52 results about "Future climate" patented technology

The invention discloses a sudden snow disaster risk degree probability prediction method oriented to insufficient historical time sequences. The method comprises the following steps: step 1), constructing a multi-index risk index based on related data such as historical snow disasters; 2) performing time sequence modeling on the violent snow disaster risk index by adopting an autoregressive integral moving average model, and capturing trend and periodic characteristics of the violent snow disaster risk index; 3) discretizing a continuous risk index of modeling production into a plurality of risk levels, and constructing a state transition probability matrix; step 4), introducing space neighborhood influence and carrier exposure constraint, and constructing a Markov space weighted state transition model; 5) coupling CMIP6 climate scene data, and dynamically adjusting future state transition probability, and 6) fusing ARIMA trend prediction and Markov state transition probability, and outputting future multi-period snowstorm disaster risk level probability distribution.According to the method, the accuracy of snowstorm disaster risk prediction and the modeling ability and prediction adaptability of future climate change scenes are improved.

The invention provides a drought and flood sudden change prediction method and system based on artificial intelligence, and the method comprises the steps: collecting the in-out reservoir runoff observation data of a reservoir in a target region, obtaining a restored natural reservoir runoff series based on a water balance method, calibrating a long-short-term memory model through a minimum batch gradient descent method, and carrying out the prediction of the drought and flood sudden change. Based on the natural reservoir runoff, the actual reservoir runoff and meteorological data, constructing a long-short-term memory model to simulate the influence of water conservancy project regulation and storage on the runoff; based on an earth system mode set and a multivariable deviation correction method, performing spatial downscaling on output data of earth system modes to obtain corrected meteorological variables; driving a multi-member set of an earth system mode, and separating by adopting a detection attribution technology to obtain a contribution rate of man-made climate compulsion to drought and flood sudden turning event evolution; and predicting a daily runoff process under future climate change, and predicting a future drought and flood sudden turning event by adopting deep learning and superposition of regulation and storage influence of a water conservancy project.

The invention discloses an alpine grassland livestock capacity monitoring and predicting method and system based on climate change. The alpine grassland livestock capacity monitoring and predicting method based on climate change comprises the following steps: 1) modeling data preprocessing: acquiring grassland type data and meteorological factor data, and carrying out space-time mask fusion, year / month scale space-time alignment and dynamic standardization to obtain grassland type standardized data; 2) feature screening and XGBoost modeling: performing feature importance screening and multi-parameter grid search optimization on grassland type standardized data by using a random forest model, and establishing and verifying a model; 3) preprocessing future CMIP6 climate factor data; and (4) predicting the livestock carrying capacity by using the results of the steps (2) and (3). According to the method, the potential influence of the climate change on the alpine grassland ecosystem can be evaluated on the regional scale, the variation trend of the future livestock capacity is accurately predicted, and a scientific basis is provided for formulating pasturing area management, ecological restoration and climate adaptability strategies.

The invention relates to a flood and drought disaster risk assessment and deduction system and method based on slope multi-level regulation and storage, and belongs to the technical field of flood and drought disaster risk prevention and control and watershed hydrological regulation and control. The method comprises the following steps: acquiring multi-source data of a target drainage basin, performing water source refined analysis through coupling of water transfer coefficients (K1 and K2) based on a hierarchical synergistic effect of'upper closure-middle retention-lower dissipation 'of a slope, and calculating a controllable, effective and renewable water source amount; in combination with the regulation and storage capacity difference of each level, carrying out flood and drought disaster chain risk collaborative assessment, and outputting hierarchical risks and 0-10 level set risk indexes; a physical mechanism and data driving coupled scene deduction model is driven, and historical extreme event redisk and future climate scene prediction are completed; and after rendering by a visual engine, through hierarchical slope regulation and storage threshold comparison, outputting a regulation and storage measure layout optimization scheme. The system adopts a micro-service architecture and comprises a data management module, a multi-level regulation and storage collaborative analysis module, a water source analysis module, a risk evaluation module, a scene deduction module, a decision support module and a threshold comparison module. According to the method, high-resolution accurate positioning, low-error simulation and efficiency improvement of flood and drought disaster risks are realized, a decision closed loop from risk identification to optimization regulation is formed, and direct and quantitative technical support can be provided for flood and drought prevention and control engineering practice of drainage basins.

