The phenomenological and physically based models,using the true stress–true strain curve data obtained under various hot working conditions of 850–1200°C and 0.001–10 s−1,were developed and improved for AerMet...The phenomenological and physically based models,using the true stress–true strain curve data obtained under various hot working conditions of 850–1200°C and 0.001–10 s−1,were developed and improved for AerMet 100 ultra-high strength steel.The predictability of the developed constitutive models was verified and compared.The determination coefficient and average absolute relative error were 0.9988 and 3.72%for the improved version of the modified Zerilli–Armstrong model,0.9985 and 3.96%for the improved version of the modified Johnson–Cook model,0.9947 and 4.59%for the strain-compensated Arrhenius-type model and 0.9913 and 5.43%for the improved Khan–Huang–Liang model,respectively.The results showed that the improved versions of the modified Zerilli–Armstrong model have the best predictability among the studied constitutive models.Comparing the predictability before and after the improvement,the average absolute relative error was increased by 65.14%for the modified Zerilli–Armstrong model and 58.45%for the modified Johnson–Cook model.This indicates that the phenomenological improvement of physically based constitutive models allows us to develop effectively constitutive equations with high prediction accuracy.展开更多
Swarm Intelligence (SI) is a collective behavior that emerges from interaction between individuals in a group. Typical SI includes fish schooling, ant foraging, bird migration, and so on. A great deal of models have b...Swarm Intelligence (SI) is a collective behavior that emerges from interaction between individuals in a group. Typical SI includes fish schooling, ant foraging, bird migration, and so on. A great deal of models have been introduced to characterize the mechanism of SI. This article reviews several typical models and classifies them into four categories: self-driven particle models, with Boids model as the primary example;pheromone communication models, including the ant colony pheromone model which serves as the foundation for ant colony optimization;leadership decision models, utilizing the hierarchical dynamics model of pigeon flock as a prime instance;empirical research models, which employ the topological rule model of starling flock as a classic model. On this basis, each type of model is elaborated upon in terms of its typical model overview, applications, and model evaluation. More specifically, multi-agent swarm control, path optimization and obstacle avoidance, formation and consensus control, trajectory tracking in the dense crowd and social networks analysis are surveyed in the application of each category, respectively. Furthermore, the more precise and effective modeling techniques for leadership decision and empirical research models are described. Limitations and potential directions for further exploration in the study of SI are presented.展开更多
Peridynamics(PD)is an effective method for simulating the spontaneous initiation and propagation of tensile cracks in materials.However,it faces great challenges in simulating compression-shear cracking of geomaterial...Peridynamics(PD)is an effective method for simulating the spontaneous initiation and propagation of tensile cracks in materials.However,it faces great challenges in simulating compression-shear cracking of geomaterials due to the lack of efficient contact-friction models.This paper introduces an original contact-friction model that leverages twin mesh and potential function principles within PD to model rock cracking under tensile and compressive stresses.The contact detection algorithm,based on space segmentation axis-aligned bounding box(AABB)tree data structure,is used to address the significant challenge of highly efficient contact detection in compression and shear problems.In this method,the twin mesh and potential function are utilized to quantify contact detection and contact degree,as well as friction behavior.This is in contrast to the distance and circular contact area model,which lacks physical significance in the classical PD method.As demonstrated by the tests on specimens containing cracks,the proposed model can capture 8 types of secondary fractures,reduce the contact detection error by about 29%e56%,and increase the contact retrieval efficiency by over 1600 times compared to the classic PD models.This significantly enhances the capability of PD to simulate the initiation,expansion,and coalescence of intricate compression-shear cracks.展开更多
In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes tech...In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes technical exchanges and learning globally.Second,resources required for large model R&D are difficult for a single institution to obtain.The evaluation of general large models also requires the participation of experts from various industries.Third,without open source collaboration,it is difficult to form a unified upper-layer software ecosystem.Therefore,open source has become an important cooperation mechanism to promote the development of AI and large models.There are two cases to illustrate how open source and international standards interact with each other.展开更多
Numerical models are crucial for quantifying the ocean-atmosphere interactions associated with the El Niño-Southern Oscillation(ENSO)phenomenon in the tropical Pacific.Current coupled models often exhibit signifi...Numerical models are crucial for quantifying the ocean-atmosphere interactions associated with the El Niño-Southern Oscillation(ENSO)phenomenon in the tropical Pacific.Current coupled models often exhibit significant biases and inter-model differences in simulating ENSO,underscoring the need for alternative modeling approaches.The Regional Ocean Modeling System(ROMS)is a sophisticated ocean model widely used for regional studies and has been coupled with various atmospheric models.However,its application in simulating ENSO processes on a basin scale in the tropical Pacific has not been explored.For the first time,this study presents the development of a basin-scale hybrid coupled model(HCM)for the tropical Pacific,integrating ROMS with a statistical atmospheric model that captures the interannual relationships between sea surface temperature(SST)and wind stress anomalies.The HCM is evaluated for its capability to simulate the annual mean,seasonal,and interannual variations of the oceanic state in the tropical Pacific.Results demonstrate that the model effectively reproduces the ENSO cycle,with a dominant oscillation period of approximately two years.The ROMS-based HCM developed here offers an efficient and robust tool for investigating climate variability in the tropical Pacific.展开更多
Architecture framework has become an effective method recently to describe the system of systems(SoS)architecture,such as the United States(US)Department of Defense Architecture Framework Version 2.0(DoDAF2.0).As a vi...Architecture framework has become an effective method recently to describe the system of systems(SoS)architecture,such as the United States(US)Department of Defense Architecture Framework Version 2.0(DoDAF2.0).As a viewpoint in DoDAF2.0,the operational viewpoint(OV)describes operational activities,nodes,and resource flows.The OV models are important for SoS architecture development.However,as the SoS complexity increases,constructing OV models with traditional methods exposes shortcomings,such as inefficient data collection and low modeling standards.Therefore,we propose an intelligent modeling method for five OV models,including operational resource flow OV-2,organizational relationships OV-4,operational activity hierarchy OV-5a,operational activities model OV-5b,and operational activity sequences OV-6c.The main idea of the method is to extract OV architecture data from text and generate interoperable OV models.First,we construct the OV meta model based on the DoDAF2.0 meta model(DM2).Second,OV architecture named entities is recognized from text based on the bidirectional long short-term memory and conditional random field(BiLSTM-CRF)model.And OV architecture relationships are collected with relationship extraction rules.Finally,we define the generation rules for OV models and develop an OV modeling tool.We use unmanned surface vehicles(USV)swarm target defense SoS architecture as a case to verify the feasibility and effectiveness of the intelligent modeling method.展开更多
