Nonlinear Relationship Research Articles

Comparison of different intensive care scoring systems and Glasgow Aneurysm score for aortic aneurysm in predicting 28-day mortality: a retrospective cohort study from MIMIC-IV database

ObjectiveThis study aims to assess the performance of various scoring systems in predicting the 28-day mortality of patients with aortic aneurysms (AA) admitted to the intensive care unit (ICU).MethodsWe utilized data from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) to perform a comparative analysis of various predictive systems, including the Glasgow Aneurysm Score (GAS), Simplified Acute Physiology Score (SAPS) III, SAPS II, Logical Organ Dysfunction System (LODS), Sequential Organ Failure Assessment (SOFA), Systemic Inflammatory Response Syndrome (SIRS), and The Oxford Acute Illness Severity Score (OASIS). The discrimination abilities of these systems were compared using the area under the receiver operating characteristic curve (AUROC). Additionally, a 4-knotted restricted cubic spline regression was employed to evaluate the association between the different scoring systems and the risk of 28-day mortality. Finally, we conducted a subgroup analysis focusing on patients with abdominal aortic aneurysms (AAA).ResultsThis study enrolled 586 patients with AA (68.39% male). Among them, 26 patients (4.4%) died within 28 days. Comparative analysis revealed higher SAPS II, SAPS III, SOFA, LODS, OASIS, and SIRS scores in the deceased group, while no statistically significant difference was observed in GAS scores between the survivor and deceased groups (P = 0.148). The SAPS III system exhibited superior predictive value for the 28-day mortality rate (AUROC 0.805) compared to the LODS system (AUROC 0.771), SOFA (AUROC 0.757), SAPS II (AUROC 0.759), OASIS (AUROC 0.742), SIRS (AUROC 0.638), and GAS (AUROC 0.586) systems. The results of the univariate and multivariate logistic analyses showed that SAPS III was statistically significant for both 28-day and 1-year mortality. Subgroup analyses yielded results consistent with the overall findings. No nonlinear relationship was identified between these scoring systems and 28-day all-cause mortality (P for nonlinear > 0.05).ConclusionThe SAPS III system demonstrated superior discriminatory ability for both 28-day and 1-year mortality compared to the GAS, SAPS II SIRS, SOFA, and OASIS systems among patients with AA.

Plasma atherogenicity index is a powerful indicator for identifying metabolic syndrome in adults with type 2 diabetes mellitus: A cross-sectional study.

Metabolic syndrome (MetS) is an important risk factor for atherosclerotic cardiovascular disease (ASCVD). Elevated triglyceride (TG) levels and decreased high-density lipoprotein levels (HDL-C) are predisposing factors for the development of ASCVD. Evidence on the association between atherosclerotic index of plasma [AIP = log (TG/HDL-C)] and MetS is limited. Our study aimed to investigate the association between AIP and MetS. This is a cross-sectional study that determines the presence of MetS by assessing anthropometric and biochemical parameters. Multivariate log-binomial regression models were used to analyze the relationship between AIP and MetS risk. To further test the stability of the results, we performed sensitivity analyses in young, non-obese, and normal lipid population. Smoothing plots explored the potential nonlinear relationship between the AIP index for MetS and the estimated potential risk threshold. Predictive power of AIP for MetS using respondent operating characteristic (ROC) curves. The prevalence of MetS was 67.35%. Multivariate logistic regression analysis showed an independent and positive association between AIP and MetS (Per 1 SD increase, PR = 1.31, 95% CI: 1.15-1.47). Sensitivity analysis demonstrated the stability of the results. Smoothing plot showed a nonlinear relationship between AIP and MetS, with an inflection point of 0.66. ROC curve analysis, AIP was an accurate indicator for assessing MetS in type 2 diabetics (AUC = 0.840, 95% CI: 0.819-0.862). AIP is a stable and independently powerful predictor of MetS in T2DM patients. AIP can be used as a simple assessment tool for the early detection of MetS and disease management for the prevention of cardiovascular disease.

