Saad Arif

The integration of wireless power transfer (WPT) technology with fuel cell hybrid electric vehicles (FCHEVs) offers a viable solution to enhance energy accessibility, increase flexibility, and improve driving sustainability. However, the coordinated operation of wireless charging systems, hybrid energy-storage systems (HESSs), and vehicular propulsion units introduces significant control challenges due to nonlinear dynamics, parameter uncertainties, and rapidly varying load demands. To overcome these limitations, this paper presents a WPT–FCHEV architecture comprising an LCC-S-based wireless charging system, a fuel cell source, a battery, ultracapacitor-based HESS, and an induction-motor (IM) drive. Comprehensive mathematical models of the respective charging and propulsion stages are developed by incorporating the dynamics of the LCC-S-compensated WPT, energy-storage units, bidirectional DC–DC converter interfaces, DC bus, and IM drive. A constant-coefficient-based finite-time sliding mode controller (CCFTSMC) is designed to regulate fuel cell current, battery current, ultracapacitor current, DC bus voltage, and motor speed, thereby ensuring finite-time convergence and robust tracking performance under dynamic operating conditions. The efficacy of the proposed framework is investigated through extensive simulation in MATLAB/Simulink under charging and propulsion scenarios. The obtained results demonstrate accurate current regulation and power allocation, effective DC bus voltage regulation, good speed tracking, and stable wireless charging operation. Compared with the conventional Backstepping controller, the proposed CCFTSMC achieves a 93.41% reduction in settling time and a 38.06% reduction in rise time, demonstrating superior transient and steady-state tracking performance with reduced RMSE and MAE tracking errors. Robustness analysis under ±20% parameter variations further demonstrates stable operation without controller retuning, with WPT power maintained between 309.5 and 310.6 W under mutual-inductance variation and only marginal speed RMSE variation under converter inductance and resistance uncertainties. These results confirm the robustness and effectiveness of the proposed control strategy for WPT–FCHEV applications.

Efficient monitoring and recognition of excavator operational activities are critical for improving productivity, safety, and intelligent construction management. Traditional vision-based activity recognition systems remain sensitive to challenging construction site conditions, including occlusion, dust, variable illumination, and limited visibility. This paper presents an excavator activity recognition framework that, for the first time, applies Gramian Angular Field (GAF) encoding to multi-sensor excavator operational data, transforming them into spatial image representations and enabling deep convolutional neural networks (CNNs) to extract discriminative features. The proposed approach integrates synchronized excavator sensor signals including bucket positional coordinates, body orientation, fuel consumption, engine RPM, and joint angles to characterize operational behavior during four representative activities: digging, dumping, idle, and levelling. Unlike conventional sensor-based methods that directly process sequential time-series data, our GAF-based framework transforms operational signals into structured spatial representations that preserve temporal correlations while enabling effective feature learning through image-based deep learning architectures. Experimental results demonstrate that the proposed GAF-CNN-LSTM framework achieves 8.24 percentage points higher classification accuracy compared with LSTM networks, and 6.24 percentage points higher than 1D CNN baselines trained on the same sensor data. The method effectively captures discriminative operational signatures across multiple excavator activities in the collected dataset. This work bridges the limitations of both vision-based and traditional sensor-based approaches, providing a promising framework for excavator activity monitoring in construction and fleet management contexts, pending further validation under operational deployment conditions.

Economic Dispatch (ED) is a fundamental optimization problem in power system operation, often characterized by nonlinearity, non-convexity, and practical constraints such as valve-point effects (VPE), ramp-rate limits (RRL), and transmission losses (TL). This paper presents a systematic evaluation of the Hunger Games Search (HGS) algorithm for constrained economic dispatch (ED) problems under diverse operating conditions. A comprehensive and systematically controlled evaluation framework is developed, considering benchmark systems ranging from small-scale (6 units) to large-scale (up to 160 generating units). All experiments are conducted over 30 independent runs, and performance is assessed using statistical indicators including best, mean, worst, and standard deviation of fuel cost. Constraint feasibility is enforced using the hybrid strategy, with power balance error maintained below 1e-6. To ensure fair comparison, computational effort is standardized using the number of function evaluations (NFE). In addition to conventional benchmarking, a detailed parameter tuning and sensitivity analysis is performed to examine the influence Wilcoxon signed-rank test and Friedman test, are employed to rigorously validate the significance of performance differences across algorithms. The results indicate that HGS achieves generally stable performance with variability in complex cases, across a wide range of ED formulations, particularly in large-scale and highly constrained scenarios. The combined use of sensitivity analysis and statistical validation provides deeper insight into the behavior of HGS beyond outcome-based comparison. These findings demonstrate that HGS is a competitive and effective metaheuristic, with performance dependent on problem complexity and constraint severity for solving complex ED problems, while also providing a reproducible and statistically grounded framework for performance evaluation.

Medical image reconstruction and enhancement play a critical role in improving the reliability of computer assisted disease analysis. In this study, a simulation-based hybrid framework is proposed that integrates a deep neural network (DNN) with a cubature Kalman filter (CKF) to combine nonlinear state estimation with learning-based refinement for improved image reconstruction and organ-wise disease classification. The framework is evaluated in a controlled simulation environment using synthetic multimodal radiology-pathology images representing four organs (liver, kidney, lung, and heart) and three disease severity levels (normal, mild, and severe). The proposed hybrid approach consistently demonstrates higher reconstruction fidelity and classification performance than standalone CKF and DNN models. Quantitative evaluation using the peak signal-to-noise ratio and the structural similarity index measure indicates improved structural preservation and noise reduction. In addition, classification analysis using confusion matrices and derived performance metrics demonstrates reliable discrimination between disease severity levels. Across the simulated cases, improvements of approximately 5–10% in classification accuracy and higher reconstruction quality are achieved relative to baseline methods. These findings suggest that the integration of nonlinear filtering with deep learning can provide a robust computational framework for multimodal medical image reconstruction and organ-wise disease analysis. Although the results are based on simulation experiments, the proposed approach demonstrates potential for future validation using real clinical imaging datasets.

