Search Results (7,271)

Background/Objectives: Periodontitis is a highly prevalent chronic disease requiring accurate diagnosis for effective treatment planning. Artificial intelligence (AI) has emerged as a potential tool to support clinical decision-making. This study aimed to compare the diagnostic performance and classification error patterns of a large language model (LLM) and a neural network (NN) in periodontitis classification according to the current staging and grading system. Methods: This retrospective study included 110 patients with periodontal disease. Clinical and demographic variables (age, sex, smoking status, number of teeth, API, BOP, PPD, and CAL) were analyzed. Reference diagnoses were established by two experts. Cases were evaluated using an LLM and a neural network. Model performance was assessed using accuracy, confusion matrices, and Cohen’s kappa coefficient, along with error analysis. Results: The LLM achieved 62% accuracy for stage and 63% for grade classification (κ = 0.48). The neural network showed higher performance, with 85% accuracy for stage and 79% for grade (κ = 0.79 and κ = 0.67, respectively). The LLM more often underestimated disease severity, whereas the neural network tended to overestimate progression. Differences between models were statistically significant (p < 0.0001). Conclusions: In this dataset and classification task, the task-specific neural network demonstrated higher diagnostic performance than the evaluated large language model. However, the findings should be interpreted in light of the fundamentally different training paradigms and intended applications of these AI systems. Further research is required to optimize and validate AI-based approaches for clinical use. Full article

Fatigue cracking remains a critical challenge in asphalt pavement design, yet conventional prediction methods fail to capture the fundamental damage mechanisms governing failure evolution. This study proposes an innovative residual strain-based approach to predict the fatigue life of crumb rubber modified asphalt (CRMA) mixtures. Through semi-circular bending (SCB) tests under varying aging conditions and stress ratios, a modified Burgers model was employed to decompose residual strain into residual viscoelastic strain (RVES) and residual viscous-flow strain (RVFS) components. The key innovation lies in establishing the residual strain ratio (RSR) as a damage evaluation parameter, with its plateau value (PV) serving as the independent variable in a novel fatigue prediction equation. Results demonstrate that while RVES stabilizes after initial loading, RVFS accumulation drives fatigue damage progression. The RSR-defined damage factor exhibits a distinct three-stage evolution accurately characterized by the ExpAssoc model (R² > 0.97). The proposed PV-based fatigue equation achieves prediction errors below 15% when validated against field core samples, offering a mechanistically sound and practically viable alternative to conventional phenomenological approaches. Full article

There is a growing trend towards vehicles powered by alternative energy sources due to the environmental pollution caused by fossil fuel vehicles. Electric vehicles (EVs) are thought to make a significant contribution to reducing environmental pollution. This study presents a performance comparison of field-oriented control (FOC) and finite control set-based model predictive current control (FCS-MPCC) methods for controlling PMSM motors, which are commonly preferred for EV applications. A multilevel ANPC inverter topology, which has a higher-quality power flow than classical two-level inverters, was preferred to power the PMSM. While the classical FOC method has a fixed switching frequency by including cascaded PI controllers and a pulse width modulation (PWM) modulator, the FCS-MPCC method determines a variable frequency-switching signal that minimizes the cost function by predicting the future current behavior of the PMSM using the mathematical model of the system. The performance comparison of FOC and FCS-MPCC methods was carried out by conducting real-time experimental studies. Both control algorithms were analyzed under variable speed and load conditions using the same motor and drive structure. Performance analysis of FOC and FCS-MPCC control algorithms was carried out in terms of speed tracking, torque, current, and harmonics. According to the results obtained, the total harmonic distortion (THD) value of the stator current was 7.03% in the FOC method, while it was 22.19% in the FCS-MPCC method. Furthermore, a comparative analysis was conducted on the dynamic performance of the two methods in different scenarios using the mean absolute error (MAE), root mean square error (RMSE), integral absolute error (IAE), integrated time absolute error (ITAE), and integral squared error (ISE) criteria. The FCS-MPCC method was observed to be superior in different speed scenarios according to these criteria. In terms of processor load, it was calculated as 17.09% in the FOC method and 63.75% in the FCS-MPCC method. This study is important for determining the control strategy of PMSMs used in EV drives. Full article