Tunable low-frequency broadband band gap in nonlinear locally resonant metamaterial inspired by sarcomere structure

Since traditional locally resonant metamaterials (LRMs) cannot achieve a broadband band gap in the low-frequency band, the suppression and isolation of low-frequency broadband vibration has always been a difficult problem in the field of vibration control. Therefore, in this paper, inspired by the structure of the sarcomere, a novel nonlinear LRM is proposed, aiming to open a wider band gap in the low-frequency band. Firstly, a statics model of the nonlinear local resonator (LR) is established to explore the mechanical nonlinear characteristics of the structure with different geometric parameters. And the inflection point fitting method is used to approximate the nonlinear statics model of nonlinear LR. Then a dynamics model was established, the nonlinear dispersion relationship of the infinite periodic system was obtained by Bloch's theorem and l-P perturbation method. And an analytical investigation of the theoretically calculated band gap is conducted through the band gap ratio and relative band gap width. In addition, according to the influence of different structural parameters on the real and imaginary parts of the wave vector, the regulation rules and mechanisms for nonlinear LRM to open low-frequency broadband band gaps are revealed. Finally, the Runge-Kutta method was used to numerically calculate the vibration transmissibility curve of the nonlinear LRM composed of a limited number of nonlinear LRs. And the correctness of the theoretically calculated band gap was verified. The research results provide theoretical support for the application of nonlinear LRMs in the suppression and isolation of low-frequency broadband vibrations.

Study on the correlation between the number of mushroom poisoning cases and meteorological factors based on the generalized additive model in Guizhou Province, 2023

BackgroundMushroom poisoning is a significant food safety issue in Guizhou Province, China. Studies on this topic are essential for its prevention and control. We aimed to analyze the epidemiological characteristics of mushroom poisoning and study the correlation between its cases and meteorological factors in Guizhou Province.MethodsWe collected data on mushroom poisoning and meteorological factors in Guizhou Province in 2023. A descriptive analysis was conducted on the epidemiological features of mushroom poisoning and meteorological factors. We used Spearman correlation analysis and the generalized additive model to examine the relationship between meteorological factors and the number of mushroom poisoning cases.ResultsIn 2023, mushroom poisoning cases in Guizhou Province were concentrated among individuals aged 20–59. Clinical symptoms were primarily gastrointestinal symptoms and occurrences peaked from June to October, mainly in the northeastern region of the province. Most incidents occurred in households. In 72 mushroom poisoning incidents where species were identified, 33 poisonous mushrooms were found. The number of mushroom poisoning cases in Guizhou Province was positively correlated with each meteorological factor(P < 0.05). The generalized additive model showed a significant nonlinear relationship between DGT, PRE, RHU, SSD, and the number of mushroom poisoning cases (P < 0.05). Interaction analysis showed that the risk of mushroom poisoning in Guizhou Province increased with the rising values of any two of these four meteorological factors.ConclusionMushroom poisoning incidents in Guizhou Province are characterized by high-risk groups, seasonality, and specific high-incidence regions and places. Public awareness for high-risk groups and early warnings for high-incidence regions and places should be strengthened every summer and fall. There is a correlation between meteorological factors and the number of mushroom poisoning cases, suggesting that these factors could serve as early warning indicators for the prevention and control of mushroom poisoning.

Geopolitical risks and energy transition: the impact of environmental regulation and green innovation

This study examines the impact of geopolitical risk on energy transition, focusing on the moderating roles of environmental regulations and green innovation within OECD countries. By employing a multivariate linear and nonlinear regression model, we identify a substantial positive effect of geopolitical risk on energy transition. Our analysis indicates that stronger environmental regulations and advancements in green innovation significantly amplify this effect. Through threshold effect bootstrap sampling tests, we detect a nonlinear relationship between geopolitical risk and energy transition at varying levels of environmental regulation and green innovation. We also explore lag effects, revealing that the influence of geopolitical risk on energy transition grows stronger over time. The inclusion of interaction terms in our analysis further clarifies the moderating influences of environmental regulation and green innovation. Utilizing a range of geopolitical risk indicators and regression methods, our findings are robust, consistently highlighting the proactive role of geopolitical risk in fostering energy transition. These insights highlight the importance of integrated strategies that harness environmental regulations and technological innovation to facilitate a resilient and efficient energy transition in the face of challenges posed by geopolitical uncertainties.

