Applied Sciences

28 pages, 2272 KB

Open AccessArticle

A Novel Gene Expression Programming Algorithm for Forecasting Carbon Dioxide Emissions in G7 Countries

by Kasım Zor, Ali Can Ozdemir and Iclal Cetin Tas

Appl. Sci. 2026, 16(10), 4676; https://doi.org/10.3390/app16104676 - 8 May 2026

The increase in the carbon dioxide (CO₂) emissions, nearly a quarter of those originating from the G7 countries, threatens not only the sustainability of the Earth but also the lives of future generations of humanity. Shedding light on future projections of [...] Read more.

The increase in the carbon dioxide (CO₂) emissions, nearly a quarter of those originating from the G7 countries, threatens not only the sustainability of the Earth but also the lives of future generations of humanity. Shedding light on future projections of the CO₂ emissions is vital in achieving the target of carbon neutrality, and machine learning-based algorithms are frequently applied to forecast the CO₂ emissions in the literature. However, the majority of these algorithms create model equations that are abstruse and irreproducible. In the current study, a novel gene expression programming (GEP) algorithm is proposed to produce genuine and easily understandable mathematical models for forecasting the CO₂ emissions of the G7 countries. The proposed algorithm is comprehensively compared with both the simple GEP and the previous studies in terms of several error metrics and computational time. Consequently, the obtained results unveiled that the proposed algorithm surpassed the simple GEP by the improvements of 26% in nMAE, 24% in nRMSE, and 27% in MAPE, respectively. Notably, the proposed algorithm maintains essentially the same computational efficiency as the simple GEP (a 0.2% difference in duration) despite its richer function set. In addition to those, the estimated model equations belonging to the year of 2035 were meticulously presented to guide the researchers in the field for the sake of applicability and reproducibility. Full article

► Show Figures

Figure 1

25 pages, 1909 KB

Open AccessArticle

Shale Cap Breakthrough Pressure Prediction Method Based on Machine Learning

by Huanping Wu, Meiling Zhang, Zheng Wu and Zongli Liu

Appl. Sci. 2026, 16(10), 4675; https://doi.org/10.3390/app16104675 - 8 May 2026

Abstract

Breakthrough pressure (BP) is a key parameter for evaluating the sealing capacity of shale caprocks, whereas direct laboratory measurements are time-consuming and costly, limiting their use in continuous regional assessment. This study develops a conventional-log-based workflow for BP prediction in the Quan-4 Member [...] Read more.

Breakthrough pressure (BP) is a key parameter for evaluating the sealing capacity of shale caprocks, whereas direct laboratory measurements are time-consuming and costly, limiting their use in continuous regional assessment. This study develops a conventional-log-based workflow for BP prediction in the Quan-4 Member (K1q4) caprock underlying the Qingshankou shale oil interval in the Gulong Sag, Songliao Basin. Four routinely available logs—gamma ray (GR), acoustic interval transit time (AC), bulk density (DEN), and compensated neutron log (CNL)—were integrated with core-measured BP data. A GR-AC multiple-regression baseline and five machine-learning algorithms, including stochastic gradient descent (SGD), extremely randomized trees (ERT), Random Forest (RF), extreme gradient boosting (XGBoost), and adaptive boosting (AdaBoost), were compared under a unified workflow. The training set was used for normalization, model fitting, grid search, and internal five-fold cross-validation, whereas the held-out test set and external prediction wells were kept separate for performance evaluation. The results show that BP generally increases with GR and DEN and decreases with AC and CNL, indicating that clay content, compaction, and pore connectivity jointly control the logging response of caprock sealing capacity. Among the evaluated models, AdaBoost achieved the best overall performance, with RMSE, MAE, and R² values of 1.33 MPa, 0.97 MPa, and 0.89 on the held-out test set, and 1.19 MPa, 0.92 MPa, and 0.94 in external prediction wells. Train–test diagnostics, learning curves, and SHAP analysis indicate stable performance and physically plausible feature contributions within the present dataset. The proposed workflow can therefore provide a practical supplement to laboratory BP measurements for caprock evaluation in the study area, although broader application still requires well-level independent validation and explicit prediction-uncertainty quantification. Full article

(This article belongs to the Section Earth Sciences)

40 pages, 3212 KB

Open AccessArticle

An Empirical Study of Spatial and Spectral Feature Fusion for Robust Lung Cancer Histopathology Classification Under Domain Shift and Image Perturbations

by Pavan Kumar Illa and Senthil Kumar Thillaigovindan

Appl. Sci. 2026, 16(10), 4674; https://doi.org/10.3390/app16104674 - 8 May 2026

Abstract

Deep learning has demonstrated high efficiency in histopathological image analysis, particularly in lung cancer classification. However, the stability of these models with image corruption and cross-dataset validation remains an important practical concern. In this study, we explored the potential of adding spectral information [...] Read more.

Deep learning has demonstrated high efficiency in histopathological image analysis, particularly in lung cancer classification. However, the stability of these models with image corruption and cross-dataset validation remains an important practical concern. In this study, we explored the potential of adding spectral information derived from the discrete wavelet transform (DWT) and spatial convolutional representations to enhance the robustness of multi-class lung cancer classification between Normal, Adenocarcinoma and Squamous cell carcinoma. The lightweight ResNet18 backbone was used to obtain spatial features, and spectral descriptors were obtained through wavelet sub-bands and integrated through early feature-level fusion. The models were trained and evaluated using the LC25000 dataset. Subsequently, it was tested under controlled perturbations, such as Gaussian noise and Gaussian blur. Three random seeds were used to assess performance variability, and paired t-tests were conducted as an indicative statistical measure of the results. Under clean conditions, the spatial and hybrid models were nearly saturated, and there was no significant difference between them (spatial: 99.85 ± 0.26; hybrid: 99.72 ± 0.22; p = 0.1217). The hybrid model exhibited higher robustness when Gaussian noise (σ = 0.05) was added, which resulted in 84.89% ± 4.52% accuracy versus 74.99% ± 7.20% of the spatial baseline (p = 0.0443) with an observed effect size (Cohen’s d = 2.64), noting that these estimates are based on a limited number of runs and should be interpreted with caution. The same behavior was observed in Gaussian blur perturbations, where the hybrid representation was slightly more stable. We also investigated a simplified adaptive gating mechanism process and found that the learned gate parameter also tends to converge towards spatial feature dominance with a model trained with clean data. Finally, cross-dataset validation with LungHist700 showed a slight increase in the balanced accuracy of the hybrid model (0.5158) over the spatial baseline (0.4722). These results indicate that spectral and spatial features can be used to enhance robustness to image corruption and still yield high classification accuracy, indicating that spectral–spatial representations can improve robustness under controlled perturbations, whereas their impact on cross-dataset generalization remains limited. The results further indicate that robustness improvements are strongly influenced by training strategies, such as noise augmentation, whereas the contribution of fusion is comparatively moderate. Full article

24 pages, 409 KB

Open AccessArticle

Enhancing Cross-Dataset Mental Workload Detection Using Electrodermal Activity and Domain Adaptation

by Luis Sigcha, Eduarda Pereira, Luigi Borzì, Diego Gachet, Paulo Cardoso and Nélson Costa

Appl. Sci. 2026, 16(10), 4673; https://doi.org/10.3390/app16104673 - 8 May 2026

Abstract

Mental workload assessment using physiological signals has gained increasing attention for applications in human–computer interaction and occupational monitoring. Among these signals, electrodermal activity (EDA) is widely recognised as a reliable indicator of sympathetic activation associated with cognitive effort. However, most existing machine learning- [...] Read more.