We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpr...We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpreting these parameters is crucial for effectively exploring and developing oil and gas.However,with the increasing complexity of geological conditions in this industry,there is a growing demand for improved accuracy in reservoir parameter prediction,leading to higher costs associated with manual interpretation.The conventional logging interpretation methods rely on empirical relationships between logging data and reservoir parameters,which suffer from low interpretation efficiency,intense subjectivity,and suitability for ideal conditions.The application of artificial intelligence in the interpretation of logging data provides a new solution to the problems existing in traditional methods.It is expected to improve the accuracy and efficiency of the interpretation.If large and high-quality datasets exist,data-driven models can reveal relationships of arbitrary complexity.Nevertheless,constructing sufficiently large logging datasets with reliable labels remains challenging,making it difficult to apply data-driven models effectively in logging data interpretation.Furthermore,data-driven models often act as“black boxes”without explaining their predictions or ensuring compliance with primary physical constraints.This paper proposes a machine learning method with strong physical constraints by integrating mechanism and data-driven models.Prior knowledge of logging data interpretation is embedded into machine learning regarding network structure,loss function,and optimization algorithm.We employ the Physically Informed Auto-Encoder(PIAE)to predict porosity and water saturation,which can be trained without labeled reservoir parameters using self-supervised learning techniques.This approach effectively achieves automated interpretation and facilitates generalization across diverse datasets.展开更多
Modeling HIV/AIDS progression is critical for understanding disease dynamics and improving patient care. This study compares the Exponential and Weibull survival models, focusing on their ability to capture state-spec...Modeling HIV/AIDS progression is critical for understanding disease dynamics and improving patient care. This study compares the Exponential and Weibull survival models, focusing on their ability to capture state-specific failure rates in HIV/AIDS progression. While the Exponential model offers simplicity with a constant hazard rate, it often fails to accommodate the complexities of dynamic disease progression. In contrast, the Weibull model provides flexibility by allowing hazard rates to vary over time. Both models are evaluated within the frameworks of the Cox Proportional Hazards (Cox PH) and Accelerated Failure Time (AFT) models, incorporating critical covariates such as age, gender, CD4 count, and ART status. Statistical evaluation metrics, including Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), log-likelihood, and Pseudo-R2, were employed to assess model performance across diverse patient subgroups. Results indicate that the Weibull model consistently outperforms the Exponential model in dynamic scenarios, such as younger patients and those with co-infections, while maintaining robustness in stable contexts. This study highlights the trade-off between flexibility and simplicity in survival modeling, advocating for tailored model selection to balance interpretability and predictive accuracy. These findings provide valuable insights for optimizing HIV/AIDS management strategies and advancing survival analysis methodologies.展开更多
Accurate Global Horizontal Irradiance(GHI)forecasting has become vital for successfully integrating solar energy into the electrical grid because of the expanding demand for green power and the worldwide shift favouri...Accurate Global Horizontal Irradiance(GHI)forecasting has become vital for successfully integrating solar energy into the electrical grid because of the expanding demand for green power and the worldwide shift favouring green energy resources.Particularly considering the implications of the aggressive GHG emission targets,accurate GHI forecasting has become vital for developing,designing,and operational managing solar energy systems.This research presented the core concepts of modelling and performance analysis of the application of various forecasting models such as ARIMA(Autoregressive Integrated Moving Average),Elaman NN(Elman Neural Network),RBFN(Radial Basis Function Neural Network),SVM(Support Vector Machine),LSTM(Long Short-Term Memory),Persistent,BPN(Back Propagation Neural Network),MLP(Multilayer Perceptron Neural Network),RF(Random Forest),and XGBoost(eXtreme Gradient Boosting)for assessing multi-seasonal forecasting of GHI.Used the India region data to evaluate the models’performance and forecasting ability.Research using forecasting models for seasonal Global Horizontal Irradiance(GHI)forecasting in winter,spring,summer,monsoon,and autumn.Substantiated performance effectiveness through evaluation metrics,such as Mean Absolute Error(MAE),Root Mean Squared Error(RMSE),and R-squared(R^(2)),coded using Python programming.The performance experimentation analysis inferred that the most accurate forecasts in all the seasons compared to the other forecasting models the Random Forest and eXtreme Gradient Boosting,are the superior and competing models that yield Winter season-based forecasting XGBoost is the best forecasting model with MAE:1.6325,RMSE:4.8338,and R^(2):0.9998.Spring season-based forecasting XGBoost is the best forecasting model with MAE:2.599599,RMSE:5.58539,and R^(2):0.999784.Summer season-based forecasting RF is the best forecasting model with MAE:1.03843,RMSE:2.116325,and R^(2):0.999967.Monsoon season-based forecasting RF is the best forecasting model with MAE:0.892385,RMSE:2.417587,and R^(2):0.999942.Autumn season-based forecasting RF is the best forecasting model with MAE:0.810462,RMSE:1.928215,and R^(2):0.999958.Based on seasonal variations and computing constraints,the findings enable energy system operators to make helpful recommendations for choosing the most effective forecasting models.展开更多
Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have in...Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific accuracy.To address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model co-evolution.First,we examine methods of integrating external knowledge into LLMs to enhance factual accuracy,reasoning capabilities,and interpretability,including incorporating knowledge into training objectives,instruction tuning,retrieval-augmented inference,and knowledge prompting.Second,we discuss model collaboration strategies that leverage the complementary strengths of LLMs and smaller models to improve efficiency and domain-specific performance through techniques such as model merging,functional model collaboration,and knowledge injection.Third,we delve into model co-evolution,in which multiple models collaboratively evolve by sharing knowledge,parameters,and learning strategies to adapt to dynamic environments and tasks,thereby enhancing their adaptability and continual learning.We illustrate how the integration of these techniques advances AI capabilities in science,engineering,and society—particularly in hypothesis development,problem formulation,problem-solving,and interpretability across various domains.We conclude by outlining future pathways for further advancement and applications.展开更多