Natural resource extraction - Sustainable development relationship and energy productivity moderation in resource-rich countries: A panel Bayesian regression analysis

Social development is essential for improving quality of life, fostering stability, driving economic growth, achieving sustainable goals, enhancing human capital, encouraging civic engagement, building resilience, and fostering innovation, thereby ensuring inclusive, equitable, and sustainable progress. The Social Progress Index (SPI), also known as social development, contains all these features. This study aims to examine the impact of natural resource extraction on social development quadratically using data from 2011 to 2022. The focus of the study is on resource-rich countries. In this study, productive capacity in the energy sector is used as a social development determinant and a natural resource consumption moderator, ensuring sustainability in social development. Analyzing the quadratic relationship allows us to model and understand nonlinear relationships between the predictor and the outcome variable, capturing the effects that change direction or strength at different levels of the predictor. In this study, Bayesian regression is incorporated as an econometric approach. It provides more flexible modelling, incorporates prior knowledge, and offers a full distribution of parameter estimates, leading to better uncertainty quantification and more robust predictions. These estimates confirms a U-shaped natural resource extraction and social development relationship. Energy productive capacities in resource-rich countries not only enhance social development but also ensure the sustainable extraction of natural resources leading to sustainable social development. Population density and biodiversity are incorporated as control variables of the model. The impact of both is positive on social development.

Association of systemic immune-inflammation index with body mass index, waist circumference and prevalence of obesity in US adults

This study aims to investigate the potential relationships between the systemic immune-inflammation index (SII) and body mass index (BMI), waist circumference, and the prevalence of obesity. A cross-sectional analysis was conducted on 7,645 individuals aged 20 and above from the NHANES 2017–2020. Multivariate linear regression analyses were conducted to evaluate the association of the logarithmically transformed SII (lgSII) with BMI and waist circumference. Additionally, multivariable logistic regression was utilized to explore the relationship between lgSII and the prevalence of obesity. Fitted smoothing curves and threshold-effect analysis were applied to elucidate nonlinear relationships. In the fully adjusted model, a positive relationship was observed between lgSII and BMI, waist circumference, and obesity prevalence (β = 3.13, 95% CI 2.10–4.16; β = 7.81, 95% CI 5.50–10.13; OR = 1.44, 95% CI 1.12–1.86). The variables of gender, age, race, education, marital status, poverty income ratio (PIR), energy intake, sleep disorder, smoking status, and alcohol use did not significantly modify the positive association between lgSII and obesity. However, physical activity appeared to influence the positive correlation between lgSII and obesity. Using a two-segment linear regression model, an inverted U-shaped relationship was observed between lgSII and both BMI and waist circumference. Furthermore, lgSII demonstrated a linear positive correlation with obesity prevalence. When stratified by physical activity, lgSII showed a non-significant negative correlation with obesity in the physically active group. Our findings underscore a robust association between the logarithmically transformed SII and BMI, waist circumference, and the prevalence of obesity.