High-fidelity computational fluid dynamics (CFD) remains computationally expensive for steady aerodynamic prediction under multi-condition and variable-geometry configurations, which limits rapid design iteration. To address this issue, this study proposes a data-driven surrogate framework for aircraft surface flow-field prediction on irregular meshes. The framework combines a geometry-unification strategy for variable rudder-deflection configurations with KM-GAT, a hybrid neural architecture that integrates graph attention and KAN-based nonlinear feature transformation. Geometry unification maps the surface flow fields associated with different rudder-deflection states onto a common zero-deflection reference template, thereby establishing consistent mesh correspondence and fixed prediction locations across samples while retaining the rudder angle as an operating-condition variable. The KM-GAT model further combines topology-aware message passing with localized nonlinear refinement, while the Huber loss is adopted to improve training robustness for CFD-derived data. Experiments on the F-22 research model show that the proposed framework achieves lower prediction errors and more concentrated error distributions than baseline MLP and GNN-based models. Qualitative comparisons further indicate that KM-GAT better preserves localized high-gradient structures, including pressure transitions and vortex-dominated regions. These results suggest that the proposed framework provides an effective surrogate modeling strategy for variable-geometry aerodynamic flow field prediction. Full article

Road tunnel ventilation systems have traditionally been designed to dilute vehicle-generated pollutants and control smoke during fires. However, the thermal environment, including temperature and humidity, is not the variable taken into consideration. Despite the operation of its ventilation system, Boryeong Subsea Tunnel (6.9 km), the longest subsea road tunnel in Korea, has experienced severe condensation since its opening in December 2021. As hot, humid ambient air enters the tunnel and meets wall surfaces cooled by seawater and the surrounding ground, condensation and fog may form, reducing visibility. To investigate the causes of condensation and develop a decision-making tool for prediction, a variety of tasks were carried out: (1) field measurements of temperature, humidity, tunnel wall temperature, and tunnel air velocity; (2) development of a 1D model for condensation rate quantification; and (3) 3D CFD simulations. Condensation occurred mainly from June to September, with the most severe conditions in July and August. Both the 1D model analysis and the CFD simulations showed good agreement with field measurement data, with wall temperature errors within 7.3%. Under current traffic conditions (with a peak of approximately 250 veh/h), the annual condensation volume was estimated at approximately 12,415 ton/year. Under the design traffic volume (1550 veh/h), heat from vehicles was found to effectively suppress condensation. The Condensation Contour Map (CCM) was developed as a decision support tool to predict the likelihood and amount of condensation based on the tunnel air temperature and humidity conditions. The results of this study clearly indicate that condensation should be explicitly considered in the design and operation of long subsea road tunnels. Full article

This study consists of two main components. The first part establishes a seakeeping assessment method, while the second part focuses on hull-form optimization with seakeeping performance as the objective. For the seakeeping analysis, the Lewis conformal mapping method is used to calculate the sectional hydrodynamic coefficients. Strip theory is then applied to obtain the global hydrodynamic coefficients of the hull. The coupled heave and pitch motion responses are calculated and compared with nonlinear time-domain simulation results and experimental data, showing good agreement. A multivariate linear regression model is established to approximate the relationship between the principal hull-form parameters and the heave and pitch RAOs. The comparison between the regression model and strip theory results shows that the prediction error remains within 5%, indicating that the regression model can provide an efficient surrogate objective function for hull-form optimization. The particle swarm optimization (PSO) algorithm is then employed to optimize the hull form, with the ship length, breadth, draft, and block coefficient considered as design variables. To further evaluate the optimized hull, additional calculations are conducted under different Froude numbers and encounter angles. Under head sea conditions with varying Froude numbers, the optimized hull reduces the peak heave RAO by 11.6–31.1% and the peak pitch RAO by 8.6–17.9%. Under different encounter angles at F_r = 0.3, the reductions in peak heave and pitch RAOs are 31.1–33.9% and 16.5–18.8%, respectively. These results demonstrate that the proposed regression assisted PSO optimization framework can effectively reduce the heave and pitch responses of the Wigley hull under the investigated regular wave conditions. Full article

An empirical model is developed to predict significant wave height and significant wave period using only wind speed at 10 m height, fetch, and water depth. The model distinguishes between fetch-limited and duration-limited sea states within a conditional empirical framework that incorporates modified empirical exponents and corrections into classical wave formulations. Validation was performed using wind and wave data from the Global Forecast System at 26 coastal and offshore stations distributed across eleven different pilot seas and oceans worldwide, encompassing a broad spectrum of marine environments and climatic conditions. The proposed model was benchmarked against established empirical approaches. Results indicate a mean prediction error of 6.6% for the significant wave height and 9.6% for the significant wave period, substantially outperforming conventional formulations whose errors exceed 50% under comparable conditions. Unlike existing empirical models that are restricted to specific regions or sea-state conditions, the proposed model demonstrated strong predictive performance across diverse seas, oceans, and climatic conditions, enabling more reliable wave predictions in data-scarce and dynamically complex marine environments. The developed model was further applied to the Red Sea, where it successfully reproduced the spatial variability of significant wave height and wave period. From the results, it has been found that the developed model provides a practical and transferable tool for wave forecasting, coastal engineering, and offshore renewable energy applications. Full article

