Numerical Prediction Research Articles

A machine learning model that outperforms conventional global subseasonal forecast models

Skillful subseasonal forecasts are crucial for various sectors of society but pose a grand scientific challenge. Recently, machine learning-based weather forecasting models outperform the most successful numerical weather predictions generated by the European Centre for Medium-Range Weather Forecasts (ECMWF), but have not yet surpassed conventional models at subseasonal timescales. This paper introduces FuXi Subseasonal-to-Seasonal (FuXi-S2S), a machine learning model that provides global daily mean forecasts up to 42 days, encompassing five upper-air atmospheric variables at 13 pressure levels and 11 surface variables. FuXi-S2S, trained on 72 years of daily statistics from ECMWF ERA5 reanalysis data, outperforms the ECMWF’s state-of-the-art Subseasonal-to-Seasonal model in ensemble mean and ensemble forecasts for total precipitation and outgoing longwave radiation, notably enhancing global precipitation forecast. The improved performance of FuXi-S2S can be primarily attributed to its superior capability to capture forecast uncertainty and accurately predict the Madden-Julian Oscillation (MJO), extending the skillful MJO prediction from 30 days to 36 days. Moreover, FuXi-S2S not only captures realistic teleconnections associated with the MJO but also emerges as a valuable tool for discovering precursor signals, offering researchers insights and potentially establishing a new paradigm in Earth system science research.

Numerical Study on the Effect of Geometrical Changes on the Deformation Behaviour of Recycled Aluminium Alloy AA6061 undergoing High Velocity Impact using Hyperworks-Radioss

Numerical analysis is important to predict material behaviour without require an experimental implementation. This manuscript focuses on a numerical prediction establishment of a direct recycled aluminium alloys AA6061 undergoing Taylor Cylinder Impact test using Johnson-Cook model in HyperWorks Radioss. The numerical setup was first validated against the experimental data at the velocity range of 179 to 212 m/s. Good agreement between simulation and experimental data was obtained within the range that exhibits a mushrooming shape fracture mode. A parametric study was then conducted to study the deformation behaviour of the selected recycled aluminum alloys within the validated range at various geometrical settings. The analysis was made by focusing on the post-impact configuration of the projectile at different impact velocities in terms of residual length, deformed diameter, and the final length-to-diameter ratio. It was found that a broader projectile experienced a less significant reduction in its final length (Lf/Lo goes from 0.87 to 0.9 for projectile diameter 9mm to 34mm) and a smaller increase in the deformed diameter compared to a thinner projectile (Df/Do goes from 1.18 to 1.12 for projectile diameter 9mm to 34mm). It was found that a thinner projectile experienced more diameter expansion than length reduction post impact. In addition, a longer projectile experienced more residual length reduction (Lf/Lo goes from 0.92 to 0.87 for projectile length 2mm to 14mm) and more radial deformation compared to the one with a smaller initial length (Df/Do goes from 1.06 to 1.18 for projectile length 2mm to 14mm). All projectiles showed more significant changes on the deformed diameter compared to the changes in residual length post-impact. The results helped the understanding of a critical aspect of the deformation behaviour of recycled aluminum alloy AA6061 more effectively compared to experimental work implementation.

Development and validation of a patient-level model to predict dementia across a network of observational databases