Concrete material model plays an important role in numerical predictions of its dynamic responses subjected to projectile impact and charge explosion.Current concrete material models could be distinguished into two ki...Concrete material model plays an important role in numerical predictions of its dynamic responses subjected to projectile impact and charge explosion.Current concrete material models could be distinguished into two kinds,i.e.,the hydro-elastoplastic-damage model with independent equation of state and the cap-elastoplastic-damage model with continuous cap surface.The essential differences between the two kind models are vital for researchers to choose an appropriate kind of concrete material model for their concerned problems,while existing studies have contradictory conclusions.To resolve this issue,the constitutive theories of the two kinds of models are firstly overviewed.Then,the constitutive theories between the two kinds of models are comprehensively compared and the main similarities and differences are clarified,which are demonstrated by single element numerical examples.Finally,numerical predictions for projectile penetration and charge explosion experiments on concrete targets are compared to further demonstrate the conclusion made by constitutive comparison.It is found that both the two kind models could be used to simulate the dynamic responses of concrete under projectile impact and blast loadings,if the parameter needed in material models are well calibrated,although some discrepancies between them may exist.展开更多
The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can si...The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can simultaneously process multi-modality data such as medical images and medical reports.These models can not only recognize images,but also understand the semantic relationship between images and texts,effectively realize the integration of medical information,and provide strong support for clinical decision-making and disease diagnosis.The visual-language large model has good performance for specific medical tasks,and also shows strong potential and high intelligence in the general task models.This paper provides a comprehensive review of the visual-language large model in the field of medical health.Specifically,this paper first introduces the basic theoretical basis and technical principles.Then,this paper introduces the specific application scenarios in the field of medical health,including modality fusion,semi-supervised learning,weakly supervised learning,unsupervised learning,cross-domain model and general models.Finally,the challenges including insufficient data,interpretability,and practical deployment are discussed.According to the existing challenges,four potential future development directions are given.展开更多
Climate model prediction has been improved by enhancing model resolution as well as the implementation of sophisticated physical parameterization and refinement of data assimilation systems[section 6.1 in Wang et al.(...Climate model prediction has been improved by enhancing model resolution as well as the implementation of sophisticated physical parameterization and refinement of data assimilation systems[section 6.1 in Wang et al.(2025)].In relation to seasonal forecasting and climate projection in the East Asian summer monsoon season,proper simulation of the seasonal migration of rain bands by models is a challenging and limiting factor[section 7.1 in Wang et al.(2025)].展开更多
Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate ...Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate the biological characteristics and treatment outcomes seen in human diseases.Currently,various lung cancer models have been established,including chemical induction models,orthotopic transplan-tation models,ectopic transplantation models,metastasis models,and gene editing mouse models.Additionally,lung cancer grafts can be categorized into two types:tissue-based and cell-based grafts.This paper summarizes the phenotypes,advan-tages,and disadvantages of various induction methods based on their modeling tech-niques.The goal is to enhance the simulation of clinical lung cancer characteristics and to establish a solid foundation for future clinical research.展开更多
In this paper,a physical model of RIS of bistatic polarized radar cross section is derived starting from the Stratton-Chu equations under the assumptions of physical optics,PEC,far field and rectangular RIS element.In...In this paper,a physical model of RIS of bistatic polarized radar cross section is derived starting from the Stratton-Chu equations under the assumptions of physical optics,PEC,far field and rectangular RIS element.In the context of important physical characteristics of the backscattering polarization of RIS,the modeling of the RIS wireless channel requires a tradeoff between complexity and accuracy,as well as usability and simplicity.For channel modeling of RIS systems,RIS is modelled as multi-equivalent virtual base stations(BSs)induced by multi polarized electromagnetic waves from different incident directions.The comparison between test and simulation results demonstrates that the proposed algorithm effectively captures the key characteristics of the general RIS element polarization physical model and provides accurate results.展开更多
In this study,we used an extensive sampling network established in central Romania to develop tree height and crown length models.Our analysis included more than 18,000 tree measurements from five different species.In...In this study,we used an extensive sampling network established in central Romania to develop tree height and crown length models.Our analysis included more than 18,000 tree measurements from five different species.Instead of building univariate models for each response variable,we employed a multivariate approach using seemingly unrelated mixed-effects models.These models incorporated variables related to species mixture,tree and stand size,competition,and stand structure.With the inclusion of additional variables in the multivariate seemingly unrelated mixed-effects models,the accuracy of the height prediction models improved by over 10% for all species,whereas the improvement in the crown length models was considerably smaller.Our findings indicate that trees in mixed stands tend to have shorter heights but longer crowns than those in pure stands.We also observed that trees in homogeneous stand structures have shorter crown lengths than those in heterogeneous stands.By employing a multivariate mixed-effects modelling framework,we were able to perform cross-model random-effect predictions,leading to a significant increase in accuracy when both responses were used to calibrate the model.In contrast,the improvement in accuracy was marginal when only height was used for calibration.We demonstrate how multivariate mixed-effects models can be effectively used to develop multi-response allometric models that can be easily calibrated with a limited number of observations while simultaneously achieving better-aligned projections.展开更多