Mental workload assessment using physiological signals has gained increasing attention for applications in human–computer interaction and occupational monitoring. Among these signals, electrodermal activity (EDA) is widely recognised as a reliable indicator of sympathetic activation associated with cognitive effort. However, most existing machine learning- based approaches are evaluated within a single dataset, limiting their generalisability across different populations and experimental conditions. This study investigates the cross-dataset performance of machine learning models for mental workload detection using EDA features. Two independent datasets were employed, and a cross-dataset evaluation framework was adopted to simulate realistic deployment scenarios under domain shift. Three classifiers (Random Forest, XGBoost, and Support Vector Classifier (SVC)) were evaluated, together with two domain adaptation techniques: Correlation Alignment (CORAL) and Subspace Alignment (SA). The results show that model performance is strongly dependent on the direction of transfer, with a notable performance drop when generalising across datasets. Domain adaptation improved performance in several configurations, particularly for SVC with CORAL, achieving the best overall F1-score (0.815). However, improvements were not consistent across all models and target domains. Overall, this study highlights the challenges of cross-dataset generalisation in EDA-based workload detection and demonstrates the potential, yet limited robustness, of domain adaptation techniques in mitigating distribution shifts. Full article

(This article belongs to the Special Issue Emerging Technologies of Accident Analysis and Prevention in Safety Engineering)

19 pages, 3179 KB

Open AccessArticle

Localized Resonance Mechanism of Rail Corrugation and Active Suppression via Wheel–Rail Self-Grinding on Urban Express Line with Different Tracks

by Jie Zhong, Jing Tong, Chunqiang Shao, Chaozhi Ma and Peng Zhou

Appl. Sci. 2026, 16(10), 4672; https://doi.org/10.3390/app16104672 - 8 May 2026

Abstract

The occurrence of short-wave corrugation with wavelengths of 32–44 mm on curved sections of urban express railway lines is particularly pronounced, yet the underlying initiation mechanisms have remained insufficiently understood. Furthermore, conventional mitigation strategies—including the installation of rail dampers and passive grinding—entail substantial [...] Read more.

The occurrence of short-wave corrugation with wavelengths of 32–44 mm on curved sections of urban express railway lines is particularly pronounced, yet the underlying initiation mechanisms have remained insufficiently understood. Furthermore, conventional mitigation strategies—including the installation of rail dampers and passive grinding—entail substantial maintenance expenditures, thereby hindering their large-scale application. To elucidate the initiation mechanisms of rail corrugation and to formulate effective control measures, the characteristic corrugation parameters under various track structure configurations across an entire alignment were first measured and systematically analyzed. Dynamic interaction models between vehicles and three distinct track typologies were subsequently developed, together with a comprehensive analytical framework for corrugation evolution. The wheel–rail dynamic response characteristics and corrugation growth rates corresponding to each track type were examined, and the wheel–rail coupled vibration modes that exacerbate corrugation propagation in urban express lines were identified. The instantaneous wear behavior of the rail under differing creep regimes was also investigated, leading to the proposal of a novel self-mitigating approach for rail corrugation. The results demonstrate that the excitation frequency of rail corrugation is predominantly confined to the 600–700 Hz range, exhibiting a fixed-frequency characteristic that remains invariant with respect to curve radius, track structure type, and operational speed. An interesting finding is that, although the intrinsic vibration properties of different track structures diverge significantly, the third-order bending resonance of the rail segment situated between bogie wheels is largely unaffected by track-borne vibrations and manifests as a localized wheel–rail resonance within the vehicle–track coupled system. This particular resonance markedly accelerates corrugation development and is identified as the critical governing factor for corrugation initiation in urban express lines, regardless of the underlying track configuration. Furthermore, rail instantaneous wear displays a substantial phase shift under varying creep conditions, with the wear profiles under creep saturation (full sliding) and low creep (rolling–sliding) exhibiting a distinct anti-phase relationship. This insight underpins a novel self-wear suppression strategy: by intentionally mixing rolling–sliding and full-sliding operational regimes, destructive interference between the out-of-phase wear contributions is achieved, resulting in a considerably attenuated corrugation growth rate compared with exclusive rolling–sliding operation. This methodology thus offers a promising and fundamentally new alternative for the long-term management of rail corrugation through intrinsic wheel–rail interaction. Full article

(This article belongs to the Special Issue Advances in Tunnel Excavation and Underground Construction)

26 pages, 3290 KB

Open AccessArticle

DEGC-TransUNet: A Dual-Encoder TransUNet with Global Context Enhancement for Mountaintop Area Extraction from Grid DEMs

by Fangbin Zhou, Junwei Bian and Jiamin Huang

Appl. Sci. 2026, 16(10), 4671; https://doi.org/10.3390/app16104671 - 8 May 2026

Abstract

Accurate extraction of mountaintop areas from grid digital elevation models (DEMs) is essential for terrain analysis, geomorphological research, hydrological modeling, natural disaster monitoring, and emergency communication site selection. However, existing deep-learning-based methods often suffer from inadequate representation of local details and limited global [...] Read more.

Accurate extraction of mountaintop areas from grid digital elevation models (DEMs) is essential for terrain analysis, geomorphological research, hydrological modeling, natural disaster monitoring, and emergency communication site selection. However, existing deep-learning-based methods often suffer from inadequate representation of local details and limited global contextual awareness, leading to blurred boundaries and reduced segmentation accuracy in complex mountainous terrains. To address these limitations, this study proposes a dual-encoder and global-context-enhanced TransUNet framework, named DEGC-TransUNet, for automated mountaintop delineation. The architecture integrates a convolutional encoder to capture fine-grained local terrain features and a MaxViT-based encoder to model multi-scale global context by encoding low-dimensional topographic attributes such as slope and curvature. A dedicated feature fusion module harmonizes complementary representations from both encoding paths, while a BiFormer-based strategy is introduced at the bottleneck to strengthen long-range dependencies and enhance convergence. The experimental results demonstrate that DEGC-TransUNet significantly outperforms baseline models such as TransUNet, DE-TransUNet, and GC-TransUNet, with relative improvements of 19.8% in Intersection over Union (IoU), 10.4% in overall accuracy (ACC), and 10.9% in F1-score. These findings provide a robust solution for mountaintop extraction, with significant potential in analyzing geomorphological evolution, simulating soil erosion, modeling species distribution in “sky island” ecosystems, and optimizing strategic placements for communication base stations and wind energy infrastructures. Full article

(This article belongs to the Section Earth Sciences)

► Show Figures

Figure 1

22 pages, 1144 KB

Open AccessArticle

A Study on Improving the Accuracy of Accident Reports Through Event-Based Information Structuring of Accident Occurrence Processes

by Jung Nam Kim, Young Beom Kwon and Jong Yill Park

Appl. Sci. 2026, 16(10), 4670; https://doi.org/10.3390/app16104670 - 8 May 2026

Abstract

Industrial accident reporting systems provide the foundation for establishing future prevention strategies by collecting and analyzing accident-related data. While some industrial accidents occur as isolated events, many exhibit a process-oriented nature in which a sequence of temporally connected events accumulates and ultimately leads [...] Read more.

Industrial accident reporting systems provide the foundation for establishing future prevention strategies by collecting and analyzing accident-related data. While some industrial accidents occur as isolated events, many exhibit a process-oriented nature in which a sequence of temporally connected events accumulates and ultimately leads to a final accident. Nevertheless, a substantial proportion of accident reports are prepared by injured workers or employers who lack specialized safety knowledge. As a result, critical information about the conditions, procedures, and actions involved in accident progression is often insufficiently documented. Such information loss hinders a comprehensive understanding of accident causation and, consequently, reduces the effectiveness of preventive measures. To address this limitation, this study proposes an event-based accident information reporting framework that enables injured workers and employers without professional safety expertise to record accidents in a structured manner following their temporal sequence. The proposed framework defines the observed actions and conditions throughout the accident occurrence process as a series of discrete “events,” each of which is classified by an occurrence type. Furthermore, each occurrence type is linked to a corresponding object that reflects its characteristics, allowing accident components to be described in a standardized and systematic form. The framework is designed to be easily completed through a simple selection-and-entry process centered on occurrence types, thereby facilitating consistent and uniform reporting. When applied to 462 fatal industrial accident cases that occurred in South Korea in 2018, the proposed method indicated that approximately 55% of accidents involved multi-stage event sequences, highlighting the importance of process-related information that is not captured by conventional outcome-centered classification systems. In addition, the distribution of occurrence types differed substantially from patterns observed in existing reporting practices. The structured reporting approach proposed in this study may contribute to the preservation and accumulation of essential information on accident occurrence processes, thereby supporting more effective accident prevention efforts. This study does not propose a new investigation methodologies. Instead, it aims to improve accident reporting quality at the data input stage. Full article

► Show Figures

Figure 1

26 pages, 1140 KB

Open AccessReview

Relationship Between the Risk of Cardiovascular Disease and Mitochondrial Dysfunction

by Ida Manna, Annamaria Cerantonio, Federico Rocca, Antonio Cerasa, Luigi Citrigno and Domenico Bosco

Appl. Sci. 2026, 16(10), 4669; https://doi.org/10.3390/app16104669 - 8 May 2026

Abstract

Cardiovascular diseases (CVDs) remain a leading cause of death worldwide. Mitochondria, essential organelles within cells, play a crucial role in maintaining cardiovascular health by producing energy through ATP synthesis. The heart’s high energy demand makes it particularly sensitive to mitochondrial function. In CVDs, [...] Read more.

