast are met.

The invention discloses a garden partition intelligent water-saving irrigation regulation and control system based on the Internet of Things, and belongs to the field of irrigation regulation and control systems. According to the system, a basic water demand model containing static characteristics of plants, soil and the like is established for each subarea through a central processing controller, real-time environment data and weather forecast are fused, and future irrigation water demand and optimal time are calculated in a prospective mode. The core innovation of the method lies in that after irrigation, the system performs feedback iterative optimization on a basic water demand model by comparing the efficiency deviation of the actual change and the predicted change of the soil humidity to form a self-learning closed loop. According to the method, conversion from passive response to active prediction is achieved, the method has the advantages of being accurate, prospective and self-adaptive, the method can continuously adapt to environmental changes, and the intelligent level and the water saving effect of irrigation are remarkably improved.

The invention belongs to the technical field of wind power weather forecasting, and provides an all-region three-dimensional wind speed correction method and system, and the method comprises the steps: constructing a weather numerical forecasting model, and obtaining wind field forecasting data; fusing the preprocessed multi-source data by adopting an optimal interpolation method to obtain three-dimensional space-time continuous wind field analysis data; based on the wind field forecast data and the wind field analysis data, features are extracted and fused, then a historical forecast error sample set is constructed, a wind speed correction model is constructed, and the historical forecast error sample set is utilized to train the wind speed correction model; introducing an initial value, a physical parameter and boundary condition disturbance, calculating a mean value and a standard deviation of each set result, extracting a probability distribution feature of a wind speed, constructing a confidence interval, estimating a probability density function, and quantifying an occurrence probability of an extreme wind speed event; and the prediction result of the wind speed correction model and the multi-source wind field observation data are fused to generate final three-dimensional wind field data, so that the actual requirements of wind power prediction and power grid dispatching can be met.

The invention relates to the field of business forecasting, and discloses an intelligent weather forecasting system for numerical mode simulation, which is used for timely adjusting decisions according to business execution states and comprehensively improving weather forecasting accuracy and business decision scientificity, adaptability and real-time performance. The intelligent weather forecasting system based on numerical mode simulation comprises the steps that a cross-scale continuous correction vortex field is generated through spatial fractional order Laplace operation in combination with a bit vortex conservation law, meteorological phase change events such as cyclones and frontal surfaces are recognized through continuous coherence analysis topology, and invalid data are filtered; a Riemannian metric field of meteorological-commercial curvature conservation mapping is constructed based on conformal differential transformation, a decision manifold is constructed in combination with symplectic structure tensor and Gaussian curvature constraint, decision parameters such as a minimum entropy generation path output logistics curvature coefficient and an inventory elastic factor are solved, and real-time commercial data feedback is introduced to realize dynamic closed-loop calibration of the decision manifold. And the decision-making precision and response timeliness of commercial systems such as supply chains and energy sources in extreme weather are remarkably improved.

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 discloses a radar echo extrapolation method based on a space-time attention mechanism, and the method comprises the steps: firstly obtaining the radar echo data of a target region, carrying out the preprocessing of the obtained radar echo data, carrying out the sliding grouping of the preprocessed radar echo data, carrying out the data augmentation, and obtaining a radar echo sequence data set, dividing the data into a training set and a test set; a radar echo prediction network model based on a SimVP architecture is constructed, training and testing are carried out, and a multi-target loss function is adopted to supervise model training; and a future radar echo image is obtained through real-time prediction. Through the method, the dynamic evolution of complex weather phenomena such as storm can be more accurately captured, the precision and interpretability of short-time approaching weather forecast can be remarkably improved, more reliable technical support is provided for timely and accurate early warning of meteorological disasters, and the method has important practical application value.

The invention relates to the technical field of weather forecast, and discloses a wind speed forecast correction method, device and equipment and a readable storage medium, and the wind speed forecast correction method comprises the steps: carrying out the space-time alignment processing of multi-source wind speed data, and obtaining the wind speed alignment data corresponding to an observation station; constructing a corresponding initial spatial-temporal feature based on the wind speed alignment data, and performing standardization processing on the initial spatial-temporal feature to obtain a standardized spatial-temporal feature; inputting the standardized spatial-temporal characteristics into a prediction model to obtain predicted wind speed information; and generating a wind speed forecast correction result according to a dynamic fusion strategy based on the predicted wind speed information and the multi-source wind speed data. The accuracy and the stability of wind speed forecasting are remarkably improved, and the problem that the correction capability is insufficient under complex terrains and extreme weather is effectively solved.