Firstly,based on the data of air quality and the meteorological data in Baoding City from 2017 to 2021,the correlations of meteorological elements and pollutants with O_(3)concentration were explored to determine the ...Firstly,based on the data of air quality and the meteorological data in Baoding City from 2017 to 2021,the correlations of meteorological elements and pollutants with O_(3)concentration were explored to determine the forecast factors of forecast models.Secondly,the O_(3)-8h concentration in Baoding City in 2021 was predicted based on the constructed models of multiple linear regression(MLR),backward propagation neural network(BPNN),and auto regressive integrated moving average(ARIMA),and the predicted values were compared with the observed values to test their prediction effects.The results show that overall,the MLR,BPNN and ARIMA models were able to forecast the changing trend of O_(3)-8h concentration in Baoding in 2021,but the BPNN model gave better forecast results than the ARIMA and MLR models,especially for the prediction of the high values of O_(3)-8h concentration,and the correlation coefficients between the predicted values and the observed values were all higher than 0.9 during June-September.The mean error(ME),mean absolute error(MAE),and root mean square error(RMSE)of the predicted values and the observed values of daily O_(3)-8h concentration based on the BPNN model were 0.45,19.11 and 24.41μg/m 3,respectively,which were significantly better than those of the MLR and ARIMA models.The prediction effects of the MLR,BPNN and ARIMA models were the best at the pollution level,followed by the excellent level,and it was the worst at the good level.In comparison,the prediction effect of BPNN model was better than that of the MLR and ARIMA models as a whole,especially for the pollution and excellent levels.The TS scores of the BPNN model were all above 66%,and the PC values were above 86%.The BPNN model can forecast the changing trend of O_(3)concentration more accurately,and has a good practical application value,but at the same time,the predicted high values of O_(3)concentration should be appropriately increased according to error characteristics of the model.展开更多
As one of the main characteristics of atmospheric pollutants,PM_(2.5) severely affects human health and has received widespread attention in recent years.How to predict the variations of PM_(2.5) concentrations with h...As one of the main characteristics of atmospheric pollutants,PM_(2.5) severely affects human health and has received widespread attention in recent years.How to predict the variations of PM_(2.5) concentrations with high accuracy is an important topic.The PM_(2.5) monitoring stations in Xinjiang Uygur Autonomous Region,China,are unevenly distributed,which makes it challenging to conduct comprehensive analyses and predictions.Therefore,this study primarily addresses the limitations mentioned above and the poor generalization ability of PM_(2.5) concentration prediction models across different monitoring stations.We chose the northern slope of the Tianshan Mountains as the study area and took the January−December in 2019 as the research period.On the basis of data from 21 PM_(2.5) monitoring stations as well as meteorological data(temperature,instantaneous wind speed,and pressure),we developed an improved model,namely GCN−TCN−AR(where GCN is the graph convolution network,TCN is the temporal convolutional network,and AR is the autoregression),for predicting PM_(2.5) concentrations on the northern slope of the Tianshan Mountains.The GCN−TCN−AR model is composed of an improved GCN model,a TCN model,and an AR model.The results revealed that the R2 values predicted by the GCN−TCN−AR model at the four monitoring stations(Urumqi,Wujiaqu,Shihezi,and Changji)were 0.93,0.91,0.93,and 0.92,respectively,and the RMSE(root mean square error)values were 6.85,7.52,7.01,and 7.28μg/m^(3),respectively.The performance of the GCN−TCN−AR model was also compared with the currently neural network models,including the GCN−TCN,GCN,TCN,Support Vector Regression(SVR),and AR.The GCN−TCN−AR outperformed the other current neural network models,with high prediction accuracy and good stability,making it especially suitable for the predictions of PM_(2.5)concentrations.This study revealed the significant spatiotemporal variations of PM_(2.5)concentrations.First,the PM_(2.5) concentrations exhibited clear seasonal fluctuations,with higher levels typically observed in winter and differences presented between months.Second,the spatial distribution analysis revealed that cities such as Urumqi and Wujiaqu have high PM_(2.5) concentrations,with a noticeable geographical clustering of pollutions.Understanding the variations in PM_(2.5) concentrations is highly important for the sustainable development of ecological environment in arid areas.展开更多
Large models,such as large language models(LLMs),vision-language models(VLMs),and multimodal agents,have become key elements in artificial intelli⁃gence(AI)systems.Their rapid development has greatly improved percepti...Large models,such as large language models(LLMs),vision-language models(VLMs),and multimodal agents,have become key elements in artificial intelli⁃gence(AI)systems.Their rapid development has greatly improved perception,generation,and decision-making in various fields.However,their vast scale and complexity bring about new security challenges.Issues such as backdoor vulnerabilities during training,jailbreaking in multimodal rea⁃soning,and data provenance and copyright auditing have made security a critical focus for both academia and industry.展开更多
Conducting predictability studies is essential for tracing the source of forecast errors,which not only leads to the improvement of observation and forecasting systems,but also enhances the understanding of weather an...Conducting predictability studies is essential for tracing the source of forecast errors,which not only leads to the improvement of observation and forecasting systems,but also enhances the understanding of weather and climate phenomena.In the past few decades,dynamical numerical models have been the primary tools for predictability studies,achieving significant progress.Nowadays,with the advances in artificial intelligence(AI)techniques and accumulations of vast meteorological data,modeling weather and climate events using modern data-driven approaches is becoming trendy,where FourCastNet,Pangu-Weather,and GraphCast are successful pioneers.In this perspective article,we suggest AI models should not be limited to forecasting but be expanded to predictability studies,leveraging AI's advantages of high efficiency and self-contained optimization modules.To this end,we first remark that AI models should possess high simulation capability with fine spatiotemporal resolution for two kinds of predictability studies.AI models with high simulation capabilities comparable to numerical models can be considered to provide solutions to partial differential equations in a data-driven way.Then,we highlight several specific predictability issues with well-determined nonlinear optimization formulizations,which can be well-studied using AI models,holding significant scientific value.In addition,we advocate for the incorporation of AI models into the synergistic cycle of the cognition–observation–model paradigm.Comprehensive predictability studies have the potential to transform“big data”to“big and better data”and shift the focus from“AI for forecasts”to“AI for science”,ultimately advancing the development of the atmospheric and oceanic sciences.展开更多
基金support from the Central Research Fund of Kim Chaek University of Technology(MIRAE 2023779).