BackgroundA prediction model can be a useful tool to quantify the risk of a patient developing dementia in the next years and take risk-factor-targeted intervention. Numerous dementia prediction models have been developed, but few have been externally validated, likely limiting their clinical uptake. In our previous work, we had limited success in externally validating some of these existing models due to inadequate reporting. As a result, we are compelled to develop and externally validate novel models to predict dementia in the general population across a network of observational databases. We assess regularization methods to obtain parsimonious models that are of lower complexity and easier to implement.MethodsLogistic regression models were developed across a network of five observational databases with electronic health records (EHRs) and claims data to predict 5-year dementia risk in persons aged 55–84. The regularization methods L1 and Broken Adaptive Ridge (BAR) as well as three candidate predictor sets to optimize prediction performance were assessed. The predictor sets include a baseline set using only age and sex, a full set including all available candidate predictors, and a phenotype set which includes a limited number of clinically relevant predictors.ResultsBAR can be used for variable selection, outperforming L1 when a parsimonious model is desired. Adding candidate predictors for disease diagnosis and drug exposure generally improves the performance of baseline models using only age and sex. While a model trained on German EHR data saw an increase in AUROC from 0.74 to 0.83 with additional predictors, a model trained on US EHR data showed only minimal improvement from 0.79 to 0.81 AUROC. Nevertheless, the latter model developed using BAR regularization on the clinically relevant predictor set was ultimately chosen as best performing model as it demonstrated more consistent external validation performance and improved calibration.ConclusionsWe developed and externally validated patient-level models to predict dementia. Our results show that although dementia prediction is highly driven by demographic age, adding predictors based on condition diagnoses and drug exposures further improves prediction performance. BAR regularization outperforms L1 regularization to yield the most parsimonious yet still well-performing prediction model for dementia.

Evaluation of Hub‐Height Wind Forecasts Over the New York Bight

ABSTRACTAs offshore wind energy development accelerates in the United States, it is important to assess the accuracy of hub‐height wind forecasts from numerical weather prediction models over the ocean. Leveraging approximately 2 years of Doppler lidar observations from buoys in the New York Bight, we evaluate 80‐m wind speed forecasts from two weather models: the High‐Resolution Rapid Refresh (HRRR) atmospheric model and the Global Forecast System (GFS) coupled atmosphere‐ocean model. These models have different horizontal grid spacing, vertical layering, initialization methods, and parameterizations of boundary layer mixing and surface–atmosphere interactions. Despite these differences, the models demonstrate similar and highly skillful short‐term forecasts at three measurement sites. At the Hudson Southwest location that provides a full year of data, their performance is statistically indistinguishable: root mean square error (RMSE) = 2.1 m/s and the Pearson correlation coefficient for 24‐h forecasts of both models, and RMSE = 2.6 m/s and 0.83 for 48‐h forecasts. Twenty‐four‐hour forecasts also exhibit skill in predicting quiescent winds and winds associated with maximum turbine power. By Day 10, GFS forecasts on average have almost no skill. The short‐term forecast skill by the HRRR and GFS does not strongly depend on season or time of day, yet we find some dependence of the models' performance on near‐surface stability. Additionally, 4‐ to 14‐day forecasts by the GFS exhibit lower RMSE during summer relative to other seasons. The high skill of the HRRR and GFS short‐term forecasts establishes confidence in their utility for offshore wind energy maintenance and operation.

Numerical investigation to predict the impact of pre-heated air on the tar formation of cocoa pod husk gasification in a fixed bed downdraft gasifier

Tar production is a critical aspect of gasification thermochemistry, and it hinders the widening of biofuel utilisation in several applications. Both the gasifier configuration and the operating conditions influence the quantity of tar in PG. Since temperature plays a major role, this study examines the performance parameters of a gasifier across a spectrum of input air temperatures, from 27 to 627 °C, utilising computational fluid dynamics with species transport and porous media models. Since the deviation of numerical predictions from experimental results is within 10 %, the model's validity is confirmed. CO and H2 gas production increased from 19.23 % to 27.12 % and 9.23 %–16.2 %, respectively, over the tested air temperature range; however, the variation in methane content was not significant. Furthermore, the higher heating values of PG and cold gas efficiency were found to be 5–6.8 MJ/Nm3 and 65–69 %, respectively. The study also investigates reaction rates for gasification, combustion, and volatiles' breakdown under various operating conditions. Among the tar species studied, benzene is the most concentrated at 1.6 g/Nm3, followed by toluene at 0.9 g/Nm3, with naphthalene and phenol present in comparatively smaller amounts. Furthermore, the research indicates that an appropriate temperature of gasifying medium can control tar in PG.