Floods remain among the most devastating natural disasters globally, disproportionately impacting low-resource regions where real-time flood forecasting is constrained by limited computational infrastructure and the scarcity of fine-resolution predictive models. Although state-of-the-art global climate models achieve high predictive accuracy, their scale and computational complexity restrict their applicability in localized and resource-constrained settings. This study proposes a deep climate model distillation framework that transfers knowledge from a high-capacity Fourier Neural Operator (FNO)-based global climate model inspired by FourCastNet into lightweight, regionally adaptive student networks suitable for edge deployment. The framework combines climate variables, satellite observations, and hydrological measurements to improve localized flood prediction. Knowledge transfer is achieved through a multi-objective distillation strategy that combines supervised learning, soft-target alignment, and intermediate feature matching. Experimental evaluation across multiple flood-prone regions in Sub-Saharan Africa and South Asia shows that the distilled student model achieves an average classification accuracy of 0.89, an AUC of 0.91, and an F1-score of 0.88, retaining approximately 96.7% of the teacher model’s predictive performance. In continuous discharge estimation, the model attains a mean absolute error of 0.17, RMSE of 0.24, and an R² score of 0.85. The proposed distillation approach yields an 8× reduction in inference latency and over a 20× reduction in model size, enabling real-time execution on low-power edge devices such as the Raspberry Pi 4 and NVIDIA Jetson Nano. The student model further demonstrates robust regional and temporal generalization, with limited performance degradation in unseen geographic areas and during extreme flood years. Full article

Accurate housing price prediction is important for market efficiency and purchasing decisions. However, multicollinearity among independent variables remains a major challenge in linear regression, causing variance inflation and reducing the reliability of the ordinary least squares (OLS) estimator. Although regularization methods such as ridge regression, least absolute shrinkage and selection operator (LASSO), and elastic net (EN) are widely used, evidence regarding their variance behavior under controlled multicollinearity structures remains limited. This study addresses this gap through simulation experiments conducted under controlled correlation structures with sample sizes ranging from 100 to 2000, 5 to 70 independent variables, and correlation coefficients between 0.50 and 0.99. The findings are further validated using the California Housing Dataset, where mean squared prediction error (MSPE) is computed on the full dataset, while root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination (R²) are evaluated on a hold-out test set. Simulation results show that LASSO generally yields the lowest variance estimates under moderate multicollinearity, whereas EN becomes more competitive as multicollinearity and dimensionality increase. In the California Housing application, EN reduces MSPE by approximately 95.5% relative to OLS. These findings provide insight into the behavior of linear regression estimators and offer practical guidance for researchers in selecting appropriate models for housing price modelling. Full article

Hip joint position sense (JPS), a key component of neuromuscular function arising from muscle spindle and periarticular mechanoreceptor input, remains underexplored, with no standardized and reliable clinical protocol available to assess hip proprioception. This study evaluated the intra- and inter-rater reliability of a laser- and inclinometer-based active hip JPS protocol and established preliminary references in healthy adults. A two-phase reliability study was conducted in accordance with GRRAS and COSMIN guidelines: 17 participants for reliability analyses and 57 for preliminary references. Six movement directions were assessed (flexion, extension, abduction, adduction, medial and lateral rotations). Reliability was quantified using intraclass correlation coefficients with their 95% confidence intervals, using two-way random-effects models with absolute agreement (ICC(3,1) for intra-rater and ICC(2,1) for inter-rater analyses), interpreted as poor (<0.50), moderate (0.50–0.70), or good (≥0.70). Absolute measurement error was reported as standard error of measurement (SEM%) and 95% minimal detectable change (MDC95%), normalized to target amplitudes to allow direct cross-direction comparison. Intra-rater reliability ranged from poor to moderate, with experienced raters reaching ICC = 0.64 (95% CI [0.39; 0.80]) for medial rotation. Inter-rater reliability improved across sessions, peaking for medial rotation (ICC = 0.78; 95% CI [0.50; 0.91]). Rotational movements yielded the lowest SEM% (3–6%), indicating high measurement precision despite trial-to-trial variability (MDC% 9–31%). Normative errors were largest in flexion (21.4 cm) and smallest in rotations (≈2.2–2.3°). Despite overall low-to-moderate reliability, the protocol achieved clinically acceptable measurement precision (SEM% < 10%) for rotational tasks, whereas the laser-based sagittal and frontal-plane components remained exploratory. The protocol provides preliminary reference values for hip JPS in healthy adults and requires further validation before clinical use. Full article