文摘The phenomenological and physically based models,using the true stress–true strain curve data obtained under various hot working conditions of 850–1200°C and 0.001–10 s−1,were developed and improved for AerMet 100 ultra-high strength steel.The predictability of the developed constitutive models was verified and compared.The determination coefficient and average absolute relative error were 0.9988 and 3.72%for the improved version of the modified Zerilli–Armstrong model,0.9985 and 3.96%for the improved version of the modified Johnson–Cook model,0.9947 and 4.59%for the strain-compensated Arrhenius-type model and 0.9913 and 5.43%for the improved Khan–Huang–Liang model,respectively.The results showed that the improved versions of the modified Zerilli–Armstrong model have the best predictability among the studied constitutive models.Comparing the predictability before and after the improvement,the average absolute relative error was increased by 65.14%for the modified Zerilli–Armstrong model and 58.45%for the modified Johnson–Cook model.This indicates that the phenomenological improvement of physically based constitutive models allows us to develop effectively constitutive equations with high prediction accuracy.
基金co-supported by the National Natural Science Foundation of China(No.61873017)the Academic Excellence Foundation of Beihang University for PhD Students,China.
文摘Swarm Intelligence (SI) is a collective behavior that emerges from interaction between individuals in a group. Typical SI includes fish schooling, ant foraging, bird migration, and so on. A great deal of models have been introduced to characterize the mechanism of SI. This article reviews several typical models and classifies them into four categories: self-driven particle models, with Boids model as the primary example;pheromone communication models, including the ant colony pheromone model which serves as the foundation for ant colony optimization;leadership decision models, utilizing the hierarchical dynamics model of pigeon flock as a prime instance;empirical research models, which employ the topological rule model of starling flock as a classic model. On this basis, each type of model is elaborated upon in terms of its typical model overview, applications, and model evaluation. More specifically, multi-agent swarm control, path optimization and obstacle avoidance, formation and consensus control, trajectory tracking in the dense crowd and social networks analysis are surveyed in the application of each category, respectively. Furthermore, the more precise and effective modeling techniques for leadership decision and empirical research models are described. Limitations and potential directions for further exploration in the study of SI are presented.
基金supported by the National Natural Science Foundation of China(Grant No.52278333)the China Scholarship Council(CSC)and the Science and Technology Department of Liaoning Province(Grant No.2024JH2/102500069).
文摘Peridynamics(PD)is an effective method for simulating the spontaneous initiation and propagation of tensile cracks in materials.However,it faces great challenges in simulating compression-shear cracking of geomaterials due to the lack of efficient contact-friction models.This paper introduces an original contact-friction model that leverages twin mesh and potential function principles within PD to model rock cracking under tensile and compressive stresses.The contact detection algorithm,based on space segmentation axis-aligned bounding box(AABB)tree data structure,is used to address the significant challenge of highly efficient contact detection in compression and shear problems.In this method,the twin mesh and potential function are utilized to quantify contact detection and contact degree,as well as friction behavior.This is in contrast to the distance and circular contact area model,which lacks physical significance in the classical PD method.As demonstrated by the tests on specimens containing cracks,the proposed model can capture 8 types of secondary fractures,reduce the contact detection error by about 29%e56%,and increase the contact retrieval efficiency by over 1600 times compared to the classic PD models.This significantly enhances the capability of PD to simulate the initiation,expansion,and coalescence of intricate compression-shear cracks.
文摘In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes technical exchanges and learning globally.Second,resources required for large model R&D are difficult for a single institution to obtain.The evaluation of general large models also requires the participation of experts from various industries.Third,without open source collaboration,it is difficult to form a unified upper-layer software ecosystem.Therefore,open source has become an important cooperation mechanism to promote the development of AI and large models.There are two cases to illustrate how open source and international standards interact with each other.
基金Supported by the Laoshan Laboratory(No.LSKJ 202202404)the Strategic Priority Research Program of the Chinese Academy of Sciences(No.XDB 42000000)+1 种基金the National Natural Science Foundation of China(NSFC)(No.42030410)the Startup Foundation for Introducing Talent of NUIST,and the Jiangsu Innovation Research Group(No.JSSCTD 202346)。
文摘Numerical models are crucial for quantifying the ocean-atmosphere interactions associated with the El Niño-Southern Oscillation(ENSO)phenomenon in the tropical Pacific.Current coupled models often exhibit significant biases and inter-model differences in simulating ENSO,underscoring the need for alternative modeling approaches.The Regional Ocean Modeling System(ROMS)is a sophisticated ocean model widely used for regional studies and has been coupled with various atmospheric models.However,its application in simulating ENSO processes on a basin scale in the tropical Pacific has not been explored.For the first time,this study presents the development of a basin-scale hybrid coupled model(HCM)for the tropical Pacific,integrating ROMS with a statistical atmospheric model that captures the interannual relationships between sea surface temperature(SST)and wind stress anomalies.The HCM is evaluated for its capability to simulate the annual mean,seasonal,and interannual variations of the oceanic state in the tropical Pacific.Results demonstrate that the model effectively reproduces the ENSO cycle,with a dominant oscillation period of approximately two years.The ROMS-based HCM developed here offers an efficient and robust tool for investigating climate variability in the tropical Pacific.
基金National Natural Science Foundation of China(71690233,71971213,71901214)。
文摘Architecture framework has become an effective method recently to describe the system of systems(SoS)architecture,such as the United States(US)Department of Defense Architecture Framework Version 2.0(DoDAF2.0).As a viewpoint in DoDAF2.0,the operational viewpoint(OV)describes operational activities,nodes,and resource flows.The OV models are important for SoS architecture development.However,as the SoS complexity increases,constructing OV models with traditional methods exposes shortcomings,such as inefficient data collection and low modeling standards.Therefore,we propose an intelligent modeling method for five OV models,including operational resource flow OV-2,organizational relationships OV-4,operational activity hierarchy OV-5a,operational activities model OV-5b,and operational activity sequences OV-6c.The main idea of the method is to extract OV architecture data from text and generate interoperable OV models.First,we construct the OV meta model based on the DoDAF2.0 meta model(DM2).Second,OV architecture named entities is recognized from text based on the bidirectional long short-term memory and conditional random field(BiLSTM-CRF)model.And OV architecture relationships are collected with relationship extraction rules.Finally,we define the generation rules for OV models and develop an OV modeling tool.We use unmanned surface vehicles(USV)swarm target defense SoS architecture as a case to verify the feasibility and effectiveness of the intelligent modeling method.