This disclosure relates generally to system and method for estimating crop water requirement using multi-sensor data fusion. Increasing global population is imparting pressure on both agriculture for food demand and limited freshwater resources for consumption. Estimating crop water requirement reduces water demand for crop production. The method divides soil and crop into multiple vertical and horizontal profiles to estimate water balance thereby reducing the errors in estimation of crop water requirement. Additionally, the method has capability to interlink the multiple data sets such as satellite based earth observations, weather observations from IoT sensors, Weather forecasts from global circulation models, and crop knowledge base for crop water requirement estimation. The method is based on spatio-temporal modeling for multi-layer crop and soil water balance and helps to generate the additional insights on crop water requirement like moisture at different levels in soil profile, crop canopy growth at different locations.

The invention discloses a super-set deterministic weather forecast method and device based on machine learning, and the method comprises the steps: obtaining multi-source meteorological data of a target region, carrying out the meshing of the multi-source meteorological data, carrying out the historical static feature analysis and dynamic feature analysis of the meshing features, obtaining the feature weight of each mode, and carrying out the recognition of the multi-source meteorological data. The method comprises the following steps: constructing a prediction sub-model according to an extreme event, obtaining enhanced numerical prediction data, obtaining posterior probability distribution of grid points through a conditional generative adversarial network and a Bayesian neural network based on the numerical prediction data, a gridding feature and a feature weight, taking a maximum probability value as a deterministic weather forecast, and calculating a confidence interval. And obtaining a joint probability product including wind speed and rainfall joint distribution and the characteristic contribution degree. According to the method, numerical forecasting set products of different mode centers are utilized to fuse probability forecasting information, deterministic weather forecasting is obtained, smoothing of extreme events is reduced, deterministic maximum value output is provided, and meanwhile good interpretability is achieved.

The invention discloses a crop irrigation strategy real-time optimization method considering weather forecast. The method comprises the following steps: S1, acquiring historical meteorological data; s2, a localized AquaCrop crop growth model is constructed; s3, coupling the AquaCrop crop growth model and the NSGA-III multi-objective optimization algorithm model to obtain a coupling model, and solving to obtain a fixed irrigation strategy; s4, formulating optimization time periods, and dynamically integrating new all-season meteorological data in each optimization time period by adopting a DTW (Dynamic Time Warning) algorithm; and S5, correspondingly taking the fixed irrigation strategy obtained in the S3 as an irrigation strategy of the first optimization time period of the growing season according to time, taking various data of the optimization time period as input of the coupling model again to obtain a new irrigation strategy, applying the new irrigation strategy to the next optimization time period, and continuously optimizing. According to the method, the irrigation scheme can be adjusted in time in the growing season, and the method has important significance on guaranteeing grain safety and promoting rainfall flood resource utilization.

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 belongs to the technical field of meteorological monitoring, and provides a sky-sea intelligent eye meteorological perception system for correcting airspace weather forecast in real time based on an unmanned aerial vehicle. An air-sea-land intelligent coupling mechanism is established, meteorological foundation observation, an ocean circulation field and unmanned aerial vehicle detection data are fused in real time based on an ensemble Kalman filtering algorithm, and an efficient three-dimensional wind field reconstruction engine is constructed; a physical enhancement AI correction strategy is adopted, an LSTM-Transform hybrid model is used for driving forecast updating, and meanwhile, a gradient constraint mechanism is introduced to effectively inhibit non-physical mutation of a meteorological field. The final value of the system is reflected in deep coordination of airspace management and control, a risk thermodynamic diagram can be automatically generated, an obstacle avoidance path can be planned, and an air traffic management system is linked to trigger a control instruction. The comprehensive application of the system breaks through the limitation of a traditional method in temporal-spatial resolution, and provides high temporal-spatial resolution early warning support for scenes such as port scheduling and unmanned aerial vehicle logistics.

The invention discloses a complex terrain power transmission line icing thickness probability forecasting method based on a multi-mode set, and relates to the technical field of disaster prevention and reduction and weather forecasting of a power system. Constructing a plurality of icing thickness prediction models based on a physical mechanism and artificial intelligence based on the forecast data; and integrating the icing thickness prediction models into one icing thickness probability prediction method, and quantifying the uncertainty of icing prediction. Through the method, the accuracy and robustness of icing thickness prediction under complex terrain conditions can be effectively improved, a scientific basis with certainty and probabilistic is provided for power grid ice prevention and disaster reduction decisions, and thus the risk early warning and prevention and control capability of a power grid to deal with ice disasters is enhanced.