Urinary metals are associated with obesity in U.S. children and adolescents: A cross-sectional study

Heavy metals are pervasive in the environment, and exposure to these metals may contribute to obesity in children and adolescents. We hypothesized that metal exposures are associated with obesity in children and adolescents. Data were drawn from children and adolescents aged 6 to 19 years from the 2007 to 2018 National Health and Nutrition Examination Survey. We employed weighted multivariate logistic regression and restricted cubic spline to explore the effects of individual metal exposures on obesity, and weighted quantile sum regression, quantile g-computed regression, and Bayesian kernel machine regression to explore the effects of mixed metal exposures on obesity. Subgroup analyses by gender were also performed. All models were adjusted for age, gender, race, poverty to income ratio, and serum cotinine. Among the 3,650 children and adolescents studied, 21.04% had obesity. Logistic regression revealed positive associations between barium (OR = 1.23, 95% CI: 1.07-1.40) and thallium (OR = 1.55, 95% CI: 1.23-2.15) with obesity, while cadmium (OR = 0.74, 95% CI: 0.61-0.89), cobalt (OR = 0.51, 95% CI: 0.41-0.62), and lead (OR = 0.70, 95% CI: 0.57-0.86) were negatively associated with obesity. Restricted cubic spline indicated a nonlinear relationship between lead and thallium and obesity. Quantile g-computed regression demonstrated that mixed metal exposure was negatively associated with obesity (OR = 0.50, 95% CI: 0.42-0.59). Subgroup analyses revealed a gender-specific effect for mercury (P for interaction = 0.03), which was negatively associated with obesity in females (OR = 0.83, 95% CI: 0.69-0.99). In conclusion, metal exposures are associated with obesity in children and adolescents, with gender differences.

Prediction of Dielectric Constant in Series of Polymers by Quantitative Structure-Property Relationship (QSPR)

This work is devoted to the investigation of dielectric permittivity which is influenced by electronic, ionic, and dipolar polarization mechanisms, contributing to the material’s capacity to store electrical energy. In this study, an extended dataset of 86 polymers was analyzed, and two quantitative structure–property relationship (QSPR) models were developed to predict dielectric permittivity. From an initial set of 1273 descriptors, the most relevant ones were selected using a genetic algorithm, and machine learning models were built using the Gradient Boosting Regressor (GBR). In contrast to Multiple Linear Regression (MLR)- and Partial Least Squares (PLS)-based models, the gradient boosting models excel in handling nonlinear relationships and multicollinearity, iteratively optimizing decision trees to improve accuracy without overfitting. The developed GBR models showed high R2 coefficients of 0.938 and 0.822, for the training and test sets, respectively. An Accumulated Local Effect (ALE) technique was applied to assess the relationship between the selected descriptors—eight for the GB_A model and six for the GB_B model, and their impact on target property. ALE analysis revealed that descriptors such as TDB09m had a strong positive effect on permittivity, while MLOGP2 showed a negative effect. These results highlight the effectiveness of the GBR approach in predicting the dielectric properties of polymers, offering improved accuracy and interpretability.

Remaining useful life prognostics of bearings based on convolution attention networks and enhanced transformer

Rolling bearings are critical components of industrial equipment, and predicting their remaining useful life (RUL) is a challenging task. This article proposes a new end-to-end model (MDSCT) to achieve efficient and accurate prediction of bearing RUL. MDSCT uses the raw vibration signals collected by sensors for prediction, without relying on a large amount of prior knowledge. The feature extraction backbone of this model is composed of an MDSC attention module that integrates multiple parallel depth-wise separable convolutions and an efficient attention mechanism, combined with a tansformer encoder (PPSformer) optimized using patch embedding and probesparse self attention techniques. They are respectively used to capture local subtle features and global dependent features in degraded signals. This article also proposes an improved adaptive activation function AdaptH_Swish to enhance the model's ability to model nonlinear relationships. To comprehensively verify the comprehensive performance of the model, this paper conducted detailed ablation and comparative experiments using two standard datasets, PHM2012 and XJTU-SY. The experimental results not only confirm the rationality and efficiency of the model structure design, but also demonstrate significant advantages in three evaluation indicators compared to various existing methods, fully demonstrating the high generalization and strong robustness of the MDSCT model in bearing RUL prediction tasks.