Cardiovascular diseases (CVDs) remain a leading cause of death worldwide. Mitochondria, essential organelles within cells, play a crucial role in maintaining cardiovascular health by producing energy through ATP synthesis. The heart’s high energy demand makes it particularly sensitive to mitochondrial function. In CVDs, mitochondrial adaptability is compromised, resulting in dysfunction characterized by impaired respiratory chain activity, decreased ATP production, oxidative stress, and structural damage. This review consolidates current research on mitochondrial roles in CVD development, focusing on mitochondrial respiration, ATP synthesis, and the processes involved in maintaining mitochondrial quality, such as mitophagy. It discusses the challenges in developing therapies aimed at restoring mitochondrial function, including drug delivery issues and targeting specificity. The assessment includes analysis of mitochondrial anomalies associated with cardiac disease progression and potential therapeutic strategies. Mitochondrial dysfunction contributes to the progression of various CVDs by reducing energy output and increasing oxidative stress, leading to cardiomyocyte injury and death. Damaged mitochondria produce excessive reactive oxygen species (ROS), exacerbating cellular damage. Repairing mitochondrial components, especially the respiratory chain and ATP synthesis pathways, has shown potential in mitigating cellular injury and improving cardiac function. Restoring mitochondrial function is vital for preventing and treating CVDs. Targeted therapies that repair mitochondrial respiratory activity and enhance ATP production may reduce cellular damage, promote cardiomyocyte survival, and improve clinical outcomes. Understanding mitochondrial dynamics offers promising avenues for innovative interventions in cardiovascular health management. Full article

(This article belongs to the Special Issue Diagnosis and Pharmacological Treatment of Neurological Diseases)

15 pages, 3774 KB

Open AccessArticle

Hybrid Analytical–Numerical Modeling and Dynamic Response Evaluation of Vehicle–Track–Tunnel–Soil System

by Yuwang Liang, Hao Xu, Tao Wang, Zonghao Yuan and Fengxi Zhou

Appl. Sci. 2026, 16(10), 4668; https://doi.org/10.3390/app16104668 - 8 May 2026

Abstract

With the rapid development of urban rail transportation, environmental vibrations caused by subways operating are becoming more serious. They have affected people’s living comfort, the stability of precision scientific instruments, and the safety of buildings. In order to make a good and real [...] Read more.

With the rapid development of urban rail transportation, environmental vibrations caused by subways operating are becoming more serious. They have affected people’s living comfort, the stability of precision scientific instruments, and the safety of buildings. In order to make a good and real prediction of metro-induced vibration, this paper constructs a kind of analytical–numerical hybrid method to evaluate the dynamic response of the whole vehicle–track–tunnel–soil system. First, an analytical metro vehicle–track coupled dynamic model is established according to the Zhai method, and then the wheel–rail force acting on trains can be acquired. And then we use a 3D numeric model which incorporates the track, tunnel and ground with a finite element method. At last, the wheel–rail force is used as an excitatory load, and then the train–track–tunnel–ground coupling system dynamic response analytic–numeric model can be created. Based on this, the influence of track quality level, train speed and tunnel burial depth on the characteristics of ground surface vibration is studied in sequence. From the results, we can see that given the same track quality condition, increasing train speed greatly increases the response amplitude on the ground surface; improving track quality is very effective to reduce response amplitude; and with the tunnel buried deeper, the vibration level would be obviously reduced, local peak response would be suppressed and deep tunnels would have better vibration suppression under higher-speed operation conditions. Full article

(This article belongs to the Section Civil Engineering)

► Show Figures

Figure 1

23 pages, 725 KB

Open AccessArticle

A Comprehensive Assessment of UPFC-Based Power Flow Control for Voltage Stability Enhancement in Large-Scale Power Systems

by Mohammed Mirghani Hassan, Mohammed Gmal Osman and Gheorghe Lazaroiu

Appl. Sci. 2026, 16(10), 4667; https://doi.org/10.3390/app16104667 - 8 May 2026

Abstract

This study presents a comprehensive investigation into the optimal deployment of Unified Power Flow Controllers (UPFCs) to enhance voltage stability and reduce power losses in the Sudanese national grid. With the increasing demand for electricity driven by population growth, urban expansion, and industrial [...] Read more.

This study presents a comprehensive investigation into the optimal deployment of Unified Power Flow Controllers (UPFCs) to enhance voltage stability and reduce power losses in the Sudanese national grid. With the increasing demand for electricity driven by population growth, urban expansion, and industrial development, modern power systems require advanced control strategies to ensure reliable and efficient operation. In this work, the Line Stability Index (Lmn) is employed as a key indicator to identify the most critical transmission lines prone to voltage instability. Based on this index, optimal locations for UPFC installation are determined. Furthermore, an Optimal Power Flow (OPF) framework is utilized to calculate the control parameters of the UPFC devices, aiming to minimize system losses while maintaining operational constraints. The proposed methodology is validated using a real large-scale network model of the Sudanese power system implemented in MATLAB (24b) and NEPLAN (v10) environments. The results demonstrate that installing seven UPFC devices leads to a significant improvement in voltage profiles, maintaining all bus voltages within ±5% of nominal values. Additionally, the system experiences a reduction in total active and reactive power losses by 6.96% and 0.74%, respectively. These findings highlight the effectiveness of UPFC-based control strategies in improving system stability, enhancing transmission efficiency, and supporting the integration of future energy resources. Full article

20 pages, 1238 KB

Open AccessArticle

Screed Mortars Containing Recycled Plastic Waste: Influence on Physical Properties and Durability

by Alejandra Vidales-Barriguete and Carolina Piña Ramírez

Appl. Sci. 2026, 16(10), 4666; https://doi.org/10.3390/app16104666 - 8 May 2026

Abstract

Screed mortars are extensively used in construction, yet their durability and environmental footprint remain key challenges. This study evaluates the effects of partially replacing natural sand with polymeric waste aggregates (25–55% by volume) on the mechanical, hygric, and deformation-related properties of cementitious screed [...] Read more.