An accurate and robust method for intensification of wastewater sludge pipe flow

Globally, environmental impacts and population growth are driving the process intensification of wastewater treatment plants (WWTPs) via transition from conventional (2–3 wt% solids) to highly concentrated (4–6 wt% solids) wastewater sludges (HCWS). This presents an industrial challenge as HCWS are complex, non-Newtonian materials whose viscosity increases nonlinearly with solids concentration. This viscosity increase is particularly relevant for sludge pipe flow as it leads to considerable pumping pressure that ultimately limits the feasibility of pipe flow transportation. Hence, process intensification demands accurate prediction of HCWS turbulent pipe flow to design and optimise pumping infrastructure and piping systems. Such prediction requires accurate rheological characterisation of HCWS and numerical prediction of HCWS turbulent pipe flow, neither of which has been achieved to date due to respective limitations associated with benchtop rheometry and numerical turbulence models. We address these challenges by first developing accurate methods for rheological characterisation of HCWS via laminar flow of digested sludge at various solids concentrations (2–5 %) in a fully instrumented pipe loop facility at a large-scale WWTP. These rheological parameters are used in direct numerical simulation (DNS) computations (that avoid turbulence models) of turbulent pipe flow of HCWS. These predictions are then validated against turbulent flow pipe loop data. This method yields accurate (2–15 % error) predictions of HCWS turbulent pipe flow, compared with up to ∼75 % error for conventional pipe flow correlations. This validation highlights the need for accurate rheological characterisation and numerical simulation to predict HCWS pipe flow and provides a sound basis for the intensification and optimisation of WWTP pipeline systems.

A novel residual-pyramid-attention super resolution model for mesoscale meteorological forecasting spatial downscaling

ABSTRACT Developing a spatial downscaling method using machine learning techniques has emerged as a significant endeavor in meteorological forecasting. In this study, we propose a novel Residual-Pyramid-Attention Super Resolution (RPASR) model that effectively addresses the checkerboard artifacts issue prevalent in existing approaches, offering a unique perspective on spatial downscaling. Moreover, we propose a unique scaling technique during the data training to enhance the prediction performance. To validate and apply the proposed method, we utilize both low and high-resolution meteorological variables obtained from a numerical weather prediction model and observations in the Zhangbei region, China. The results demonstrate the exceptional performance of the RPASR model in spatial downscaling, exhibiting superior accuracy in meteorological prediction. Notably, the model outperforms interpolation techniques, significantly reducing deviations by 50% for variables such as 2 m temperature (T2), 2 m humidity (rh2), and 10 m wind speed (wspd10). In practical applications, our proposed model surpasses the DeepSD super-resolution model and low-resolution simulations as a whole, particularly for T2, where it enhances the T2 correlation coefficient by 1.2% compared to DeepSD. Additionally, it reduces the T2 root mean square error by 10.6% and the rh2 root mean square error by 8.6% compared to the low-resolution simulations. Overall, the RPASR model demonstrates robust validation and exhibits promising potential for advancing spatial downscaling methods in meteorological forecasting.

Damage analysis in welded steel pipe elbow under strong cyclic loading

According to cyclic loading modes, highly pressurized tubular elements, such as those in elbows, differ from other straight tubular structures in their modes of stress and damage. This work aims to perform a numerical prediction of a tubular structure. It consists of an X52 steel elbow welded to X65 grade ends by straight tubular sections. This pressurized structure is subjected to strong cyclic loading in bending moment applied until damage occurs. The bending moment is combined with different orientations between the closing or opening moment and the out-of-plane moment, presenting a possible but rarely analyzed situation. This cyclic loading mode is taken as an evaluative parameter in its various amplitudes, orientations, and patterns. The cyclic hardening of the elbow and straight tubular sections, including those of the welded joints, are all formulated by the combined isotropic and kinematic model introduced in the ABAQUS calculation code with parameters calibrated according to the Voce model and experimental results, using plasticity by the Von Mises equivalent stress flow theory. The appropriate XFEM (Extended Finite Element Method) technique is used in conjunction with the hardening model to overcome numerical calculation difficulties and predict damage in traction laws, separation by crack initiation and propagation. The results, in the form of hysteresis curves in cyclic bending moment and angular displacement, have shown a strong dependency that conditions the structure’s response as well as the level of its damage.