基金supported by National Key Research and Development Program (2019YFA0708301)National Natural Science Foundation of China (51974337)+2 种基金the Strategic Cooperation Projects of CNPC and CUPB (ZLZX2020-03)Science and Technology Innovation Fund of CNPC (2021DQ02-0403)Open Fund of Petroleum Exploration and Development Research Institute of CNPC (2022-KFKT-09)
文摘We propose an integrated method of data-driven and mechanism models for well logging formation evaluation,explicitly focusing on predicting reservoir parameters,such as porosity and water saturation.Accurately interpreting these parameters is crucial for effectively exploring and developing oil and gas.However,with the increasing complexity of geological conditions in this industry,there is a growing demand for improved accuracy in reservoir parameter prediction,leading to higher costs associated with manual interpretation.The conventional logging interpretation methods rely on empirical relationships between logging data and reservoir parameters,which suffer from low interpretation efficiency,intense subjectivity,and suitability for ideal conditions.The application of artificial intelligence in the interpretation of logging data provides a new solution to the problems existing in traditional methods.It is expected to improve the accuracy and efficiency of the interpretation.If large and high-quality datasets exist,data-driven models can reveal relationships of arbitrary complexity.Nevertheless,constructing sufficiently large logging datasets with reliable labels remains challenging,making it difficult to apply data-driven models effectively in logging data interpretation.Furthermore,data-driven models often act as“black boxes”without explaining their predictions or ensuring compliance with primary physical constraints.This paper proposes a machine learning method with strong physical constraints by integrating mechanism and data-driven models.Prior knowledge of logging data interpretation is embedded into machine learning regarding network structure,loss function,and optimization algorithm.We employ the Physically Informed Auto-Encoder(PIAE)to predict porosity and water saturation,which can be trained without labeled reservoir parameters using self-supervised learning techniques.This approach effectively achieves automated interpretation and facilitates generalization across diverse datasets.
文摘Modeling HIV/AIDS progression is critical for understanding disease dynamics and improving patient care. This study compares the Exponential and Weibull survival models, focusing on their ability to capture state-specific failure rates in HIV/AIDS progression. While the Exponential model offers simplicity with a constant hazard rate, it often fails to accommodate the complexities of dynamic disease progression. In contrast, the Weibull model provides flexibility by allowing hazard rates to vary over time. Both models are evaluated within the frameworks of the Cox Proportional Hazards (Cox PH) and Accelerated Failure Time (AFT) models, incorporating critical covariates such as age, gender, CD4 count, and ART status. Statistical evaluation metrics, including Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), log-likelihood, and Pseudo-R2, were employed to assess model performance across diverse patient subgroups. Results indicate that the Weibull model consistently outperforms the Exponential model in dynamic scenarios, such as younger patients and those with co-infections, while maintaining robustness in stable contexts. This study highlights the trade-off between flexibility and simplicity in survival modeling, advocating for tailored model selection to balance interpretability and predictive accuracy. These findings provide valuable insights for optimizing HIV/AIDS management strategies and advancing survival analysis methodologies.
文摘Accurate Global Horizontal Irradiance(GHI)forecasting has become vital for successfully integrating solar energy into the electrical grid because of the expanding demand for green power and the worldwide shift favouring green energy resources.Particularly considering the implications of the aggressive GHG emission targets,accurate GHI forecasting has become vital for developing,designing,and operational managing solar energy systems.This research presented the core concepts of modelling and performance analysis of the application of various forecasting models such as ARIMA(Autoregressive Integrated Moving Average),Elaman NN(Elman Neural Network),RBFN(Radial Basis Function Neural Network),SVM(Support Vector Machine),LSTM(Long Short-Term Memory),Persistent,BPN(Back Propagation Neural Network),MLP(Multilayer Perceptron Neural Network),RF(Random Forest),and XGBoost(eXtreme Gradient Boosting)for assessing multi-seasonal forecasting of GHI.Used the India region data to evaluate the models’performance and forecasting ability.Research using forecasting models for seasonal Global Horizontal Irradiance(GHI)forecasting in winter,spring,summer,monsoon,and autumn.Substantiated performance effectiveness through evaluation metrics,such as Mean Absolute Error(MAE),Root Mean Squared Error(RMSE),and R-squared(R^(2)),coded using Python programming.The performance experimentation analysis inferred that the most accurate forecasts in all the seasons compared to the other forecasting models the Random Forest and eXtreme Gradient Boosting,are the superior and competing models that yield Winter season-based forecasting XGBoost is the best forecasting model with MAE:1.6325,RMSE:4.8338,and R^(2):0.9998.Spring season-based forecasting XGBoost is the best forecasting model with MAE:2.599599,RMSE:5.58539,and R^(2):0.999784.Summer season-based forecasting RF is the best forecasting model with MAE:1.03843,RMSE:2.116325,and R^(2):0.999967.Monsoon season-based forecasting RF is the best forecasting model with MAE:0.892385,RMSE:2.417587,and R^(2):0.999942.Autumn season-based forecasting RF is the best forecasting model with MAE:0.810462,RMSE:1.928215,and R^(2):0.999958.Based on seasonal variations and computing constraints,the findings enable energy system operators to make helpful recommendations for choosing the most effective forecasting models.
基金supported in part by National Natural Science Foundation of China(62441605)。
文摘Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple modalities.Despite these achievements,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific accuracy.To address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model co-evolution.First,we examine methods of integrating external knowledge into LLMs to enhance factual accuracy,reasoning capabilities,and interpretability,including incorporating knowledge into training objectives,instruction tuning,retrieval-augmented inference,and knowledge prompting.Second,we discuss model collaboration strategies that leverage the complementary strengths of LLMs and smaller models to improve efficiency and domain-specific performance through techniques such as model merging,functional model collaboration,and knowledge injection.Third,we delve into model co-evolution,in which multiple models collaboratively evolve by sharing knowledge,parameters,and learning strategies to adapt to dynamic environments and tasks,thereby enhancing their adaptability and continual learning.We illustrate how the integration of these techniques advances AI capabilities in science,engineering,and society—particularly in hypothesis development,problem formulation,problem-solving,and interpretability across various domains.We conclude by outlining future pathways for further advancement and applications.