Screed mortars are extensively used in construction, yet their durability and environmental footprint remain key challenges. This study evaluates the effects of partially replacing natural sand with polymeric waste aggregates (25–55% by volume) on the mechanical, hygric, and deformation-related properties of cementitious screed mortars. The proposed material solution, protected under patent ES2973008, results in a significant reduction in density of up to 26.87% while decreasing natural aggregate consumption by as much as 55%, improving workability and ease of application. Experimental results indicate reductions in flexural and compressive strength with increasing polymer content; however, the obtained strength levels remain suitable for self-leveling mortars and applications subjected to pedestrian traffic or light loads. In contrast, the incorporation of polymeric aggregates leads to marked improvements in durability-related parameters, including reductions in drying shrinkage of up to 25.2%, Young’s modulus of up to 73%, capillary water absorption of up to 72.34%, and water vapour permeability of up to 6.53%. These combined effects reflect a pronounced increase in elastic deformation capacity, dimensional stability, and resistance to moisture ingress, thereby reducing susceptibility to shrinkage-induced cracking and freeze–thaw damage. Overall, the results demonstrate that polymeric waste incorporation enables the development of lighter, more crack-resistant, and more durable screed mortars, achieving a favourable balance between mechanical performance and long-term durability while contributing to sustainability and circular economy objectives. Full article

(This article belongs to the Special Issue Recent Advances in Sustainable Construction Materials and Structures)

► Show Figures

Figure 1

17 pages, 23699 KB

Open AccessArticle

Effects of Crossflow Air on Conical Water Spray Structure Using a Laser-Based Imaging Method

by Dariusz Obracaj, Paweł Deszcz, Waldemar Wodziak and Jacek Sobczyk

Appl. Sci. 2026, 16(10), 4665; https://doi.org/10.3390/app16104665 - 8 May 2026

Abstract

The interaction between crossflows from sprinkler nozzles and airflow is crucial for engineering applications, particularly affecting the efficiency of sprayed areas. This study investigates the deformation of a continuously injected conical water spray subjected to horizontal airflow, using a planar laser imaging method [...] Read more.

The interaction between crossflows from sprinkler nozzles and airflow is crucial for engineering applications, particularly affecting the efficiency of sprayed areas. This study investigates the deformation of a continuously injected conical water spray subjected to horizontal airflow, using a planar laser imaging method as a visualisation technique. Experiments were conducted in a wind tunnel at a constant water pressure of 0.2 MPa and four airflow rates (0.1, 0.2, 0.4, and 0.6 m³·s⁻¹) to systematically vary the air-to-water momentum ratio. A grayscale-based analysis method was developed using a per-pixel Look-Up Table (LUT), enabling indirect assessment of droplet concentrations and spray structure. This approach allowed for a detailed examination of changes in the spray cone shape under flowing air. By assessing the water spray across three vertical planes intersecting the spray cone, it became possible to calculate lateral area and cone volume at different air-to-water mass flow ratios. The spray formation region exposed to airflow exhibited larger cone volumes than those with minimal airflow. The changes in apparent spray angles for the tested nozzles were determined to characterize the cone shape. The apparent spray angle varies systematically with the air-to-water mass flow ratio, confirming the dominant role of aerodynamic forces. These findings improve the understanding of spray behavior under crossflow and provide a basis for validating numerical models of air–water interactions. Full article

(This article belongs to the Section Fluid Science and Technology)

► Show Figures

Figure 1

24 pages, 13456 KB

Open AccessArticle

Dual-Subspace Network for Few-Shot Fine-Grained Image Classification

by Meijia Wang, Guochao Wang, Haozhen Chu, Bin Yao, Weichuan Zhang, Yuan Wang and Junpo Yang

Appl. Sci. 2026, 16(10), 4664; https://doi.org/10.3390/app16104664 - 8 May 2026

Abstract

Few-shot fine-grained image classification aims to recognize subcategories with high visual similarity using only a limited number of annotated samples. Existing metric learning-based methods typically rely solely on spatial-domain features. Confined to this single perspective, models inevitably suffer from inherent texture biases, entangling [...] Read more.

Few-shot fine-grained image classification aims to recognize subcategories with high visual similarity using only a limited number of annotated samples. Existing metric learning-based methods typically rely solely on spatial-domain features. Confined to this single perspective, models inevitably suffer from inherent texture biases, entangling essential structural details with high-frequency background noise. Furthermore, lacking cross-view geometric constraints, single-view metrics tend to overfit this noise, resulting in structural instability under few-shot conditions. To address these issues, this paper proposes the Dual-Subspace Network (DSNet). Specifically, DSNet utilizes the discrete cosine transform (DCT) and a low-pass filtering mechanism to explicitly isolate low-frequency global structural components from spatial features, thereby suppressing background interference. Truncated Singular Value Decomposition (SVD) is employed to construct independent, low-rank linear subspaces for both spatial texture and frequency structural features. An adaptive gating mechanism is designed to dynamically fuse the projection distances from these dual views. This strategy leverages the structural stability of the frequency subspace to prevent the spatial subspace from overfitting to background features. Extensive experiments on four benchmark datasets—CUB-200-2011, Stanford Cars, Stanford Dogs, and FGVC-Aircraft—demonstrate that DSNet exhibits excellent classification performance and robustness, achieving highly competitive results compared to existing metric learning algorithms. Complexity analysis further confirms that the proposed network achieves a favorable balance between high accuracy and computational efficiency, providing an effective new paradigm for few-shot fine-grained visual recognition. Full article

(This article belongs to the Section Computing and Artificial Intelligence)

23 pages, 3015 KB

Open AccessArticle

Effects of Fin Length on Frosting and Defrosting Characteristics of Small-Diameter Copper Tube-Fin Heat Exchangers

by Dalong Liang and Wenbin Cui

Appl. Sci. 2026, 16(10), 4663; https://doi.org/10.3390/app16104663 - 8 May 2026

Abstract

Frost buildup on copper tube-fin heat exchangers reduces their performance in cold, humid conditions. Fin length plays a key role in balancing heat transfer and frost resistance. This study experimentally examines how fin length affects thermal and frosting behavior. Four heat exchangers with [...] Read more.

Frost buildup on copper tube-fin heat exchangers reduces their performance in cold, humid conditions. Fin length plays a key role in balancing heat transfer and frost resistance. This study experimentally examines how fin length affects thermal and frosting behavior. Four heat exchangers with fin lengths of

15.1

m

m

,

18.53

m

m

,

20.3

m

m

, and

23.5

m

m

were tested at 2

^{\circ} C

/ 1

^{\circ} C

dry-bulb/wet-bulb air temperature and

- 6

^{\circ} C

coolant temperature under constant static pressure. Results show that longer fins increase total heat transfer—peak capacity rose from 512

W

to 566

W

—but reduce heat transfer per unit area by about 30%. Operating time before defrosting increased by 30.6%, from

45.7

\min

to

59.6

\min

, due to lower frost density. Total frost mass grew, but unit-area frost decreased by 12.7%. During defrosting, longer fins achieved greater absolute airflow recovery (from 195 to 213

m^{3} / h

), though defrosting efficiency per gram of frost declined. Short fins ( 15

m

m

) suit space-limited systems needing high surface efficiency. Long fins ( 23

m

m

) benefit large systems requiring long run times and strong post-defrost performance. Medium lengths ( 17

m

m

to 20

m

m

) offer a practical balance for general use. These findings support better heat exchanger design in frost-prone applications. Full article

(This article belongs to the Section Applied Thermal Engineering)

31 pages, 19944 KB

Open AccessArticle

1dMC-MPR-SABinet: A UAV Rotor Blade Crack Fault Diagnosis Method Based on Vibration Signals

by Taochuan Zhang, Huiyuan Huang, Jiahui Fu, Qiang Liu and Jingliang Lin

Appl. Sci. 2026, 16(10), 4662; https://doi.org/10.3390/app16104662 - 8 May 2026

Abstract

In recent years, the application scenarios of Unmanned Aerial Vehicles (UAVs) have become increasingly widespread. As core components of UAVs, rotor blades’ health status is directly related to flight safety. Aiming at issues such as insufficient feature extraction, weak noise resistance, and low [...] Read more.