This study introduces a physics-guided deep learning architecture designed for the simulation, reconstruction, and pattern recognition of biomedical images. By explicitly integrating physical priors into the learning model, the framework addresses the black-box nature of traditional artificial intelligence (AI). It provides an explainable AI pathway that enhances diagnostic accuracy, robustness, and clinical interpretation. The proposed framework was evaluated through systematic simulation studies. It involved complex geometric configurations, multimodal physical fields, and noise-corrupted synthetic three-dimensional brain volumes. Quantitative analysis demonstrates consistent improvements in reconstruction fidelity, with the peak signal-to-noise ratio (PSNR) reaching 47 dB and the structural similarity index exceeding 0.90 across all scenarios. Notably, at moderate noise levels (0.05), the framework maintains a PSNR greater than 32 dB, ensuring structural integrity essential for computer-aided diagnosis. Volumetric brain experiments further reveal a 38–44% reduction in activation localization errors, highlighting the framework’s utility in functional imaging and disease prognosis. By grounding deep learning in physical constraints, this study provides a transparent and robust solution for automated disease classification and advanced biomedical imaging tasks within clinical decision support systems.

Cerebral palsy (CP) is a neurological disorder caused by early brain injury. It often damages motor function, as well as generates sensory and cognitive difficulties. In this work, we proposed a virtual reality (VR)-based telerehabilitation system for CP, called Virtual Reality Telerehabilitation for Cerebral Palsy (VRTCP). The VRTCP is designed to support patients with upper-limb motor impairments. The system combines immersive VR technology with real-time surface electromyography (EMG) biofeedback sensors. The system is designed to provide interactive and task-oriented therapy sessions. We evaluated the VRTCP in two phases of studies: a) 10-participant and 100-participants. The 10-participant pilot study showed improvements across key motor-performance indicators. During the 100-participant study, we compared two intervention arms: (a) VR with EMG biofeedback and (b) EMG biofeedback only. The VR-integrated group achieved better gains, including improvements in Motor Function Score (64.8%), normalized EMG amplitude (135% increase), response time (32% reduction), and coordination index (108% improvement). Other than performance outcomes, the VRTCP provides continuous monitoring of neuromuscular control. Better engagement is observed, with a high session completion rate of around 93% in the VR with Biofeedback group. Overall, the findings suggest that VRTCP can complement conventional rehabilitation and offer a practical approach for remote therapy, particularly in areas with limited access to in-person services.

Gastric cancer remains one of the most common malignancies worldwide. The timely and accurate histopathological diagnosis plays a critical role in effective treatment. Manual interpretation of histology slides is time consuming and requires considerable expertise. To address these challenges, this study introduces a two-teacher one-student (2T–1S) knowledge distillation framework for gastric cancer classification using the GasHisSDB dataset. The framework leverages DenseNet-121 and ResNet-50 as teacher networks to guide a lightweight MobileNet-V2 student. This approach provided high accuracy with significantly reduced computational cost. Experiments on multi-resolution patches (80 × 80, 120 × 120, and 160 × 160) show that the MobileNet-V2 student achieved accuracies of 95.78%, 97.46%, and 98.33%, respectively. Also, the teacher model DenseNet-121 achieved the accuracies of 96.44%, 98.75% and 98.19% and the ResNet-50 teacher reached 96.63%, 97.87% and 98.31% respectively. In addition, the student network was more than thirty times smaller and nearly twice as fast during inference. This fast light-weight model is well-suited for real-time inference on resource-constrained devices. Integrated Gradients were applied to explain the model was paying attention to actual features and focus on meaningful regions like nuclei clusters and gland boundaries. Compared with many existing techniques this framework act as balance trade-off between accuracy, speed and interpretability. This balance positions the framework as a viable tool for digital pathology workflows and further refinement could extend its utility to clinical decision support.

Synthetic polymer composites exhibit exceptional strength, making them ideal candidate for manufacturing industrial applications. Besides their advantages these composites have lower tendency under impact loads, which severely hinders their widespread utilization. Considering the facts, this study involves fabrication of hybrid composite materials incorporating a stacked combination of kevlar/glass fabrics reinforced into epoxy/vinylester hybrid resins at varied ratios (70:30, 50:50, and 30:70). The prepared hybrid composites were then subjected to mechanical and thermal analysis to observe the influence of fiber hybridization as well as hybrid resins effect. The kevlar/glass hybrid composite with balanced resins ratio (50:50) outperformed the (70:30) sample by 8.44% in tensile strength (227.8 MPa) and 8.63% in flexural strength (134.11 MPa). This superior strength is attributed to the better bond formation between the resins, which ultimately supported the hybrid reinforcement. A better fiber/matrix interfacial adhesion was observed in balanced resin‐based hybrid composite during scanning electron microscopic analysis, including reduced fiber pull‐outs, and delamination. Furthermore, thermogravimetric analysis revealed that the proposed hybrid composite exhibited enhanced thermal stability as well as residual char. Thus, the result of this study provides valuable insights into the hybrid composite laminates, and such composites are promising for various industrial applications. These newly developed materials are auspicious for high‐performance engineering applications without conferring environmental problems.