The application discloses a simulation method and system for engineering geological survey in alpine permafrost regions based on digital twinning, and relates to the technical field of engineering geological survey.The method comprises the following steps: obtaining multi-source heterogeneous data to construct a digital base; constructing a thermal-water-force-fracture multi-field coupling physical model based on the phase field theory; constructing a generative adversarial network as a time reversal engine to learn the mapping from the current state to the past state; inputting the current measured state into the time reversal engine to generate a historical evolution sequence, and comparing the historical evolution sequence with actual geological relics to dynamically calibrate model parameters; loading future climate scenarios to perform forward prediction, and fusing monitoring data through Kalman filtering to form a closed-loop corrected digital twin; and outputting a risk thermodynamic map, a fracture connectivity atlas and an engineering warning level based on the prediction result.The method can realize dynamic calibration and real-time early warning of the engineering geological model in the permafrost region, and improve the accuracy and calculation efficiency of the prediction of engineering geological disasters in cold regions.

The invention discloses a multi-lake system low-carbon-ecology collaborative scheduling method and system adapting to extreme climate. The method comprises the following steps: constructing a future extreme climate scene set; constructing an urban-natural drainage basin coupling hydrological model, and obtaining multiple groups of lake inflow runoff sequences driven by various future extreme climate scenes according to the urban-natural drainage basin coupling hydrological model; taking minimization of annual total diversion carbon emission as a single target, and constructing a low-carbon-ecological collaborative optimization model according to a future extreme climate scene set and an urban-natural drainage basin coupling hydrological model; inputting each group of into-lake runoff sequences into a low-carbon-ecological collaborative optimization model for solving to obtain a plurality of groups of candidate scheduling schemes; and screening the candidate scheduling schemes to obtain an optimal scheduling scheme. The method can fully consider the uncertainty of future climate, improves the reliability of a scheduling scheme, solves the problem of difficulty in decision making of a multi-target model, and can be widely applied to the technical field of water resource scheduling.

The present application discloses a wetland vegetation carbon sequestration potential prediction method and system under the influence of future climate change, belongs to the technical field of wetland vegetation carbon sequestration potential prediction, and comprises the following steps: obtaining the net primary productivity of vegetation, the vegetation coverage, the monthly temperature, the precipitation and the atmospheric carbon dioxide concentration in the historical period and the future period; calculating the annual average net primary productivity of vegetation, the coverage and the multi-year average vegetation coverage; calculating the annual average temperature, the precipitation and the carbon dioxide concentration; extracting the wetland vegetation pixel distribution as the research area range; calculating the ratio of the annual average net primary productivity to the vegetation coverage to obtain the wetland vegetation carbon sequestration potential value in the historical period; adopting the multiple stepwise regression analysis method to build a per-pixel wetland vegetation carbon sequestration potential estimation model; and utilizing the annual temperature, the precipitation and the carbon dioxide concentration in the future period to realize the accurate prediction of the wetland vegetation carbon sequestration potential in the research area range in the future period.

The invention discloses an artificial wetland space optimization arrangement method and system, a non-point source pollution model is constructed based on a hydrological response unit, target meteorological data, target pollution data and an SWAT model, and the simulation precision of the non-point source pollution process can be improved. Climate scene data is constructed through future climate data, a preset path rule and a simple quantile algorithm, and the influence of climate change on pollution load can be reflected. Furthermore, a wetland optimization model is constructed based on a non-point source pollution model and a target function, the scientificity of an optimization strategy is enhanced, candidate wetland areas are identified by a spatial analysis algorithm based on a complex terrain index CTI, and the maximum potential area of each candidate wetland area is counted, so that efficient identification of potential arrangement areas can be realized. The constructed wetland optimization regulation and control scheme is generated based on the preset genetic algorithm, and the constructed wetland space optimization arrangement is performed according to the scheme, so that the optimization and practicability of the constructed wetland layout can be realized.

The invention discloses an ecological system productivity estimation method for predicting the optimal vegetation temperature based on LSTM, and relates to the technical field of ecological information, and the method comprises the steps: obtaining historical weather, terrain and future climate data; constructing a spatial neighborhood of a target pixel, quantifying climate connectivity cost based on terrain and a wind field, and generating anisotropic spatial features; resolving a dynamic space-time confidence factor representing the microclimate complexity; inputting the anisotropic spatial features and the historical temperature into a long short-term memory network, regulating and controlling fusion weight through a dynamic gating mechanism by using confidence factors, and predicting the optimal vegetation temperature; constructing a decoupling model containing thermal kinetic energy and nonlinear thermal stress terms, and calculating productivity based on a predicted value and future temperature; according to the method, the microclimate heterogeneity and the vegetation thermal adaptation mechanism can be accurately depicted, and the estimation precision of the productivity of the ecological system under the complex climate scene is effectively improved.