In recent years, the application scenarios of Unmanned Aerial Vehicles (UAVs) have become increasingly widespread. As core components of UAVs, rotor blades’ health status is directly related to flight safety. Aiming at issues such as insufficient feature extraction, weak noise resistance, and low diagnostic accuracy in the crack fault diagnosis of UAV rotor blades, this study proposes a one-dimensional deep network integrating multi-scale convolution, a multi-path residual module, BiLSTM, and a self-attention mechanism, referred to as 1dMC-MPR-SABinet. Taking the triaxial (X, Y, Z) vibration signals of rotor blades as input, the method integrates a multi-scale convolution module and a multi-path residual module, models the bidirectional temporal dependencies of signals through Bi-LSTM, and is combined with a self-attention mechanism to enhance the capture of subtle fault features. Meanwhile, it adopts the Northern Goshawk Optimization algorithm to optimize hyperparameters, thereby improving stability in noisy environments. Experiments are validated based on a self-collected fault vibration dataset, with precision, recall, and F1-score as evaluation metrics. The results show that the proposed model achieves a diagnostic accuracy of 99.37% under noise-free conditions without NGO-based hyperparameter optimization, representing a maximum improvement of 6.50% over the comparative models. Under a strong noisy condition with SNR = 1, the base model achieves 91.95% accuracy, while after NGO-based hyperparameter optimization, the model performance is further improved, with the precision, recall, and F1-score reaching 97.64%, 97.78%, and 97.01%, respectively. Ablation experiments and generalization experiments further verify the rationality and effectiveness of the proposed architecture. Full article

(This article belongs to the Section Acoustics and Vibrations)

► Show Figures

Figure 1

25 pages, 2329 KB

Open AccessArticle

A Multi-Dimensional Joint Quantitative Evaluation Method for Table Tennis Techniques Based on OpenPose and YOLO

by Yukai Yang, Hanqi Shi and Yuqiang Li

Appl. Sci. 2026, 16(10), 4661; https://doi.org/10.3390/app16104661 - 8 May 2026

Abstract

Traditional table tennis technique evaluation relies heavily on coaches’ subjective judgment, which limits the objectivity, consistency, and scalability of instructional feedback. To address this problem, this study proposes a multi-dimensional joint quantitative evaluation method for table tennis techniques based on OpenPose and YOLOv8 [...] Read more.

Traditional table tennis technique evaluation relies heavily on coaches’ subjective judgment, which limits the objectivity, consistency, and scalability of instructional feedback. To address this problem, this study proposes a multi-dimensional joint quantitative evaluation method for table tennis techniques based on OpenPose and YOLOv8 using consumer-grade high-frame-rate video. A total of 50 participants were recruited and divided into a high-level group and a low-level group. Standardized forehand drive and backhand push tasks were recorded using a synchronized dual-view camera setup. OpenPose was used to extract upper-body keypoint trajectories for kinematic analysis, while YOLOv8 was employed to detect and track the ball, racket, and net for outcome-related feature extraction. Based on these data, seven core indicators covering movement stability, coordination, timing, smoothness, and hitting effectiveness were selected to construct a quantitative scoring model, which was further optimized by ridge regression and validated against expert ratings from three senior athletes/coaches. The results show significant between-group differences in multiple technical dimensions, including impact accuracy, smoothness, trajectory consistency, and limb coordination (

p < 0.001

). The model score was strongly correlated with expert ratings (

r = 0.882

,

p < 0.001

) and demonstrated high reliability (

I C C = 0.915

). These findings indicate that the proposed framework can provide a low-cost, non-invasive, and practically effective solution for intelligent table tennis teaching, technical diagnosis, and skill-level evaluation. Full article

26 pages, 13314 KB

Open AccessReview

Synergy of Carbon Sequestration and Solid Waste Resource Utilization: A Review on Carbonation Behavior of Fly Ash Concrete

by Yubo Wang, Zhenzhao Ding, Dandan Zheng and Zhiwei Pang

Appl. Sci. 2026, 16(10), 4660; https://doi.org/10.3390/app16104660 - 8 May 2026

Abstract

In recent years, the application of fly ash concrete (FAC) has witnessed a remarkable expansion worldwide. Compared with ordinary Portland cement (OPC), the incorporation of fly ash (FA) reduces the consumption of cement, realizes solid waste resource utilization, and concurrently cuts down carbon [...] Read more.

In recent years, the application of fly ash concrete (FAC) has witnessed a remarkable expansion worldwide. Compared with ordinary Portland cement (OPC), the incorporation of fly ash (FA) reduces the consumption of cement, realizes solid waste resource utilization, and concurrently cuts down carbon emissions from cement production, thus yielding notable environmental benefits. With the gradual popularization of concrete carbon sequestration technology, the research focus of academic circles on concrete carbonation behavior has shifted from the traditional orientation of “optimizing carbonation resistance” to the new direction of “enhancing carbon sequestration efficiency”. Nevertheless, current research on the mechanical properties, durability, and other behaviors of FAC after carbonation remains scarce, lacking systematic and in-depth exploration, and the mechanism underlying the impacts of carbonation on material properties still requires further systematic collation and generalization. Consequently, research on the carbonation behavior of FAC holds profound academic significance and promising application value. This paper reviews the microscopic mechanisms and influencing factors of FAC carbonation; summarizes and analyzes the effects of FAC carbonation on its various properties and microscopic pore structure; introduces the innovative breakthroughs in FAC technology in recent years; and finally, prospects future research directions. It is anticipated to provide a valuable reference for subsequent relevant studies. Full article

(This article belongs to the Topic Innovations in Sustainable Building Materials and Structural Design for Net-Zero Carbon Emissions)

► Show Figures

Figure 1

20 pages, 2126 KB

Open AccessArticle

From Experimental Characterization to Numerical Reconstruction: Modeling Bubble Spiral Trajectory Dynamics via UDFs-Enhanced Lagrangian Methods

by Nannan Zhao, Zhiguo Luo, Lei Shao and Zongshu Zou

Appl. Sci. 2026, 16(10), 4659; https://doi.org/10.3390/app16104659 - 8 May 2026

Abstract

There are four types of bubble ascent paths: rectilinear, zigzag, spiral, and chaotic, but fewer quantitative studies on the dynamics of bubble spiral motion. In this paper, the spiral path dynamics of a single bubble with an initial diameter of 1.75–3.86 mm in [...] Read more.

There are four types of bubble ascent paths: rectilinear, zigzag, spiral, and chaotic, but fewer quantitative studies on the dynamics of bubble spiral motion. In this paper, the spiral path dynamics of a single bubble with an initial diameter of 1.75–3.86 mm in still water is investigated. The bubble position on the spiral trajectory was quantitatively characterized as a function of time using the three-dimensional shadow imaging technique combined with image digitization processing. The additional forces that induce spiral motion were derived using Newton’s second law and subsequently integrated into the Lagrangian framework through Fluent User-Defined Functions (UDFs) to reproduce the spiral trajectory of the single bubble. The simulation results for bubble velocity and trajectory closely matched the experimental data. The forces, accelerations, velocities, trajectories, and swept volumes of the bubbles are discussed. Compared to the rectilinear motion, the swept volumes of the bubbles obtained after considering the spiral paths were increased by 29.5%, 34.4%, 38.2%, 40.6%, and 37.1% for 1.75, 1.83, 1.93, 2.05, and 3.86 mm, respectively. These results will contribute to a better understanding of the dynamic behavior of the bubble spiral motion. Full article

(This article belongs to the Section Fluid Science and Technology)

14 pages, 3398 KB

Open AccessArticle

Electrical Performance of Hafnium Doped In₂O₃ Thin-Film Transistors Prepared Using a Solution Method

by Haotian Yang and Kamale Tuokedaerhan

Appl. Sci. 2026, 16(10), 4658; https://doi.org/10.3390/app16104658 - 8 May 2026

Abstract

Indium hafnium oxide thin-film transistors (TFTs) were prepared by the sol-gel method, and their crystal structures, surface morphologies, chemical compositions, optical and electrical properties were systematically investigated using X-ray diffraction (XRD), atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), ultraviolet-visible (UV-Vis) spectroscopy, and [...] Read more.