The invention discloses a satellite measurement and control remote sensing integrated task scheduling method and system, and mainly solves the problems of high failure rate of separation of measurement and control and remote sensing tasks, low scheduling efficiency and the like in the prior art. According to the implementation scheme, a low-orbit satellite network transmits a data packet generated by a measurement and control remote sensing task and receives the data packet through a ground station; the ground control center generates a task list; the ground station equipment generates equipment information and generates available forecast information through interaction with the low earth orbit satellite network; the ground control center integrates and preprocesses the equipment information and the available forecast information, calculates a task state vector, and represents the task state vector as a Deque double-end queue; performing task planning according to the double-end queue to obtain a legal forecast selected by the intelligent agent, resolving conflicts between the legal forecast and other available forecast, and updating states of other related tasks; and repeating the task planning step until the planning of all tasks is completed, and outputting a legal task scheduling plan. The method improves the task planning efficiency and the task completion rate, and can be used for environment monitoring, weather prediction, agricultural management, and disaster early warning and reconnaissance.

The invention relates to the technical field of communication management, in particular to a communication management method and system, and the method comprises the following steps: collecting overwater environment data in real time based on an overwater communication environment, obtaining weather forecast data in a future time period through weather forecast, extracting communication link node state data, and carrying out the data integration; and obtaining environment and link data. According to the invention, by predicting the weather and the communication link node state, communication faults can be timely identified and prevented, the continuity and stability of the communication link can be ensured, the reliability of the link can be evaluated by using the collected data, and the communication mode can be optimized according to the evaluation result. The whole communication system can automatically adjust and optimize the communication strategy when facing a complex and changeable environment, the control of the system on the communication state is enhanced through the real-time monitoring function, it is ensured that quick response and adjustment can be achieved when potential problems occur, and the communication quality is maintained.

The application discloses a kind of atmospheric environment ozone concentration prediction system based on chemical reaction mechanism, including obtaining weather forecast parameter, historical observation data cleaning analysis, analysis screening suitable model, model training and prediction, prediction result analysis, data visualization and so on functional module composition.This system is based on Lagrangian coordinate system, establishes the response relationship between meteorological parameter, pollutant concentration and ozone concentration, using thermodynamic equilibrium principle, chemical reaction equilibrium principle and photochemical reaction principle and other basic chemical reaction engineering theory, deduces the non-linear model of meteorological parameter, pollutant concentration and other parameters on ozone concentration influence, by the independent design of each software function module, with low energy consumption, low carbon emission, low operating cost, data automatic processing, easy to deploy, can be connected ecological cloud platform, system expansibility is strong, maintenance is easy, and computing efficiency is high.Therefore, it has broad market application prospect and good social and economic benefits.

The invention belongs to the technical field of agricultural remote sensing and monitoring, and provides an agricultural medium-short-term water shortage intelligent estimation method based on remote sensing, which comprises the steps of data collection and preprocessing, crop spatial distribution and current growth cycle extraction in a research area, medium-short-term meteorological element prediction based on a WRF model, and medium-short-term crop evapotranspiration prediction and medium-short-term agricultural water deficit dynamic intelligent estimation are realized. According to the method, the time-space continuous monitoring advantage of the remote sensing image is fully utilized, and rapid identification and early warning of the water shortage condition of the crops in a large range are achieved; meanwhile, by means of simulation of a numerical weather forecasting model on complex meteorological conditions, the future weather trend is effectively predicted, and a scientific basis is provided for agricultural water resource management.

The embodiment of the invention provides a severe convection weather forecasting method and system based on numerical mode diagnostic quantity. The method is applied to the technical field of weather prediction, and comprises the following steps: collecting initial field data, and forming an initial field capable of inputting numerical mode prediction after assimilation fusion; performing diagnostic quantity calculation based on the initial field data to obtain a convection diagnostic quantity, and constructing to obtain a fusion feature; collecting a radar echo combination reflection real-time condition, and performing calculation processing based on a radar echo extrapolation model to obtain a radar echo combination reflectivity extrapolation result; and constructing a weather identification model, carrying out identification processing on the fusion feature and radar echo combination reflectivity extrapolation result based on the model, outputting and obtaining a severe convection weather falling area identification result, and carrying out early warning prompt. In this way, weather forecast can be carried out more efficiently by fusing the numerical mode diagnostic quantity and the AI model, and the method has wide applicability.

The invention provides a distributed photovoltaic power prediction method and device and electronic equipment, and relates to the technical field of photovoltaic power generation. The method comprises the following steps: acquiring a weather forecast data sequence; for each weather type in a plurality of weather types, determining a correlation degree between the weather forecast data sequence and a comparison data sequence corresponding to the weather type, the plurality of weather types being obtained by clustering historical weather forecast data sequences based on a fuzzy C-means method; determining a target weather type corresponding to the weather forecast data sequence from the plurality of weather types according to all the correlation degrees; and inputting the weather forecast data sequence into a power prediction model corresponding to the target weather type to obtain distributed photovoltaic power prediction data, the power prediction model being obtained by pre-training based on an iTransform model. The method is used for improving the precision and model adaptability of distributed photovoltaic power prediction.