The invention provides a crop irrigation and fertilization decision-making method based on probability constraint. The crop irrigation and fertilization decision-making method comprises the steps of obtaining multi-source input data, and collecting key growth index data of crops through field trials. A Markov chain is adopted to simulate distribution and correlation of different variables in meteorological historical observation data, a day-by-day meteorological sequence is generated, and the day-by-day meteorological sequence is used for reflecting the natural variation rate and uncertainty of future climate conditions. A to-be-optimized water and fertilizer management variable set is set, the crop DSSAT model is operated in each meteorological scene, and constraint conditions are constructed; and under the constraint condition, carrying out global optimization on the target function by adopting a multi-target optimization algorithm to obtain a Pareto optimal solution set. By introducing a multi-scene meteorological simulation method and combining a crop DSSAT model and a multi-objective optimization algorithm, the adaptive capacity and robustness of a crop management scheme under actual climate conditions are improved, and the scientificity and systematicness of agricultural management decisions are further improved.

This application discloses a method, apparatus, equipment, medium, and product for predicting regional crop water productivity, relating to the field of smart agriculture. The method includes inputting historical and future climate data into an ecosystem process model to obtain simulated net primary productivity and simulated evapotranspiration for historical and future times; using a bias mapping method based on the simulated evapotranspiration, actual evapotranspiration, and simulated evapotranspiration for future times to obtain predicted evapotranspiration for future times; training multiple machine learning models based on the simulated net primary productivity, agricultural management data, and grain yield per unit area for historical times to obtain a crop yield prediction model; obtaining the predicted grain yield per unit area for future times based on the crop yield prediction model; and obtaining the regional crop water productivity based on the predicted evapotranspiration and the predicted grain yield per unit area for future times. This application can achieve high-precision, high spatiotemporal resolution, and high-reliability prediction of regional crop water productivity.

The application relates to a water resource pressure assessment method integrating social economy and climate change. It relates to the technical field of water resource pressure assessment in arid river basins. The method comprises the following steps: obtaining historical values and predicted values of social economic data, correcting the predicted values of the social economy to obtain corrected predicted values; calculating potential evapotranspiration in different scenarios based on historical scenario meteorological data and corrected future climate mode data; calculating future basin water resources based on the correlation between basin rainfall and water resources; extracting drought intensity characteristics according to the change trend of the historical scenario and the future scenario basin drought index; constructing a basin water resource pressure index calculation formula; and according to the calculated water resource pressure index of the water resource region, standardizing the index to obtain a standardized index, and evaluating the change trend of the water resource pressure of the region in different scenarios according to the standardized index. The method solves the problem of quantification and evaluation of basin water resource pressure under complex conditions.

The invention describes a method by which it is possible to predict past and future climate changes by measuring changes in local and global magnetic fields on the Earth's surface in time and space and comparing them with older measurements and effects.

This invention discloses a method for assessing ecosystem service losses due to regional climate change. The region includes counties, cities, or important ecological function zones. The steps are as follows: S1. Collect remote sensing data and preprocess the data through band combination, geometric correction, and projection transformation. Select a standard color composite scheme, establish interpretation markers, and perform visual interpretation. S2. Calculate and statistically analyze the unit area yield value and equivalent factor of grain crops. The equivalent factor is 1 / 7 of the average market value of grain in the current year. S3. Statistically analyze land use types, calculate the magnitude of land use change, reflecting the changes in the total amount of different land use types and the overall trend of land use type changes. S4. Select vegetation cover as an indicator, and adjust the ecosystem service value based on two parameters: net primary productivity and vegetation cover. This invention is of great significance for formulating ecosystem management strategies under the background of future climate change.