Indium hafnium oxide thin-film transistors (TFTs) were prepared by the sol-gel method, and their crystal structures, surface morphologies, chemical compositions, optical and electrical properties were systematically investigated using X-ray diffraction (XRD), atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), ultraviolet-visible (UV-Vis) spectroscopy, and a semiconductor parameter analyser. We mainly study the effects of hafnium doping on indium oxide-based thin-film transistors through the following electrical properties, including field-effect mobility (μ FE), carrier concentration, on/off current ratio (Ion/Ioff), threshold voltage (Vth), and subthreshold slope (SS). The oxygen defects concentration decreased from 25.83% to 17.82% when Hf doping was increased to 5 mol%. The effect of Hf doping on the structure, as well as the properties of the Hf-InO_x thin films, was explored and it was found that Hf as a carrier inhibitor can effectively suppress the carrier concentration. This reduces the oxygen vacancy defects and improves the electrical performance of In₂O₃TFTs devices. The doped thin-film transistor exhibits excellent electrical properties with a mobility (μ) of 11.69 cm²/Vs, a threshold voltage (V_TH) of 1.68 V, a subthreshold slope (SS) of 0.68 V/dec, and an on/off current ratio (I_on/I_off) of 10⁷ when the Hf doping level is 3 mol%. Research indicates that the Hf-InO_x thin film prepared by the sol-gel method is a low-cost, high-performance, and widely applicable active layer material. Full article

► Show Figures

Figure 1

21 pages, 2594 KB

Open AccessArticle

Comparative Evaluation of Fluorescence, NIR, and FT-IR Spectroscopy Combined with Machine Learning for Geographical Origin and Species Identification of Brown Algae

by Kana Suzuki, Meryem Taskaya, Sora Hoshino, Rikuto Akiyama, Rio Chikura, Mai Kanetsuna, Yvan Llave and Takashi Matsumoto

Appl. Sci. 2026, 16(10), 4657; https://doi.org/10.3390/app16104657 - 8 May 2026

Abstract

Ensuring the authenticity of origin labeling is a challenge for brown algae products such as kelp, wakame, and hijiki. While conventional DNA analysis has high discriminatory capabilities, it is not necessarily suitable for routine screening of large quantities of samples due to time [...] Read more.

Ensuring the authenticity of origin labeling is a challenge for brown algae products such as kelp, wakame, and hijiki. While conventional DNA analysis has high discriminatory capabilities, it is not necessarily suitable for routine screening of large quantities of samples due to time and cost considerations. This study aimed to identify the species (variety) and geographical origin of brown algae (kelp, wakame, and hijiki), with both variety and origin evaluated for kelp and geographical origin evaluated for wakame and hijiki. To achieve this, classification methods were developed for each of three spectroscopic analysis techniques—excitation emission matrix (EEM), Fourier transform infrared spectroscopy (FT-IR), and near-infrared spectroscopy (NIR)—using machine learning algorithms, and their classification performance was systematically compared and evaluated. Six classification models, k-nearest neighbors (KNN), convolutional neural networks (CNN), LightGBM, XGBoost, random forest, and support vector machines, were constructed to distinguish varieties and origins based on EEM, NIR, and FT-IR data. Depending on the combination of methods, high-precision identification was obtained (>99%), especially for kelp variety identification using NIR + KNN (≈100%). These results suggest that each spectral dataset contains characteristic information specific to each sample and that selecting a model suited to these characteristics is effective for highly accurate identification of variety (in kelp) and geographical origin. The selected method can serve as a rapid and simple identification tool that contributes to verifying the authenticity of brown algae products and improving raw material traceability. Full article

(This article belongs to the Special Issue Advanced Spectroscopy Technologies)

25 pages, 7729 KB

Open AccessArticle

Residual Decomposition for Lithotype-Aware Characterization of Rock Mechanical Parameters from Well Logs Under Lithological Heterogeneity

by Xugang Liu, Binghua Dang, Lei Li, Weixian Zhang and Wenze Zhou

Appl. Sci. 2026, 16(10), 4656; https://doi.org/10.3390/app16104656 - 8 May 2026

Abstract

Accurate characterization of rock mechanical parameters in heterogeneous geological formations remains challenging because lithological variations alter the relationship between logging signals and geomechanical responses. Existing approaches, including empirical formulas, pure machine learning models, and feature-augmented learning methods, often compress these variations into a [...] Read more.

Accurate characterization of rock mechanical parameters in heterogeneous geological formations remains challenging because lithological variations alter the relationship between logging signals and geomechanical responses. Existing approaches, including empirical formulas, pure machine learning models, and feature-augmented learning methods, often compress these variations into a single predictor, which can lead to biased estimates. To address this issue, this study proposes a heterogeneity-aware residual learning framework for rock mechanical parameter characterization from well logs. The method separates the prediction into a global component and a lithotype-conditioned correction, allowing lithological effects to be represented as structured residual behavior. This framework was developed and validated on deep coal-bearing formations in the Ordos Basin. By accounting for lithology-controlled response shifts, it produces predictions that better follow observed geological controls. Cross-well validation demonstrates reduced lithotype-induced bias and stable generalization within the studied formation. Further analysis shows that the performance gain is linked to the residual decomposition structure rather than to the addition of lithotype information alone. Compared with single-stage feature augmentation, the main advantage of the proposed framework is its ability to reduce systematic bias in lithological transition zones while preserving a transparent global–residual structure. Its demonstrated applicability is limited to wells within the studied coal-bearing formation, and broader transferability requires further validation. Full article

(This article belongs to the Special Issue Advanced Technologies in Intelligent and Sustainable Coal Mining)

24 pages, 2431 KB

Open AccessArticle

Data-Driven Multi-Objective Crashworthiness Optimization of NEV Sill Beam Under Side-Pole Impact

by Kewei Li, Peijie Xiao, Hao Li, Jianbing Wu, Jianyu Li, Zhuoran Zeng and Shiwei Xu

Appl. Sci. 2026, 16(10), 4655; https://doi.org/10.3390/app16104655 - 8 May 2026

Abstract

Protecting the power battery pack during side-pole impact poses a critical challenge for new energy vehicle (NEV) passive safety, where traditional methods struggle to balance computational efficiency and predictive accuracy due to strong nonlinearity. This study proposes a multi-stage optimization framework integrating the [...] Read more.

Protecting the power battery pack during side-pole impact poses a critical challenge for new energy vehicle (NEV) passive safety, where traditional methods struggle to balance computational efficiency and predictive accuracy due to strong nonlinearity. This study proposes a multi-stage optimization framework integrating the equivalent static load (ESL) method for nonlinear topology optimization and machine learning-based surrogate modeling. A multi-cell sill beam configuration is first developed via ESL-based topology optimization with deformation mode correction to address localized buckling. Subsequently, independent XGBoost models for intrusion, energy absorption, and mass are constructed, with their hyperparameters efficiently tuned using the Optuna framework with the Tree-structured Parzen Estimator (TPE). Results demonstrate that the optimized XGBoost model achieves superior prediction accuracy (R² > 0.96) compared to conventional models (SVR, RSM, GPR, and RBF), with RMSE approximately 40% lower than the second-best model (GPR). The Non-dominated Sorting Genetic Algorithm III (NSGA-III) is then employed for tri-objective optimization, and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) selects an optimal design, which is validated by finite element simulation with high agreement (maximum prediction error < 3%). Compared to the baseline design, the optimized sill beam achieves an 11.44% reduction in maximum intrusion, a 9.86% increase in energy absorption, and a 13.45% reduction in structural mass, demonstrating the effectiveness of the proposed data-driven crashworthiness optimization framework. Full article

(This article belongs to the Section Mechanical Engineering)

18 pages, 3210 KB

Open AccessArticle

Mechanical Characteristics of Gravel-Block Soil Considering Particle Fragmentation Fractals

by Jiamin Quan, Tao Wen, Yunpeng Yang and Bocheng Zhang

Appl. Sci. 2026, 16(10), 4654; https://doi.org/10.3390/app16104654 - 8 May 2026

Abstract

To investigate the mechanical characteristics of gravel-block soils in the cold regions, four large direct shear tests were designed under different coarse particle contents and three dry density conditions. The stress variations during shearing and the particle fragmentation rate after shearing were measured. [...] Read more.