Numerical fatigue damage analysis and mathematical modeling of articular cartilage under cyclic load via hyperelasticity theory

The paper presents the modeling of damage to articular cartilage under cyclic daily loads using a curved beam model and hyper-elasticity theory. Mainly, articular cartilage (AC) is a kind of very important biological tissue that has the fundamental role of withstanding applied mechanical loads and providing smooth movement of joints. The existence of mechanical loads, however, has a huge influence on the behavior and the entire healthiness of AC. These loads, over time, can cause injury through fatigue-type damage because of frequent stresses. The basic aim of this study is to mainly offer a detailed mathematical model measuring the damage caused in AC under the action of mechanical forces incorporating different variants like age, body mass index, metabolic activity, functionally graded, porosity and prestresses. The structural energies including potential energy according to the neoHookean model as well as kinetic energy and external work are achieved through the use of strain-displacement and stress-strain relations. Then, the nonlinear governing equations of the articular cartilage are derived using Hamilton's principle. Furthermore, the mathematical model has been implemented numerically through the differential quadrature method (DQM). A qualitative correspondence between the numerical predictions and experimental data has led us to conclude that this model has the potential to serve as a valuable tool for physicians and therapists. The results of this research indicate that among the factors affecting the increase of damage in cartilage, the most important factor is the body mass index, followed by a person's age, hormonal conditions, and cartilage thickness with a negative effect. The probability of damage for an athlete is about 33 percent higher than a normal person, and for a weightlifter (heavy sports) it is about 140 percent higher than a normal person.

The Year of Polar Prediction site Model Intercomparison Project (YOPPsiteMIP) phase 1: project overview and Arctic winter forecast evaluation

Abstract. Although the quality of weather forecasts in the polar regions is improving, forecast skill there still lags behind lower latitudes. So far there have been relatively few efforts to evaluate processes in numerical weather prediction systems using in situ and remote sensing datasets from meteorological observatories in the terrestrial Arctic and Antarctic compared to the mid-latitudes. Progress has been limited both by the heterogeneous nature of observatory and forecast data and by limited availability of the parameters needed to perform process-oriented evaluation in multi-model forecast archives. The Year of Polar Prediction (YOPP) site Model Inter-comparison Project (YOPPsiteMIP) is addressing this gap by producing merged observatory data files (MODFs) and merged model data files (MMDFs), bringing together observations and forecast data at polar meteorological observatories in a format designed to facilitate process-oriented evaluation. An evaluation of forecast performance was performed at seven Arctic sites, focussing on the first YOPP Special Observing Period in the Northern Hemisphere (NH-SOP1) in February and March 2018. It demonstrated that although the characteristics of forecast skill vary between the different sites and systems, an underestimation in boundary layer temperature variability across models, which goes hand in hand with an inability to capture cold extremes, is a common issue at several sites. It is found that many models tend to underestimate the sensitivity of the 2 m air temperature (T2m) and the surface skin temperature to variations in radiative forcing, and the reasons for this are discussed.

Sea surface barometry with an O2 differential absorption radar: retrieval algorithm development and simulation

Sea surface air pressure observations are a significant gap in the current Earth observing systems. This study addresses retrieval algorithm development and the evaluation of the potential impact of instrumental and environmental uncertainties on sea level pressure retrievals for the measurements of O2 differential absorption radar systems operating at three spectrally evenly spaced close-frequency bands (65.5, 67.75, and 70.0 GHz). A simulated northern hemispheric summer case is used to simulate retrieval uncertainties. To avoid high attenuation and a low signal-to-noise ratio, radar measurements from weather conditions with a rain rate ≥1 mm/h are not used in the retrieval. This study finds that a retrieval algorithm combining all three channels, i.e., the 3-channel approach, can effectively mitigate major atmospheric and sea surface influences on sea surface air pressure retrieval. The major uncertainty of sea surface pressure retrieval is due to the standard deviation in radar power returns. Analysis and simulation demonstrate the potential of global sea surface pressure observations with errors of about 1∼2 mb, which is urgently needed for the improvement of numerical weather prediction models. Future work will emphasize instrument development and field experiments. It is anticipated that an O2 differential absorption radar system will be available for meteorological applications in a few years.

Evaluation of the potential food allergy risks of human lactoferrin expressed in Komagataella phaffii.

Prior to the introduction of novel food ingredients into the food supply, safety risk assessments are required, and numerous prediction models have been developed and validated to evaluate safety. The allergenic risk potential of Helaina recombinant human lactoferrin (rhLF, Effera™), produced in Komagataella phaffii (K. phaffii) was assessed by literature search, bioinformatics sequence comparisons to known allergens, glycan allergenicity assessment, and a simulated pepsin digestion model. The literature search identified no allergenic risk for Helaina rhLF, K. phaffii, or its glycans. Bioinformatics search strategies showed no significant risk for cross-reactivity or allergenicity between rhLF or the 36 residual host proteins and known human allergens. Helaina rhLF was also rapidly digested in simulated gastric fluid and its digestibility profile was comparable to human milk lactoferrin (hmLF), further demonstrating a low allergenic risk and similarity to the hmLF protein. Collectively, these results demonstrate a low allergenic risk potential of Helaina rhLF and do not indicate the need for further clinical testing or serum IgE binding to evaluate Helaina rhLF for risk of food allergy prior to introduction into the food supply.

Prediction of internal stress in alumina laminates induced by shrinkage mismatch based on elastic model modification

AbstractCracks induced by internal stresses compromise the structural integrity of ceramic laminates. Such stresses can be classified into thermal stress derived from thermal expansion mismatch and shrinkage stress derived from shrinkage mismatch. Analytical models have been proposed for the former, whereas only numerical predictions have been explored for the latter. In this study, an equation is constructed to predict the shrinkage stress by modifying an elastic solution proposed in previous studies. We fabricated laminates with three‐layer sandwiched structures using only alumina specimens with high and low densities to control the shrinkage and avoid thermal mismatch. Surface cracks were initiated when the shrinkage stress exceeded the flexural strength of the surface layer. Although the original strength of the surface layer was 308.8 MPa, the laminates’ flexural strength was lower than that because of shrinkage constraint. Our results demonstrate that the shrinkage stress was successfully estimated analytically using a few material properties measured in an ordinary environment. This study will ensure the structural integrity of ceramic laminates in laminate designs.

Calibration and validation of a phase-field model of brittle fracture within the damage mechanics challenge

In the context of the Damage Mechanics Challenge, we adopt a phase-field model of brittle fracture to blindly predict the behavior up to failure of a notched three-point-bending specimen loaded under mixed-mode conditions. The beam is additively manufactured using a geo-architected gypsum based on the combination of bassanite and a water-based binder. The calibration of the material parameters involved in the model is based on a set of available independent experimental tests and on a two-stage procedure. In the first stage an estimate of most of the elastic parameters is obtained, whereas the remaining parameters are optimized in the second stage to minimize the discrepancy between the numerical predictions and a set of experimental results on notched three-point-bending beams. The good agreement between numerical predictions and experimental results in terms of load–displacement curves and crack paths demonstrates the predictive ability of the model and the reliability of the calibration procedure.

Uncertainty Analysis and Maneuver Simulation of Standard Ship Model

Maneuver simulation of a standard ship model gives indication of numerical accuracy. In the numerical calculation of ship maneuvering, uncertainty analysis is a necessary step to ensure the accuracy of the calculation. In this study, uncertainty pair analysis is carried out in the simulation of the turning circle motion of the standard ship model ONRT in waves. According to the uncertainty analysis procedure recommended by the International Towing Tank Conference (ITTC), the change of ship resistance caused by the number of grids is studied to determine the influence of grid density on the numerical prediction. The simulation of turning motion in waves is carried out based on the uncertainty analysis. It is found that the minimum number of overset grids for this simulation is 1.4 million. The numerical results are fairly accurate compared to experimental results, and this technique provides a method with low calculated cost for this simulation.