Simulating multi-scale optimization and variable selection in species distribution modeling

Species distribution modeling (SDM) is a fundamental tool in theoretical and applied ecology. However, relatively little is known about the performance of different approaches for scale optimization, model selection, and algorithmic prediction in the context of nonlinear, multiscale and interactive relationships between environmental variables and species occurrence. Modelers often struggle to optimize a tradeoff between ecological relevance, model robustness, complexity, and overfitting. In this paper, we investigated several methods designed to optimize spatial scale and variable selection in SDMs, in each case evaluating model fitness, parsimony and predictive performance. We used a simulation approach to produce a large pool of alternative underlying habitat relationships that reflect a broad range of realistic habitat associations. We also compared several different modeling algorithms, including logistic regression with a generalized linear model (GLM), Lasso and Elastic-Net Regularized GLMs (GLMNet), and random forest (RF), as well as alternative variable and scale selection methods. We found that GLM methods employing all-subsets dredge routines for variable selection were consistently the best predictors based on all criteria of our model performance assessment and across all attributes of the simulated underlying relationship, including nonlinearity and interaction. We had expected machine learning approaches, such as random forest, to perform better in these more complex forms of species-environment relationships. GLM using dredge variable selection was also the method that included the fewest spurious covariates and included the most correct predictors as a proportion of all predictors. We found that univariate scaling was the most robust method of variable and scale selection, along with Minimal Redundancy Maximal Relevancy (MRMR) which performed equivalently. The simulation experiment presented here provides a robust assessment of simulated multi-species distribution model performance, complexity and fidelity. By simulating a large range of potential habitat relationships with varying spatial scale, effect sizes, linearity, and interactions, we comprehensively evaluated model performance across gradients of complexity of the underlying relationships and violations of classical statistical assumptions. This study provides a valuable assessment and a broader example of the power and utility of controlled simulation experiments in habitat relationships and other ecological spatial predictive modeling.

The onset of parametric decay of lower hybrid waves during lower hybrid current drive experiments on Experimental Advanced Superconducting Tokamak

Abstract An ad hoc calculation is presented to interpret the growth of sideband waves during lower hybrid current drive (LHCD) experiments on Experimental Advanced Superconducting Tokamak (EAST). We adopt the solution of the non-linear dispersion relation for parametric instabilities to describe the growth rate of ion cyclotron quasi-modes (ICQMs) in the peripheral plasma, and use the WKB solution of the propagation of electrostatic waves in two dimensions to model the variation of the electric field in the scrape-off-layer (SOL). We then calculate the growth rates and amplification factors of the sideband waves associated with ICQM in the parallel and perpendicular coupling limits. The onset and substantial growth of the first down-shifted sideband of lower hybrid waves (LHWs) are associated with elevation of the plasma density in the outer SOL where the growth rate of ICQM in the parallel coupling limit increases sharply after the density is elevated. The peak frequencies associated with the first harmonic of ICQM obtained from calculating the amplification factors in the parallel coupling limit ( A para ) agree very well with the observed frequencies of the first down-shifted sideband of 2.45 GHz LHWs in EAST pulse 43 772 and 86 521. The consistency between the calculation and the observation indicates that the sideband waves are associated with parametric decay instability of ICQM excited in the outer SOL through parallel coupling. The amplitude of A para obtained from a single pass of LHWs through the SOL of EAST is, however, one order of magnitude smaller than unity, meaning that it remains elusive whether the substantial growth of sideband waves mainly occurs in front of the antenna or requires multiple passes of LHWs across the SOL region.

A semi-analytical methodology for predicting the vibroacoustic response of functionally graded nanoplates under thermal loads

In this investigation, the analysis of the nonlinear vibroacoustic and sound transmission loss behaviors of nanoplates made of functionally graded materials considering the nonlocal strain gradient theory is presented. It is presumed that the properties of the functionally graded nanoplate are in the form of the simple power law scheme and continuous along the thickness, under nonlinear temperature rise and incident oblique acoustic wave as well as the first-order shear deformation theory. For this purpose, applying Hamilton’s principle, the nonlinear partial differential equations of motion are derived by the displacement field function approach and by considering the von Kármán nonlinear strain-displacement relations. To solve the equations, using the Galerkin method, the nonlinear partial differential equations of motion lead to the Duffing equation. Then, applying the homotopy analysis method, the equation of the transverse movement of the nanoplate is solved semi-analytically to obtain the nonlinear frequencies. Lastly, the nonlinear vibration and acoustic response of functionally graded nanoplate are investigated by considering the variation of the significant factors such as material gradient indices, nonlocal parameters, length scale parameters, aspect ratio, dimensionless amplitude, nonlinear temperature changes, and sound characteristics of the functionally graded nanoplate.