Anomaly detection models trained exclusively on healthy data are widely used in wind turbine condition monitoring because failure data are scarce, heterogeneous, and often unavailable. However, these models produce anomaly indicators that are sensitive not only to fault-related degradation but also to normal operational variability, transient disturbances, and changes in loading conditions. As a result, the practical behavior of an alarm system depends not only on the anomaly detection model but also on the decision rule used to activate and maintain alarm states. This study presents a decision-oriented evaluation of persistence-based alarm confirmation in wind turbine anomaly detection. Four representative techniques are analyzed within a unified framework: Isolation Forest, One-Class Support Vector Machine, Referenced Moving Window Principal Component Analysis using Q-statistic and percentage component weight indicators, and Autoencoder-based reconstruction error. The evaluation combines controlled OpenFAST simulations of rotor unbalance under different severity and noise conditions with an industrial SCADA case study involving a documented main bearing fault. Results show that temporal persistence strongly shapes alarm outcomes across methods and datasets. Low persistence values favor early detection but promote alarms from isolated threshold exceedances, whereas moderate persistence substantially reduces false positives while preserving detection capability in severe and well-observable faults. Excessive persistence increases detection latency and missed detections, particularly for weak, intermittent, or slowly evolving fault signatures. These findings indicate that persistence-based alarm confirmation should be treated as an explicit decision-level configuration variable, rather than as a fixed post-processing or alarm-state heuristic, when designing anomaly detection systems for wind turbine condition monitoring. Full article

Against the backdrop of increasingly stringent environmental regulation and increasing uncertainty in supply chain operations, this study examines how environmental penalties affect corporate supply chain resilience. Using Chinese A-share listed firms from 2009 to 2024, this paper constructs a firm-level panel dataset and employs a two-way fixed-effects model to estimate the relationship between environmental penalty intensity and supply chain resilience. Environmental penalty intensity is measured by the annual penalty amount imposed on each firm, while supply chain resilience is captured through an entropy-weighted index reflecting both resistance and recovery capacities. To alleviate endogeneity concerns, this study further uses an instrumental-variable approach based on the interaction between a firm’s one-year lagged penalty amount and city-level thermal inversion days. The results show that environmental penalties reduce corporate supply chain resilience. This negative effect is heterogeneous across firm characteristics and is partially mediated by reduced operational efficiency and crowded-out R&D investment. This conclusion remains robust after replacing the dependent variable, changing the clustering level of standard errors, and excluding observations from the COVID-19 pandemic period. Mechanism tests suggest that environmental penalties weaken supply chain resilience partly by reducing operational efficiency and crowding out R&D investment. Heterogeneity analysis indicates that the negative effect is more pronounced among young firms, non-high-tech firms, and firms located in regions with lower environmental regulation intensity. This study contributes to the literature by distinguishing environmental penalties from broader environmental regulation and by examining their implications for supply chain resilience. The findings also suggest that environmental enforcement should maintain deterrence while improving transparency, predictability, and targeted compliance guidance. Full article

This review examines the emerging role of machine learning (ML) in pharmaceutical chemistry, with emphasis on molecular design, synthetic feasibility, and structure–property–performance (SPP) relationships. By enabling pre-synthesis prediction of physicochemical properties, reaction pathways, and pharmaceutical performance, ML can reduce empirical trial-and-error experimentation and support more efficient exploration of chemical space. A structured narrative review design with PRISMA-aligned systematic search elements was used to evaluate 101 studies, enabling transparent literature identification, eligibility screening, and thematic synthesis across heterogeneous ML applications in pharmaceutical chemistry. This review examines structure–property relationships (SPRs) and property–performance relationships (PPRs), with emphasis on key pharmaceutical endpoints such as solubility, permeability, stability, dissolution, and bioavailability. An integrated SPP framework is proposed to connect molecular structure, intermediate properties, and final performance outcomes while incorporating retrosynthetic analysis and experimental feedback and closed-loop optimization. Recent frontier developments are also discussed, including molecular foundation models, multimodal language–graph models, diffusion-based molecular generation, E(3)-equivariant models, and MolMIM-like latent-space optimization. This review also covers co-folding and joint ligand–protein modeling, Boltz-2-like affinity prediction, AlphaFold 3-related biomolecular interaction modeling, and absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction. Key limitations include dataset leakage, benchmark inconsistency, assay variability, conformational and protonation-state effects, reproducibility challenges, regulatory constraints, and the gap between computational prediction and prospective experimental validation. Future progress is expected to depend on hybrid physics–ML models, uncertainty-aware prospective validation, autonomous experimentation, explainable artificial intelligence, and sustainability-aware molecular design. Overall, ML is evolving from a predictive tool into a chemically informed decision-support framework for rational, synthesis-aware, and experimentally validated pharmaceutical development. Full article