Development and application of data-driven CHF model in system analysis code

In nuclear reactor systems, when the fuel rods reach the critical heat flux (CHF), a sharp increase in fuel temperature occurs due to a drastic reduction in heat transfer capacity, thus posing a considerable risk to the reactor’s safety. Consequently, accurate CHF prediction holds paramount importance in accurately simulating accident scenarios and enhancing the overall safety of reactor systems. To tackle the challenge of limited prediction accuracy in existing CHF models, this study initially established a comprehensive database by utilizing available experimental data and lookup tables. Subsequently, various methodologies, including the Back Propagation Neural Network (BPNN), Random Forest (RF), and Physics-Informed Machine Learning (PIML), were employed to develop multiple CHF prediction models, and their performance was thoroughly evaluated. Furthermore, the optimal CHF model was integrated into the self-developed analysis code ARSAC, which was then validated using the ORNL-THTF experiment. The results indicated that the BPNN-based model not only demonstrated exceptional prediction accuracy but also exhibited rapid calculation speeds. Notably, the average relative error between the experimental data points and the calculation results of the modified code is 3.64%, while for the original code, it is 22.84%. This study effectively leverages the strengths of data-driven approaches, providing a robust technical solution for high-precision, efficient, and adaptive numerical prediction and analysis of reactor accidents.

Contrasting the Effects of X-Band Phased Array Radar and S-Band Doppler Radar Data Assimilation on Rainstorm Forecasting in the Pearl River Delta

Contrasting the X-band phased array radar (XPAR) with the conventional S-Band dual-polarization mechanical scanning radar (SMSR), the XPAR offers superior temporal and spatial resolution, enabling a more refined depiction of the internal dynamics within convective systems. While both SMSR and XPAR data are extensively used in monitoring and alerting for severe convective weather, their comparative application in numerical weather prediction through data assimilation remains a relatively unexplored area. This study harnesses the Weather Research and Forecasting Model (WRF) and its data assimilation system (WRFDA) to integrate radial velocity and reflectivity from the Guangzhou SMSR and nine XPARs across Guangdong Province. Utilizing a three-dimensional variational approach at a 1 km convective-scale grid, the assimilated data are applied to forecast a rainstorm event in the Pearl River Delta (PRD) on 6 June 2022. Through a comparative analysis of the results from assimilating SMSR and XPAR data, it was observed that the assimilation of SMSR data led to more extensive adjustments in the lower- and middle-level wind fields compared to XPAR data assimilation. This resulted in an enlarged convergence area at lower levels, prompting an overdevelopment of convective systems and an excessive concentration of internal hydrometeor particles, which in turn led to spurious precipitation forecasts. However, the sequential assimilation of both SMSR and XPAR data effectively reduced the excessive adjustments in the wind fields that were evident when only SMSR data were used. This approach diminished the generation of false echoes and enhanced the precision of quantitative precipitation forecasts. Additionally, the lower spectral width of XPAR data indicates its superior detection accuracy. Assimilating XPAR data alone yields more reasonable adjustments to the low- to middle-level wind fields, leading to the formation of small-to-medium-scale horizontal convergence lines in the lower levels of the analysis field. This enhancement significantly improves the model’s forecasts of composite reflectivity and radar echoes, aligning them more closely with actual observations. Consequently, the Threat Score (TS) and Equitable Threat Score (ETS) for heavy-rain forecasts (>10 mm/h) over the next 5 h are markedly enhanced. This study underscores the necessity of incorporating XPAR data assimilation in numerical weather prediction practices and lays the groundwork for the future joint assimilation of SMSR and XPAR data.