基金supported by the National Natural Science Foundations of China (Grant Nos. 52178515, 52078133)
文摘Concrete material model plays an important role in numerical predictions of its dynamic responses subjected to projectile impact and charge explosion.Current concrete material models could be distinguished into two kinds,i.e.,the hydro-elastoplastic-damage model with independent equation of state and the cap-elastoplastic-damage model with continuous cap surface.The essential differences between the two kind models are vital for researchers to choose an appropriate kind of concrete material model for their concerned problems,while existing studies have contradictory conclusions.To resolve this issue,the constitutive theories of the two kinds of models are firstly overviewed.Then,the constitutive theories between the two kinds of models are comprehensively compared and the main similarities and differences are clarified,which are demonstrated by single element numerical examples.Finally,numerical predictions for projectile penetration and charge explosion experiments on concrete targets are compared to further demonstrate the conclusion made by constitutive comparison.It is found that both the two kind models could be used to simulate the dynamic responses of concrete under projectile impact and blast loadings,if the parameter needed in material models are well calibrated,although some discrepancies between them may exist.
基金The Natural Science Foundation of Hebei Province(F2024501044).
文摘The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can simultaneously process multi-modality data such as medical images and medical reports.These models can not only recognize images,but also understand the semantic relationship between images and texts,effectively realize the integration of medical information,and provide strong support for clinical decision-making and disease diagnosis.The visual-language large model has good performance for specific medical tasks,and also shows strong potential and high intelligence in the general task models.This paper provides a comprehensive review of the visual-language large model in the field of medical health.Specifically,this paper first introduces the basic theoretical basis and technical principles.Then,this paper introduces the specific application scenarios in the field of medical health,including modality fusion,semi-supervised learning,weakly supervised learning,unsupervised learning,cross-domain model and general models.Finally,the challenges including insufficient data,interpretability,and practical deployment are discussed.According to the existing challenges,four potential future development directions are given.
文摘Climate model prediction has been improved by enhancing model resolution as well as the implementation of sophisticated physical parameterization and refinement of data assimilation systems[section 6.1 in Wang et al.(2025)].In relation to seasonal forecasting and climate projection in the East Asian summer monsoon season,proper simulation of the seasonal migration of rain bands by models is a challenging and limiting factor[section 7.1 in Wang et al.(2025)].
基金Sichuan Provincial Administration of Traditional Chinese Medicine,Grant/Award Number:2023MS564National Natural Science Foundation of China,Grant/Award Number:82474436。
文摘Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate the biological characteristics and treatment outcomes seen in human diseases.Currently,various lung cancer models have been established,including chemical induction models,orthotopic transplan-tation models,ectopic transplantation models,metastasis models,and gene editing mouse models.Additionally,lung cancer grafts can be categorized into two types:tissue-based and cell-based grafts.This paper summarizes the phenotypes,advan-tages,and disadvantages of various induction methods based on their modeling tech-niques.The goal is to enhance the simulation of clinical lung cancer characteristics and to establish a solid foundation for future clinical research.
基金supported by Ministry of Science and Technology of the People’s Republic of China(2020YFB1808101)the Project“5G evolution wireless air interface intelligent R&D and verification public platform project”supported by Ministry of Industry and Information Technology of the People’s Republic of China(TC220A04M).
文摘In this paper,a physical model of RIS of bistatic polarized radar cross section is derived starting from the Stratton-Chu equations under the assumptions of physical optics,PEC,far field and rectangular RIS element.In the context of important physical characteristics of the backscattering polarization of RIS,the modeling of the RIS wireless channel requires a tradeoff between complexity and accuracy,as well as usability and simplicity.For channel modeling of RIS systems,RIS is modelled as multi-equivalent virtual base stations(BSs)induced by multi polarized electromagnetic waves from different incident directions.The comparison between test and simulation results demonstrates that the proposed algorithm effectively captures the key characteristics of the general RIS element polarization physical model and provides accurate results.
基金supported by the European Union and the Romanian Government through the Competitiveness Operational Programme 2014–2020, under the project“Increasing the economic competitiveness of the forestry sector and the quality of life through knowledge transfer,technology and CDI skills”(CRESFORLIFE),ID P 40 380/105506, subsidiary contract no. 17/2020partially by the FORCLIMSOC Nucleu Programme (Contract 12N/2023)+2 种基金project PN 23090101CresPerfInst project (Contract 34PFE/December 30, 2021)“Increasing the institutional capacity and performance of INCDS ‘Marin Drǎcea’in RDI activities-CresPer”LM was financially supported by the Research Council of Finland's flagship ecosystem for Forest-Human-Machine Interplay–Building Resilience, Redefining Value Networks and Enabling Meaningful Experiences (UNITE)(decision number 357909)
文摘In this study,we used an extensive sampling network established in central Romania to develop tree height and crown length models.Our analysis included more than 18,000 tree measurements from five different species.Instead of building univariate models for each response variable,we employed a multivariate approach using seemingly unrelated mixed-effects models.These models incorporated variables related to species mixture,tree and stand size,competition,and stand structure.With the inclusion of additional variables in the multivariate seemingly unrelated mixed-effects models,the accuracy of the height prediction models improved by over 10% for all species,whereas the improvement in the crown length models was considerably smaller.Our findings indicate that trees in mixed stands tend to have shorter heights but longer crowns than those in pure stands.We also observed that trees in homogeneous stand structures have shorter crown lengths than those in heterogeneous stands.By employing a multivariate mixed-effects modelling framework,we were able to perform cross-model random-effect predictions,leading to a significant increase in accuracy when both responses were used to calibrate the model.In contrast,the improvement in accuracy was marginal when only height was used for calibration.We demonstrate how multivariate mixed-effects models can be effectively used to develop multi-response allometric models that can be easily calibrated with a limited number of observations while simultaneously achieving better-aligned projections.
基金the Project of the Key Open Laboratory of Atmospheric Detection,China Meteorological Administration(2023KLAS02M)the Second Batch of Science and Technology Project of China Meteorological Administration("Jiebangguashuai"):the Research and Development of Short-term and Near-term Warning Products for Severe Convective Weather in Beijing-Tianjin-Hebei Region(CMAJBGS202307).
文摘Firstly,based on the data of air quality and the meteorological data in Baoding City from 2017 to 2021,the correlations of meteorological elements and pollutants with O_(3)concentration were explored to determine the forecast factors of forecast models.Secondly,the O_(3)-8h concentration in Baoding City in 2021 was predicted based on the constructed models of multiple linear regression(MLR),backward propagation neural network(BPNN),and auto regressive integrated moving average(ARIMA),and the predicted values were compared with the observed values to test their prediction effects.The results show that overall,the MLR,BPNN and ARIMA models were able to forecast the changing trend of O_(3)-8h concentration in Baoding in 2021,but the BPNN model gave better forecast results than the ARIMA and MLR models,especially for the prediction of the high values of O_(3)-8h concentration,and the correlation coefficients between the predicted values and the observed values were all higher than 0.9 during June-September.The mean error(ME),mean absolute error(MAE),and root mean square error(RMSE)of the predicted values and the observed values of daily O_(3)-8h concentration based on the BPNN model were 0.45,19.11 and 24.41μg/m 3,respectively,which were significantly better than those of the MLR and ARIMA models.The prediction effects of the MLR,BPNN and ARIMA models were the best at the pollution level,followed by the excellent level,and it was the worst at the good level.In comparison,the prediction effect of BPNN model was better than that of the MLR and ARIMA models as a whole,especially for the pollution and excellent levels.The TS scores of the BPNN model were all above 66%,and the PC values were above 86%.The BPNN model can forecast the changing trend of O_(3)concentration more accurately,and has a good practical application value,but at the same time,the predicted high values of O_(3)concentration should be appropriately increased according to error characteristics of the model.
基金supported by the Program of Support Xinjiang by Technology(2024E02028,B2-2024-0359)Xinjiang Tianchi Talent Program of 2024,the Foundation of Chinese Academy of Sciences(B2-2023-0239)the Youth Foundation of Shandong Natural Science(ZR2023QD070).
文摘As one of the main characteristics of atmospheric pollutants,PM_(2.5) severely affects human health and has received widespread attention in recent years.How to predict the variations of PM_(2.5) concentrations with high accuracy is an important topic.The PM_(2.5) monitoring stations in Xinjiang Uygur Autonomous Region,China,are unevenly distributed,which makes it challenging to conduct comprehensive analyses and predictions.Therefore,this study primarily addresses the limitations mentioned above and the poor generalization ability of PM_(2.5) concentration prediction models across different monitoring stations.We chose the northern slope of the Tianshan Mountains as the study area and took the January−December in 2019 as the research period.On the basis of data from 21 PM_(2.5) monitoring stations as well as meteorological data(temperature,instantaneous wind speed,and pressure),we developed an improved model,namely GCN−TCN−AR(where GCN is the graph convolution network,TCN is the temporal convolutional network,and AR is the autoregression),for predicting PM_(2.5) concentrations on the northern slope of the Tianshan Mountains.The GCN−TCN−AR model is composed of an improved GCN model,a TCN model,and an AR model.The results revealed that the R2 values predicted by the GCN−TCN−AR model at the four monitoring stations(Urumqi,Wujiaqu,Shihezi,and Changji)were 0.93,0.91,0.93,and 0.92,respectively,and the RMSE(root mean square error)values were 6.85,7.52,7.01,and 7.28μg/m^(3),respectively.The performance of the GCN−TCN−AR model was also compared with the currently neural network models,including the GCN−TCN,GCN,TCN,Support Vector Regression(SVR),and AR.The GCN−TCN−AR outperformed the other current neural network models,with high prediction accuracy and good stability,making it especially suitable for the predictions of PM_(2.5)concentrations.This study revealed the significant spatiotemporal variations of PM_(2.5)concentrations.First,the PM_(2.5) concentrations exhibited clear seasonal fluctuations,with higher levels typically observed in winter and differences presented between months.Second,the spatial distribution analysis revealed that cities such as Urumqi and Wujiaqu have high PM_(2.5) concentrations,with a noticeable geographical clustering of pollutions.Understanding the variations in PM_(2.5) concentrations is highly important for the sustainable development of ecological environment in arid areas.
文摘Large models,such as large language models(LLMs),vision-language models(VLMs),and multimodal agents,have become key elements in artificial intelli⁃gence(AI)systems.Their rapid development has greatly improved perception,generation,and decision-making in various fields.However,their vast scale and complexity bring about new security challenges.Issues such as backdoor vulnerabilities during training,jailbreaking in multimodal rea⁃soning,and data provenance and copyright auditing have made security a critical focus for both academia and industry.
基金in part supported by the National Natural Science Foundation of China(Grant Nos.42288101,42405147 and 42475054)in part by the China National Postdoctoral Program for Innovative Talents(Grant No.BX20230071)。
文摘Conducting predictability studies is essential for tracing the source of forecast errors,which not only leads to the improvement of observation and forecasting systems,but also enhances the understanding of weather and climate phenomena.In the past few decades,dynamical numerical models have been the primary tools for predictability studies,achieving significant progress.Nowadays,with the advances in artificial intelligence(AI)techniques and accumulations of vast meteorological data,modeling weather and climate events using modern data-driven approaches is becoming trendy,where FourCastNet,Pangu-Weather,and GraphCast are successful pioneers.In this perspective article,we suggest AI models should not be limited to forecasting but be expanded to predictability studies,leveraging AI's advantages of high efficiency and self-contained optimization modules.To this end,we first remark that AI models should possess high simulation capability with fine spatiotemporal resolution for two kinds of predictability studies.AI models with high simulation capabilities comparable to numerical models can be considered to provide solutions to partial differential equations in a data-driven way.Then,we highlight several specific predictability issues with well-determined nonlinear optimization formulizations,which can be well-studied using AI models,holding significant scientific value.In addition,we advocate for the incorporation of AI models into the synergistic cycle of the cognition–observation–model paradigm.Comprehensive predictability studies have the potential to transform“big data”to“big and better data”and shift the focus from“AI for forecasts”to“AI for science”,ultimately advancing the development of the atmospheric and oceanic sciences.