The increasing penetration of renewable energy sources introduces significant variability, low-inertia behaviour, and operational uncertainty into modern power systems, resulting in frequent frequency deviations and degraded dynamic stability. Conventional Load Frequency Control (LFC) approaches based on fixed-parameter PID controllers often exhibit limited disturbance rejection capability under nonlinear and stochastic operating conditions. This study proposes an enhanced LFC framework that integrates a PID controller optimised using the Flower Pollination Algorithm (FPA) with support from a Unified Power Flow Controller (UPFC) and a Redox Flow Battery (RFB) to improve frequency regulation, damping, and robustness in renewable-integrated low-inertia power systems. This study developed a MATLAB/Simulink single-area power system model comprising governor, turbine, and generator-load dynamics to evaluate controller performance under a 0.01 pu step disturbance, stochastic load variations, renewable energy fluctuations, and ±20% parameter uncertainty conditions. The FPA optimally tuned the PID controller gains using the Integral Time Absolute Error criterion to enhance transient response and disturbance rejection capability. Comparative analyses were conducted against conventional PID and fuzzy-based controllers using settling time, overshoot, RMS deviation, ITAE, and mean frequency deviation indices. Simulation results demonstrate that the proposed FPA–PID + UPFC framework significantly outperforms the conventional PID controller by achieving approximately 66.6% settling-time reduction, 72.1% RMS reduction, and 75.5% ITAE reduction. The proposed framework reduced settling time from 18.46 s to 6.16 s and substantially improved damping performance under stochastic disturbances. The coordinated integration of the UPFC and RFB further enhanced transient stability through dynamic power-flow regulation and rapid active-power compensation during disturbances. Sensitivity analysis under parameter uncertainty and stochastic operating conditions confirmed stable and reliable operation under stochastic disturbances and parameter uncertainty conditions. The proposed architecture, therefore, provides an effective, practically applicable solution for secondary frequency regulation in renewable-rich smart grids, low-inertia transmission systems, microgrids, and future distributed power networks. Full article

Nonparametric inference for residual-life functionals is a fundamental problem in mathematical statistics, reliability theory, and survival analysis, particularly in studies with limited sample sizes where empirical plug-in estimators may exhibit substantial sampling variability. In comparative lifetime analysis, additional qualitative information is often available regarding the relative behavior of two populations; however, such information is frequently too weak to justify classical stochastic dominance assumptions. Stochastic precedence provides a natural and interpretable framework for representing this partial ordering through a pairwise probabilistic constraint. This paper develops a constraint-adjusted nonparametric inference framework for estimating the mean residual life (MRL) and quantile residual life (QRL) functions under stochastic precedence information. The proposed approach replaces the ordinary empirical distribution function in standard residual-life plug-in estimators with a constraint-adjusted empirical distribution function that enforces the stochastic precedence relation at the sample level. The adjustment is governed by a data-driven scaling factor and is asymptotically negligible, thereby preserving the large-sample behavior of the ordinary empirical estimators while incorporating meaningful structural information in finite samples. Strong consistency of the proposed MRL and QRL estimators was established under mild regularity conditions. A Monte Carlo study based on Weibull and gamma lifetime models demonstrates that in the simulation settings considered, the proposed estimators provide improved finite-sample stability and generally achieve smaller mean squared errors than their ordinary empirical counterparts, especially for small and moderate sample sizes. The methodology is further illustrated using survival data from patients with squamous cell carcinoma of the oropharynx, highlighting its practical relevance in biomedical survival analysis. The proposed method offers a flexible, interpretable, and computationally simple framework for nonparametric inference with structured lifetime data under weak stochastic ordering information. Full article