Experimental and kinetic study on the solar-driven iron-based chemical looping dry reforming of methane

The chemical looping dry reforming of methane (CL-DRM) represents a promising pathway for large-scale engineering applications of thermochemical decarbonization techniques. To elucidate the intricate reaction mechanism of CL-DRM, a comprehensive kinetic model encompassing both gas-phase reactions and ferrite surface reactions is developed in this study. This model can serve as a numerical foundation for tracing the reaction progression of comparable iron-based oxygen carriers, allowing visualization of all reaction pathways and further determination of optimal reaction conditions. Besides, foam-structured materials employing NiO-doped CoFe2O4 as oxygen carriers were prepared to conduct application experiments for the CL-DRM on a self-designed solar thermochemical platform. The experimental outcomes, obtained under temperature conditions guided by numerical predictions, validate the promising feasibility of employing Ni-doped ferrite in the CL-DRM reaction system. At a reaction temperature of only 1150 K, remarkable peaks in H2 yield of 23.5 mLmin−1g−1, CO yield of 11.6 mLmin−1g−1, and CH4 conversion around 90% are attained, with a net solar-to-fuel conversion efficiency of 1.93%. A comparative analysis conducted with the conventional two-step thermochemical cycle under the same experimental setup underscores the notable advantages of CL-DRM in terms of syngas production capacity and conversion efficiency. Consequently, CL-DRM can be regarded as a pivotal transition towards the realization of a fully carbon-free thermochemical technology.

A novel approach for the optimum cross-section design of thin-walled box section beams subject to oblique bending based on both adequate-strength and local stability conditions

This study presents a new analytical procedure developed for the optimum cross-section design of the thin-walled box sections subject to oblique bending. The box sections designed in this study have been assumed to be achieved by hollowing out the rectangular solid sections at different subtraction ratios including arbitrary, optimum, and maximum. The optimum subtraction ratio has been determined based on the adequate-strength condition while the maximum subtraction ratio has been determined taking into account both adequate-strength and local buckling conditions. The optimal design of the box sections for the arbitrary subtraction ratio has been attained by minimizing the maximum normal stress, taking into account the adequate-strength condition. Conversely, the cross-sections of the box profiles for the optimum and maximum subtraction ratios have been optimized by minimizing their cross-section area, accounting for the defined combined strength condition involving adequate- strength and local stability requirements. An equivalent box section design model that allows checking the local stability of the box sections possessing dissimilar wall segment thicknesses has been developed to perform the local buckling control on the designed optimum box sections with unequal wall thicknesses. The new local stability conditions have been defined relying on this developed equivalent box section design model. The analytical results, validated through numerical predictions obtained from linear elastic local buckling and post-buckling analyses performed using Abaqus software, have shown that the optimally designed box section achieves substantial material savings per unit length compared to its corresponding equivalent box section.

A Hybrid Machine Learning/Physics‐Based Modeling Framework for 2‐Week Extended Prediction of Tropical Cyclones

AbstractPrediction of tropical cyclones (TCs) beyond a week is challenging but of great importance for disaster prevention and mitigation. We propose a hybrid machine learning (ML)/physics‐based modeling framework to extend TC forecasts to 2 weeks. This framework integrates a recently launched ML‐based global weather prediction model (Pangu) and the high‐resolution physics‐based regional weather research and forecasting (WRF) model. The Pangu model shows promise in enhancing the accuracy of predictions for large‐scale circulation and TC tracks, while the high‐resolution WRF model is capable of capturing the core processes underlying TC evolution. To capitalize on the complementary strengths of both the Pangu and WRF models in predicting TCs, the framework comprises three key components: downscaling the Pangu model using the WRF model, adjusting large‐scale circulation through spectral nudging driven by the Pangu model forecasts, and updating sea surface temperature using an ocean mixed‐layer model. These components also ensure the framework's feasibility for real‐time TC forecasting. The prediction skill of the framework has been demonstrated for five long‐lived TCs across various basins from 2018 to 2023. Results indicate that the hybrid ML/physics‐based modeling framework decreased the 2‐week mean TC track and intensity errors by 59% and 32% compared to the global numerical weather prediction models, by 2% and 59% compared to the ERA5‐driven Pangu model, and by 32% and 23% compared to the ERA5‐driven WRF model, respectively. This implies that the framework has great potential to be used for 2‐week extended prediction of TCs.