Economic growth, income inequality and lethal violence in developed countries

PurposeThe paper aims to investigate the effect of GDP growth on crime and to test the hypothesis of nonlinearity. Additionally, we estimate the interaction between GDP and income inequality and examine its impact on the relationship between GDP and homicide rates.Design/methodology/approachThe study utilizes panel data from the Organization for Economic Cooperation and Development (OECD), spanning the period from 2000 to 2018 and estimates dynamic panel GMM models.FindingsWe found a nonlinear relationship between GDP and homicide rates, indicating a dual effect of GDP on the occurrence of lethal crimes. Moreover, income inequality conditions the effect of GDP on homicide rates, exerting a significant influence. We conclude that in contexts characterized by high levels of income inequality, GDP growth is more effective in reducing crime, as there is greater potential for improvement.Originality/valueThis paper contributes to the existing literature by providing insights into the complex nonlinearity between economic conditions, income inequality and homicide rates.

The genetic architecture of protein stability

There are more ways to synthesize a 100-amino acid (aa) protein (20100) than there are atoms in the universe. Only a very small fraction of such a vast sequence space can ever be experimentally or computationally surveyed. Deep neural networks are increasingly being used to navigate high-dimensional sequence spaces1. However, these models are extremely complicated. Here, by experimentally sampling from sequence spaces larger than 1010, we show that the genetic architecture of at least some proteins is remarkably simple, allowing accurate genetic prediction in high-dimensional sequence spaces with fully interpretable energy models. These models capture the nonlinear relationships between free energies and phenotypes but otherwise consist of additive free energy changes with a small contribution from pairwise energetic couplings. These energetic couplings are sparse and associated with structural contacts and backbone proximity. Our results indicate that protein genetics is actually both rather simple and intelligible.

Prediction of the ROP based on GA-LightGBM and drilling data

ABSTRACT Constructing oil and gas wells is one of the most expensive activities in the petroleum sector, and increasing drilling speed is crucial for reducing costs, minimizing non-productive time (NPT), and improving efficiency. To address this challenge, a data-driven predictive model for mechanical drilling speed was developed using an artificial intelligence approach. Several machine learning algorithms – including decision trees, support vector regression, k-nearest neighbors, neural networks, XGBoost, and LightGBM – were selected to thoroughly investigate the nonlinear relationship between drilling data and the rate of penetration (ROP). These models were trained, evaluated, and compared, with results indicating that the LightGBM algorithm provided the most accurate ROP predictions. Building on this, the LightGBM model was further refined using cross-validation techniques and a genetic algorithm (GA) to optimize its hyperparameters, resulting in even greater accuracy. The research findings demonstrate that the GA-LightGBM algorithm achieves high precision in ROP prediction on the test set (MAE = 2.39, MSE = 10.53, R2 = 0.93). This model effectively predicts mechanical drilling speed in various well sections, thereby aiding in optimizing the drilling process, improving production efficiency, and reducing costs.