To investigate the mechanical characteristics of gravel-block soils in the cold regions, four large direct shear tests were designed under different coarse particle contents and three dry density conditions. The stress variations during shearing and the particle fragmentation rate after shearing were measured. The experimental results indicate that when p₅ (the proportion of particles larger than 5 mm) ≥ 40%, the samples exhibit strain hardening behavior, and the stress–strain curve does not exhibit a peak within the range of the tests. The rock fragment skeleton exhibits excellent deformation resistance. With increasing coarse particle content, the internal friction angle of the soil initially decreases and then increases, while the cohesion initially decreases and then increases. Moreover, with increasing initial dry density, both the cohesion and internal friction angle of the gravel-block soils gradually increase. The fractal dimension increases with the increase in the particle fragmentation rate, indicating that the fractal dimension can also represent the degree of particle fragmentation in the soil. The relative fractal dimension increases exponentially with the increase in coarse particle content, indicating that the coarse particle content has a significant impact on the degree of particle fragmentation of gravel-block soils. The higher the coarse particle content, the greater the degree of particle fragmentation of gravel-block soils. When the coarse particle content increases from 0% to 60%, the fractal dimension decreases from 2.825 to 2.555, and the shear strength of the gravel-block soils continuously improves. During the shear process, the gravel-block soils transition from poor grading to well grading, with coarse particles breaking and fine particles filling the gaps between the coarse particles, resulting in a reduction in soil porosity and an increase in particle fragmentation rate and fractal dimension. The research outcomes of this experimental study provide guidance for the study of debris-covered slope landslides in cold regions. Full article

(This article belongs to the Special Issue Advances in Geotechnical Engineering: Stability, Mechanics, and Modeling)

27 pages, 14047 KB

Open AccessArticle

Regenerative Design for Heat-Resilient Cities: Nature-Based Microclimatic Strategies in a Mediterranean Context

by Eduardo Diz-Mellado, Juan Soto-Orozco, Victoria Patricia López-Cabeza, Francisco J. Sánchez de la Flor and Carlos Rivera-Gómez

Appl. Sci. 2026, 16(10), 4653; https://doi.org/10.3390/app16104653 - 8 May 2026

Abstract

Urban areas in Mediterranean climates are increasingly affected by extreme heat, exacerbated by the Urban Heat Island (UHI) effect and the lack of climate-responsive public spaces. This study addresses the need for integrated methodologies combining empirical monitoring and simulation tools to support regenerative [...] Read more.

Urban areas in Mediterranean climates are increasingly affected by extreme heat, exacerbated by the Urban Heat Island (UHI) effect and the lack of climate-responsive public spaces. This study addresses the need for integrated methodologies combining empirical monitoring and simulation tools to support regenerative urban design. The objective is to evaluate the effectiveness of Nature-Based Solutions (NBSs) in improving microclimatic conditions and outdoor thermal comfort during summer heatwave periods in a vulnerable urban area in Seville (Spain). A mixed-method approach combining microclimatic monitoring and ENVI-met simulations in situ was applied. A field campaign conducted in summer 2023 was used to characterize baseline conditions and calibrate the model, which simulated both current and proposed scenarios incorporating vegetation, shading systems, permeable materials, and water features. Results from the Seville case study show significant improvements, with air temperature reductions of up to 1.6 °C (daytime) and 1.9 °C (nighttime), surface temperature decreases of up to 11 °C, and thermal comfort improvements reaching 8 °C in UTCI. Beyond environmental benefits, the intervention promotes socially regenerative public space by enhancing usability, inclusivity, and comfort. Limitations include the use of a single representative summer day and inherent simplifications of the ENVI-met model. These findings demonstrate the potential of integrated NBS strategies to mitigate urban heat and support climate-adaptive and socially responsive urban design. Full article

(This article belongs to the Special Issue Urban Thermal Comfort and Microclimate: From Research to Sustainable Design)

18 pages, 50457 KB

Open AccessArticle

The Influence of SLM Process Parameters on Young’s Modulus, Poisson’s Ratio, and Impact Toughness of Ti6Al4V

by Yan Zeng, Hantao Chen, Li Yu, Shutong Dai, Jianggui Han, Zhe Wu, Man Xu and Xiaokai Wang

Appl. Sci. 2026, 16(10), 4652; https://doi.org/10.3390/app16104652 - 8 May 2026

Abstract

Selective laser melting (SLM) fabricated Ti6Al4V (TC4) has broad application potential, though its mechanical behavior is highly sensitive to processing conditions. However, the coupled response of Young’s modulus, Poisson’s ratio, and impact toughness to SLM process parameters remains insufficiently understood. In this study, [...] Read more.

Selective laser melting (SLM) fabricated Ti6Al4V (TC4) has broad application potential, though its mechanical behavior is highly sensitive to processing conditions. However, the coupled response of Young’s modulus, Poisson’s ratio, and impact toughness to SLM process parameters remains insufficiently understood. In this study, laser ultrasonic testing (LUT), Charpy impact testing, fractographic observation, and defect statistical analysis were combined to systematically investigate the effects of laser power, scan speed, and hatch spacing on the mechanical behavior of SLM-fabricated TC4 samples. The results show that the three properties exhibit a similar non-monotonic dependence on process parameters, with an overall parabola-formed trend. The optimized process parameters for Young’s modulus, Poisson’s ratio, and impact toughness were identified as laser powers of 315, 348, and 336 W, scan speeds of 1230, 1277, and 1193 mm/s, and hatch spacings of 77, 82, and 87 μm respectively. The data fitting indicated that impact toughness is the most affected by changes in both laser power, scan speed, and hatch spacing. It was followed by Young’s modulus. And Poisson’s ratio received the least impact. The combined defect, fracture, and microstructural analysis further reveal that both excessively low and excessively high volumetric energy density increase porosity and defect size, as well as reduce the β phase volume fraction and area weighted mean value of the transformed-phase plates’ width, thereby degrading mechanical performance. These results establish a process–defect/microstructure–property relationship for SLM TC4 and provide useful guidance for the optimization of mechanical performance in additively manufactured titanium alloy components. Full article

► Show Figures

Figure 1

34 pages, 4980 KB

Open AccessReview

Carbonate-Induced Self-Sealing of Near-Field Granite Fractures in Geological Disposal of High-Level Radioactive Waste: Coupled THMC Precipitation–Dissolution Mechanisms and Long-Term Performance Evaluation

by Xiao Tian, Jia-Wei Wang, Ju Wang, Zhichao Zhou, Jiebiao Li, Xianzhe Duan, Nan Li, Wentao Xu and Biao Wang

Appl. Sci. 2026, 16(10), 4651; https://doi.org/10.3390/app16104651 - 8 May 2026

Abstract

Deep geological disposal is widely recognized as the most reliable strategy for the long-term isolation of high-level radioactive waste (HLW). In granitic host rocks, fractures in the near-field represent the primary pathways for groundwater flow and potential radionuclide migration. The self-sealing capacity of [...] Read more.

Deep geological disposal is widely recognized as the most reliable strategy for the long-term isolation of high-level radioactive waste (HLW). In granitic host rocks, fractures in the near-field represent the primary pathways for groundwater flow and potential radionuclide migration. The self-sealing capacity of carbonate-filled fractures, along with its long-term effectiveness, plays a critical role in maintaining the integrity of the multi-barrier system and ensuring repository safety. Near-field fractures undergo complex thermo–hydro–mechanical–chemical (THMC) coupled evolution driven by excavation-induced disturbances, decay heat, groundwater saturation, and ongoing water–rock interactions. Within the confined fracture spaces, carbonate minerals may persistently undergo precipitation–dissolution cycling and micro- to nanoscale structural reorganization, resulting in progressive reductions in fracture connectivity and hydraulic transmissivity. However, existing studies have largely focused on short-term sealing effects, with limited systematic understanding of the long-term safety functions. In this context, this study comprehensively investigates carbonate-induced self-sealing in granitic fractures within the near-field of a repository under realistic THMC-coupled conditions. We elucidate the micro- and nanoscale heterogeneous precipitation characteristics governed by non-classical nucleation pathways, reveal how dynamic precipitation–dissolution equilibria facilitate ongoing reductions in fracture transmissivity, and propose a multi-dimensional framework for long-term hydraulic, mechanical, and chemical performance assessment. Our findings demonstrate that carbonate self-sealing operates as a dynamic, reorganizing, and multi-mineral cooperative mechanism rather than a static, one-directional process. Its core safety function lies in the sustained suppression of fracture transmissivity. The mechanistic insights and evaluation framework proposed in this study provide a foundation for integrating natural carbonate self-sealing with engineered barrier system design, thereby improving fracture control, advancing long-term safety assessment, and optimizing the design of HLW deep geological repositories. Full article