The invention discloses a man-made forest ecological function monitoring method based on meteorology and hydrology, and relates to the technical field of forestry ecological monitoring, and the method comprises the steps: obtaining the soil penetration resistance time sequence data of different soil layers of a target man-made forest area, calculating the soil physical structure toughness index of the soil maintenance pore structure and the compaction resistance through combining with soil auxiliary parameters, and calculating the soil physical structure toughness index; analyzing the historical evolution trend; collecting short-term meteorological data and long-term climate change data of the target artificial forest area, coupling the short-term meteorological data and the long-term climate change data with the soil physical structure toughness index, and establishing a meteorological-hydrological-soil structure coupling process model; quantitatively evaluating the water source conservation capability, the soil erosion risk and the climate adaptability of the artificial forest area in the target artificial forest area, and simulating the future evolution trajectory of the soil physical structure and the ecological function of the target artificial forest area by taking different climate scenes and management intervention scenes in the future as compensation factors; meteorological and hydrological dynamic monitoring and early warning of the ecological function of the artificial forest are realized. According to the invention, the accuracy of ecological function evolution prediction is improved.

The present application belongs to the technical field of geochemistry, and particularly relates to the estimation of the carbon flux of the earth system, and especially relates to a method and system for estimating the carbon flux of the deep-time earth system. The method for estimating the carbon flux of the deep-time earth system provided by the present application solves the four defects in the traditional technology caused by the fixed carbon pool assumption, the neglect of the atmospheric / biosphere carbon content, the isotopic error, and the failure to consider the fractionation effect through dynamic carbon pool segmentation, the inclusion of multi-sphere isotopic composition, and the correction of fractionation effect. The high-precision and verifiable model framework not only improves the scientificity of the deep-time carbon cycle research, but also provides a key tool for analyzing the climate evolution mechanism in the geological history period and predicting the future climate change trend, and has important theoretical innovation and practical application value.

The invention relates to the technical field of hydro meteorology, in particular to a method and a system for predicting and tracing extreme hydrological events in a reservoir area. The method comprises the following steps of S1, collecting meteorological and hydrological data, a global climate mode data set, geographic space data and stable isotope data of a reservoir area, and unifying a multi-source data scale through a time sequence fusion algorithm; according to the method, isotope data is adopted as an independent evidence constraint model, and compared with a traditional single model method, the method can more comprehensively capture a complex mechanism of occurrence of an extreme event, so that a simulation result is closer to an actual situation, and countermeasures can be taken in the face of an extreme climate event; the climate prediction and the isotope traceability technology are coupled for the first time, how causes of the extreme events change under different future climate scenes can be pre-judged in advance, and the research method and the model framework are not only suitable for specific areas, but also can be widely applied to research of the extreme events of large reservoirs and drainage basins.

The invention discloses a building demand response strategy evaluation and optimization method considering a future climate scene. The method comprises the following steps: dynamically constructing meteorological data under the future climate scene of an evaluation region; acquiring load change characteristics of the building under different climate conditions; constructing a demand response scene parameter space, and forming a demand response'variable-performance index 'data set by adopting a Latin hypercube sampling and batch simulation method; constructing a machine learning agent model to quickly predict the building load response, and optimizing the hyper-parameters of the building load response by using an NSGA-II (Non-dominated Sorting Genetic Algorithm-II) algorithm; the load adjustable capacity of the building demand response strategy in different future climate scenes is evaluated, the relative change rate is calculated, and the future climate adaptability of the building load flexible regulation and control strategy is determined; and establishing a sensitivity analysis method to identify key variables, and providing a decision basis for early-stage design and later-stage operation optimization of a building demand response strategy. The method provides important support for planning design and operation regulation and control of flexible buildings and novel energy systems.

This invention discloses a method for hydropower prediction under climate change that integrates machine learning, belonging to the field of hydropower data management technology. It includes: simulating the hydropower response under future climate change using a finite number of representative samples in a target region through a physical process-based hydrological-reservoir-hydropower coupling model; constructing a training dataset containing complex nonlinear response characteristics while ensuring the reliability of the model's physical principles and results; then, using machine learning algorithms to establish an efficient generalization model between key climate change features and hydropower generation; finally, applying this model to a larger sample of future climate prediction data covering a wider range of climate change scenarios to achieve rapid and robust hydropower output prediction and uncertainty quantification. This provides more robust and reliable technical support for hydropower prediction under climate change scenarios, as well as for uncertainty quantification and risk decision-making.

The invention discloses an intelligent identification method for a service trade-off cooperative relationship of a high and cold ecosystem, and the method comprises the steps: obtaining original multi-source heterogeneous data, and carrying out the preprocessing of the data, and obtaining a spatio-temporal data set; constructing a meta training task set based on the spatio-temporal data set so as to train a meta learner, generating corresponding optimal model parameters through the trained meta learner according to a future climate-land utilization scene, inputting the optimal model parameters into a corresponding model for chain simulation, and outputting future multi-time-sequence ecosystem service supply spatial distribution data; according to the space-time diagram convolution network model, obtaining a relation intensity space distribution diagram and further identifying a trade-off cooperative relation, generating a space activation thermodynamic diagram through a gradient weighting class activation mapping technology, and after the space activation thermodynamic diagram is fused with the relation intensity space distribution diagram, determining a key area space distribution diagram according to a quantitative decision rule. And performing rule matching in combination with the ecological management knowledge base, and determining a differentiated ecological management scheme. According to the invention, the recognition accuracy and efficiency can be improved.

The invention provides an urban power system toughness improvement method and device based on vehicle network interaction, and the method constructs a two-stage stochastic optimization model with the minimum total cost of the system as a target, and covers the long-term planning and operation scheduling of the urban power system in a disaster scene. The method can improve the toughness of an urban power system in an extreme weather long-term scene, achieves the collaborative optimization of multi-dimensional costs such as power generation capacity expansion, line reinforcement, operation maintenance and load reduction, effectively reduces the comprehensive cost, and enhances the dynamic response capability of the system. A fusion channel of vehicle-network interaction and urban power grid toughness planning is opened, and a technical path capable of being popularized is provided for safe, flexible and economical operation of a power system under future climate uncertainty.

The invention relates to the technical field of photovoltaic generating capacity prediction. The invention provides a regional photovoltaic annual generating capacity prediction method and system fused with a climate change trend. The method comprises the following steps: constructing and preprocessing a multi-source data set including historical power generation, historical meteorological observation, climate system mode output and photovoltaic module performance parameters; extracting trend components and volatility components of climate key factors in a future year, and generating a multi-path climate change scene; establishing a component performance attenuation-climatic factor dynamic response model, and outputting a photovoltaic component equivalent power generation efficiency sequence considering a climatic change cumulative effect; based on photovoltaic power station space distribution and landform data, a region-level power generation capacity space-time aggregation model is constructed, and through scale conversion and probability aggregation, a prediction result and a confidence interval of photovoltaic annual power generation capacity of a target region are output. The problem that the long-term prediction deviation is large due to the fact that an existing photovoltaic generating capacity prediction method does not fully consider the future climate change trend influence is solved.

The present application relates to the field of energy simulation, and discloses a method and device for predicting urban building energy consumption, electronic equipment and medium. The method comprises: reconstructing each component in the urban climate state of the previous time step by mapping to determine the building boundary state; determining the energy consumption state quantity of the building according to the building boundary state; updating the urban climate state of the current time step according to the energy consumption state quantity, thereby predicting the urban building energy consumption of each time step. The present application can accurately depict the evolution law of urban building energy consumption under future climate change scenarios.

The application provides a regional comprehensive energy system double-layer optimization method and system, the method comprising the following steps: calculating future meteorological data according to historical meteorological data and future prediction data; combining the future meteorological data to perform load simulation on a typical building to obtain a typical building annual dynamic load curve, and obtaining a regional typical daily dynamic load curve through accumulation and clustering; constructing a mathematical model of the regional comprehensive energy system; establishing an upper-layer capacity planning model and a lower-layer operation optimization model based on the mathematical model; determining power balance constraints by the regional typical daily dynamic load curve for the lower-layer operation optimization model; solving the model by using a non-dominated sorting genetic algorithm and an interval linear programming method; and finally determining an optimal scheme by using an entropy weight superior and inferior solution distance method. Based on the method, the application further provides a regional comprehensive energy system double-layer optimization system. The application considers the influence of future climate change and uncertainty on regional load prediction and regional comprehensive energy system planning, so that the planning result is more reasonable.

The invention provides a rape planting simulation method and device responding to and adapting to climate changes and a storage medium. The method comprises the steps that multi-source data of a target area is acquired, the multi-source data comprises climate data, soil data and historical rape crop data, and the climate data comprises historical observation data and future climate scene data; according to the soil data and historical rape crop data, optimizing model parameters of the WOFOST model by adopting a Bayesian optimization algorithm to obtain optimized model parameters; based on the climate data, obtaining changed climate data, and according to the changed climate data and the optimized model parameters, simulating the response of the rape in a climate change scene through a WOFOST model to obtain a simulation result; according to the simulation result and the soil data, an adaptive measure optimization scheme of the rape is determined, and the adaptive measure optimization scheme comprises the sowing time adjustment days and variety characteristic parameters. Therefore, the uncertainty of climate change can be captured, and the accuracy of a simulation result is improved.

The invention discloses a sponge city design method and system, a terminal and a storage medium, and the design method comprises the steps: obtaining second multi-source data of a target region, the second multi-source data comprising future predicted rainfall data, topographic data, ground coverage data and drainage system data; building future climate scene simulation of the target area according to the second multi-source data to generate a future scene ponding distribution map; based on the historical scene ponding distribution diagram and the future scene ponding distribution diagram, constructing a drainage layout optimization first model; and solving the drainage layout optimization first model to output an optimal design scheme. According to the method, historical scene ponding simulation and future scene ponding simulation are realized by fusing the historical rainfall data and the future predicted rainfall data, so that future climate changes are coped, capacity expansion requirements are planned in advance, later reworking is avoided, and a feasible technical scheme is provided for rainfall flood management of a sponge city.

The invention provides a flood risk prediction method and system based on the influence of water conservancy project regulation and storage. The flood risk prediction method comprises the steps of collecting out-and-in reservoir runoff observation data of a reservoir in a target area, obtaining a reduced natural reservoir runoff series based on a water balance method, calibrating a random forest model by adopting a Temperature Scaling method, and simulating to obtain natural reservoir runoff; a long-short-term memory model is constructed to simulate the influence of water conservancy project regulation and storage on runoff, and a minimum batch gradient descent method is adopted for parameter calibration; obtaining corrected meteorological variables based on the earth system mode set and a multivariable deviation correction method; adopting a multi-member set output data of CESM-LE to drive a calibrated random forest model, and adopting a detection attribution technology to separate to obtain the contribution rate of artificial climate compulsion to flood evolution; and predicting a daily runoff process under future climate change, and predicting a future flood event by adopting deep learning and superimposing the regulation and storage influence of a water conservancy project.

The invention belongs to the technical field of agricultural engineering and climate change influence evaluation, and particularly relates to a grain drying energy consumption prediction method based on a CMIP6 future scene, which is suitable for energy consumption evaluation of drying of main grain varieties such as rice, corn and wheat in various regions under future climate change conditions. The method comprises the following specific steps: establishing an energy consumption multiple regression model taking air temperature, rainfall, wind speed and dimension as input variables; and inputting CMIP6 data into the model to obtain an energy consumption prediction sequence in 2020-2100 years, and generating time trend and spatial distribution results under different SSP scenes. The method can quantitatively evaluate the long-term influence of the climate change on the grain drying energy demand, is suitable for agricultural production planning, energy configuration and climate change adaptability evaluation, and has the advantages of clear structure, high precision, strong applicability and the like.

The application provides a high-precision nitrogen and phosphorus simulation and prediction method based on quantile mapping and coupling of a distributed hydrological model SWAT, and belongs to the technical field of environmental science and engineering, and comprises the following steps: firstly, obtaining multi-source data of a research basin and preprocessing, then obtaining CMIP6 future climate data and performing downscaling processing, then constructing a multi-factor joint scenario system, simulating and calculating the load by using the SWAT, quantifying uncertainty and identifying risk, and finally comprehensively expressing to form nitrogen and phosphorus flux and risk distribution results. The high-precision nitrogen and phosphorus simulation and prediction method based on quantile mapping and coupling of the distributed hydrological model SWAT solves the problems of insufficient precision of existing climate data, low coupling degree of water quality models and difficulty in quantifying uncertainty of prediction results.

The application provides an atmospheric data assimilation method, device, equipment and storage medium. It relates to the technical field of computer science and atmospheric science. The method comprises the following steps: adding a cost function, a prior estimate and a gradient of the cost function into a gradient-based descent algorithm, adding a posterior estimate, and constructing a four-dimensional variational and ensemble Kalman filter combination model based on Bayesian prior information; and based on the four-dimensional variational and ensemble Kalman filter combination model, data assimilation is performed on atmospheric data. The application combines artificial intelligence technology, especially deep learning and Bayesian optimization adaptive algorithm, enhances and optimizes the traditional data assimilation algorithm, and can be widely applied to the prediction of extreme weather events such as high temperature and heat wave. Through the assimilation of historical data and real-time observation data, the prediction accuracy of future climate change trends is significantly improved, providing stronger support for extreme weather warning and decision making.