Retrieval of boundary layer precipitable water from GOES ABI using machine learning techniques

Abstract Low-level moisture is an important ingredient for forecasting severe storms, especially over the Great Plains where severe storms often develop along the dryline. Although ground-based observation systems such as lidar or radiosondes on weather balloons provide accurate information on low-level moisture, data are provided at limited locations, and the low temporal resolution of radiosondes makes it difficult to track a rapidly developing dryline. Geostationary satellites provide high spatial and temporal observation, but the channels of current geostationary satellites are mostly sensitive to water vapor at mid to upper levels. However, the split window difference (SWD) between the “clean” window channel (10.3 μm) and the “dirty” window channel (12.3 μm) is commonly used for estimating low-level water vapor. However, this estimation is complicated by surface temperature contributions, dependence on the lapse rate, and nonlinear relationships between SWD and moisture. This study applies machine learning techniques to infer boundary layer precipitable water (BLPW) from Geostationary Operational Environmental Satellite (GOES) Advanced Baseline Imager (ABI) data. Since there are few observations that cover wide regions for training convolutional neural networks, especially for the atmosphere above the surface, High-Resolution Rapid Refresh (HRRR) model outputs are used as the truth for training. Statistical results show that using ABI channel 13, 14, and 15 brightness temperatures, skin temperature, solar zenith angle, and GOES total precipitable water product as additional inputs show the lowest MSE. Case study results show that the model developed in this study is good at capturing strong moisture gradients and depicting a dryline.

The Cost-Optimal Control of Building Air Conditioner Loads Based on Machine Learning: A Case Study of an Office Building in Nanjing

Building envelopes and indoor environments exhibit thermal inertia, forming a virtual energy storage system in conjunction with the building air conditioner (AC) system. This system represents a current demand response resource for building electricity use. Thus, this study centers on the CatBoost algorithm within machine learning (ML) technology, utilizing the LASSO regression model for feature selection and applying the Optuna framework for hyperparameter optimization (HPO) to develop a cost-optimal control method for minimizing building AC loads. This method addresses the challenges associated with traditional load forecasting and control methods, which are often impacted by environmental temperature, building parameters, and user behavior uncertainties. These methods struggle to accurately capture the complex dynamics and nonlinear relationships of AC operations, making it difficult to devise AC operation and virtual energy storage scheduling strategies effectively. The proposed method was applied and validated using a case study of an office building in Nanjing, China. The prediction results showed coefficient of variation in root mean square error (CV-RMSE) values of 6.4% and 2.2%. Compared with the original operating conditions, the indoor temperature remained within a comfortable range, the AC load was reduced by 5.25%, and the operating energy costs were reduced by 24.94%. These results demonstrate that the proposed method offers improved computational efficiency, enhanced model performance, and economic benefits.

Machine Learning Prediction of Fuel Cell Remaining Life Enhanced by Variational Mode Decomposition and Improved Whale Optimization Algorithm

In predicting the remaining lifespan of Proton Exchange Membrane Fuel Cells (PEMFC), it is crucial to accurately capture the multi-scale variations in cell performance. This study employs Variational Mode Decomposition (VMD) to decompose performance data into intrinsic modes, elucidating critical multi-scale dynamics vital for understanding the complex degradation processes in fuel cells. In addition to VMD, this research utilizes an Improved Whale Optimization Algorithm (IWOA) to optimize a Back Propagation (BP) Neural Network. The IWOA focuses on precise adjustments of weights and biases, enabling the BP network to effectively interpret complex nonlinear relationships within the dataset. This optimization enhances the predictive model’s reliability and stability. Extensive experimental evaluations demonstrate that the integration of VMD, and the learning capabilities of the IWOA-optimized BP network significantly improves the model’s accuracy and stability across multiple predictions, thereby increasing the reliability of lifespan predictions for PEMFCs. This methodology offers a robust framework for extending the operational life and efficiency of fuel cells.

Variance of deviation from optimal leverage

AbstractWe show that deviations from the firm's target leverage are priced in the cross‐section of stock returns and that the relation between these quantities is nonlinear. The concave nonlinear relation between deviation from the target leverage and next‐period return is strong during economic expansions and vanishes during recessions. Our portfolio analysis provides support for the concave relation between deviation from the target leverage and next‐period returns as well. We develop a factor named variance of deviation from optimal leverage (VDOL) and show that it is an important risk factor that has been omitted in the literature.