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 discloses a weather prediction method, device and system based on federated learning, and belongs to the technical field of weather prediction, the weather prediction method based on federated learning combines a Transform model with a KAN model, and uses the advantages of the Transform model, the KAN model and a federated learning framework to predict the weather prediction. The provided first weather prediction model constructed on the basis of the Transform-KAN can have good data fitting performance. According to the method, the weather data are classified, then the weather data are placed under a federated learning framework, the automatic weather stations distributed at all places can carry out distributed collaborative learning, global features and space-time dependency relationships in the weather data can be well learned, and better data fitting ability and generalization performance can be achieved while data privacy is guaranteed.

The invention discloses a japonica rice high-yield cultivation method based on temperature and light resources, and relates to the technical field of crop precise cultivation, and the method comprises the steps: calculating a TPU increment value based on field real-time temperature and light data, obtaining a historical TPU value data set, carrying out TPU growth prediction in combination with future weather forecast data, and generating a phenological period prediction instruction; based on the phenological period prediction instruction, identifying a phenological period time interval set, extracting meteorological data from future weather forecast data according to the phenological period time interval set, and generating a phenological period meteorological risk data set; and calculating the risk probability of low-temperature cold damage and high-temperature heat damage according to the phenological period meteorological risk data set, generating a meteorological risk probability set, carrying out risk early warning judgment according to the meteorological risk probability set, and generating a third-level risk early warning instruction signal. According to the invention, through generating the three-level risk early warning instruction signal, active and graded response to potential disasters is realized, and the accuracy and timeliness of disaster prevention and control are enhanced.

The invention discloses a planar turbulence prediction method based on a hybrid prediction model, and relates to the technical field of cross of computational fluid mechanics and artificial intelligence, and the method comprises a sensor point selection method based on a multi-index assignment strategy, and is used for maximizing information acquisition from sparse deployment. A convolutional long short-term memory network, a multi-layer perceptron and a mixed deep learning model of a Fourier neural operator are fused, Conv-LSTM and MLP are used for capturing local nonlinear spatial-temporal features, and FNO is used for modeling global dependence and correcting boundary errors; a physically constrained training framework ensures the physical rationality of a prediction result by introducing undivergence,-5 / 3 energy spectrum slope and other constraint terms into a loss function; under the condition that only a small number of measuring points are used for input, the plane turbulent wind field can be reconstructed in real time in a high-precision and high-fidelity mode, prediction errors are remarkably reduced, the boundary effect problem is solved, the physical credibility of prediction results is guaranteed, and the method has important application value in the fields of wind engineering, weather forecast and the like.

The invention relates to the technical field of greenhouse temperature control, and discloses a crop growth controllable agricultural greenhouse intelligent environment control system and method based on big data, and the system comprises a central processing unit which is used for processing data and coordinating the operation of all modules; the weather forecast system comprises a data extraction module, a data processing module and a data analysis module, the data extraction module is used for acquiring local weather forecast information in real time, and the data processing module performs standardization processing on weather information and extracts key parameters; and the data analysis module predicts a cold current cooling time node in combination with historical meteorological data. According to the invention, weather forecast information sent by a local weather forecast system is processed and analyzed, key information is obtained, a time node during cold current cooling is obtained, and a geothermal line or an air heater is started in advance before the node to be cooled arrives, so that the geothermal line or the air heater works at a low temperature for preheating; power can be conveniently and rapidly increased to achieve the heat preservation purpose.

The invention discloses a rainfall short-term and imminent forecasting method and device based on GNSS-PWV and numerical weather forecast fusion, and the method comprises the steps: calculating the real-time atmospheric precipitable water amount measured through GNSS signal delay, and obtaining the meteorological observation data and numerical weather forecast data of a base station; respectively constructing a physical forecasting model and a machine learning forecasting model through the atmospheric precipitable water amount, the variable quantity of the atmospheric precipitable water amount, the base station meteorological observation data and the numerical weather forecasting data, and respectively forecasting the precipitation probability and the precipitation amount by utilizing the physical forecasting model and the machine learning forecasting model; and a hybrid forecasting model is constructed, the model fuses the precipitation probability and precipitation forecast by the physical forecasting model and the machine learning forecasting model, and the final precipitation probability and final precipitation are obtained. According to the method, on the basis of traditional numerical weather forecast, time-varying information of GNSS-PWV is intelligently fused, and adaptive optimization is realized by dynamically adjusting the weights of a physical model and a data driving model, so that the precision and reliability of short and temporary rainfall forecast are remarkably improved.