(This article belongs to the Special Issue Radioactive Waste Treatment and Environment Recovery)

22 pages, 11687 KB

Open AccessArticle

Laser-Assisted Surface Modification of Additively Manufactured WC-10Co Tools

by Gonçalo Oliveira, Patrícia Freitas Rodrigues and Maria Teresa Vieira

Appl. Sci. 2026, 16(10), 4650; https://doi.org/10.3390/app16104650 - 8 May 2026

Abstract

Tungsten carbide and cobalt cutting tools require low surface roughness to improve cutting performance by reducing the wear from machining friction. While this is achieved by conventional manufacturing processes (pressing and sintering, grinding), with additive manufacturing processes it is more difficult (layer height, [...] Read more.

Tungsten carbide and cobalt cutting tools require low surface roughness to improve cutting performance by reducing the wear from machining friction. While this is achieved by conventional manufacturing processes (pressing and sintering, grinding), with additive manufacturing processes it is more difficult (layer height, printing strategy). Since less costly and more sustainable solutions (without lubricants) are being studied as alternatives to conventional processes, a complementary technology (laser ablation) is suggested for the additive manufacturing of green WC-10Co. In this study, material extrusion (MEX) was used to produce green WC-10Co 3D objects, followed by laser ablation (50 W ytterbium fiber laser, 800–1100 nm wavelength) on their surface. Different laser strategies and parameters (power, speed, frequency, distance between lines, number of passages) were tested to find the most suitable. Most combinations were excluded by initial visual inspection, while the best ones were measured with a contact and non-contact profilometer. Further analysis was made on the composition and microstructure (with techniques such as Raman spectroscopy, scanning electron microscope, x-ray diffraction, and hardness indentation) to study what the interaction with the laser changed on the surface. Results show that with a combination of 50 W laser power, 1000 mm/s laser speed, 2000 kHz laser frequency, 0.1 mm distance between lines and three laser passages, it was possible to achieve a surface roughness of 0.6 µm (Sa) for the sintered WC-10Co, produced by MEX. No η-phase and graphite were detected, as well as microporosity and fissures. Full article

► Show Figures

Figure 1

34 pages, 1760 KB

Open AccessReview

Applications of the Kalman Filter in Physical Processes: A Review

by Ioanna Anagnostaki, Vassilis Anastassopoulos and Georgia Koukiou

Appl. Sci. 2026, 16(10), 4649; https://doi.org/10.3390/app16104649 - 8 May 2026

Abstract

This article outlines various specialized adaptations of the Kalman Filter designed to address specific estimation challenges across different domains of Physics. The work demonstrates the significant potential of the Kalman filter to enhance the accuracy and reliability of measurements in Physics, providing robust, [...] Read more.

This article outlines various specialized adaptations of the Kalman Filter designed to address specific estimation challenges across different domains of Physics. The work demonstrates the significant potential of the Kalman filter to enhance the accuracy and reliability of measurements in Physics, providing robust, real-time, and adaptive estimation capabilities. The paper starts with an extensive introduction to the core of the Kalman filter. A clear description of the different filter categories follows, along with the conditions under which each is applied. Various Kalman filter variants address nonlinear, adaptive, continuous-time, large-scale, and uncertain systems. These include the Extended and Unscented Kalman Filters for nonlinear estimation, Adaptive and Kalman–Bucy filters for changing or continuous dynamics, and Ensemble or Schmidt–Kalman filters for large or reduced-order systems. Robust, Cubature, Probabilistic, and Particle Kalman filters further improve performance under outliers, strong nonlinearities, and non-Gaussian uncertainty. To illustrate the practical relevance, detailed applications in Physics are discussed, including thermodynamics, electromagnetism, high-energy physics, quantum physics, and astrophysics, highlighting how Kalman filtering enhances both predictive accuracy and measurement-informed decision-making. Full article

27 pages, 2664 KB

Open AccessArticle

Explainable Federated Attention-Based Deep Learning for Alzheimer’s Disease Detection

by Noor Abubakr, Mustafa Kara and Hasan Hüseyin Balık

Appl. Sci. 2026, 16(10), 4648; https://doi.org/10.3390/app16104648 - 8 May 2026

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder in which early and reliable detection is critical for clinical management. This study proposes a privacy-preserving and explainable deep learning framework for Magnetic Resonance Imaging (MRI) -based AD classification that targets predictive performance, data confidentiality, [...] Read more.

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder in which early and reliable detection is critical for clinical management. This study proposes a privacy-preserving and explainable deep learning framework for Magnetic Resonance Imaging (MRI) -based AD classification that targets predictive performance, data confidentiality, and interpretability. The framework is based on an EfficientNetB0 backbone enhanced with a Convolutional Block Attention Module (CBAM). To address institutional data-sharing constraints, the model was trained in a federated learning setting, where local model updates were aggregated using Federated Averaging without sharing raw MRI data. Model interpretability was supported using Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations. The classification task was formulated as a binary distinction between cognitively normal subjects and subjects with Alzheimer’s-related impairment. Experimental evaluation on a publicly available MRI dataset showed federated performance close to centralized training, with 94.66% final test accuracy compared with 95.70% under centralized training. On an external MRI validation dataset, the model achieved 95.00% accuracy, supporting generalization to unseen data distributions. Grad-CAM heatmaps indicated that the model focused on brain regions rather than background artifacts. These results suggest that federated and explainable attention-based deep learning can support collaborative, privacy-aware AD screening while maintaining competitive predictive performance. Full article

16 pages, 1362 KB

Open AccessArticle

An Improved Transformer Early Fault Identification Method Integrating CBAM-SV2 and GAF

by Yu Yang, Liqun Liu and Xiaoyin Nie

Appl. Sci. 2026, 16(10), 4647; https://doi.org/10.3390/app16104647 - 8 May 2026

Abstract

Transformers are core equipment in power systems, and their failure may cause severe accidents. Dissolved gas analysis (DGA) is one of the dominant techniques for fault diagnosis of oil-immersed transformers. To achieve lightweight design and high identification efficiency, this paper proposes an early [...] Read more.

Transformers are core equipment in power systems, and their failure may cause severe accidents. Dissolved gas analysis (DGA) is one of the dominant techniques for fault diagnosis of oil-immersed transformers. To achieve lightweight design and high identification efficiency, this paper proposes an early fault identification method for transformers based on the integration of the Convolutional Block Attention Module-enhanced ShuffleNetV2 (CBAM-SV2) model and Gramian Angular Field (GAF). First, hybrid oversampling is used for data preprocessing. Then, the preprocessed one-dimensional gas data are converted into dual-channel two-dimensional images via GAF as the input of the classification network. Finally, a CBAM-SV2 model integrating deep convolutional networks and attention mechanisms is constructed, which combines the lightweight advantage of ShuffleNetV2 and the powerful feature representation ability of the Convolutional Block Attention Module (CBAM). Feature extraction and classification are performed by the CBAM-SV2 model to output the identification results. Additionally, t-distributed Stochastic Neighbor Embedding (t-SNE) and a confusion matrix are used to visualize classification performance for intuitive evaluation of the network’s effectiveness. The experimental results show that, compared with other mainstream algorithms, the proposed method achieves higher recognition accuracy in transformer early fault classification under imbalanced data conditions. Full article

(This article belongs to the Special Issue Advanced Technologies in Monitoring and Fault Diagnosis for Power Equipment)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Appl. Sci., Volume 16, Issue 10 (May-2 2026) – 97 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI