This study presents detailed modelling and comparative analysis of pouch lithium-ion cells (PLCs) configured in combined single and dual battery packs for electric vehicles (EVs), appraising their performance with and without active cooling systems. Pouch cells were selected due to their high volumetric energy density (VED), lightweight design, and suitability for high-power EV applications. Using SolidWorks for computer aided design (CAD) modelling, Analysis System (ANSYS) Fluent software with the multi-scale multi-domain (MSMD) approach, and equivalent circuit model (ECM), simulation based thermal and electrochemical behaviour during 1-hour charging process at specified currents was achieved. Four configurations were examined: single-pack and dual-pack, each with and without liquid active cooling. Main parameters included state of charge (SoC), temperature distribution, and thermal gradients. Results confirmed that dual-pack configuration significantly outperformed single-pack, achieving over 69.3% higher SoC (55.07% against 32.53%) under the same conditions. Without active cooling, maximum temperatures reached 541.11 K (single) and 462.77 K (dual), indicating notable hotspots in the single pack. Active cooling dramatically reduced temperatures to 295 K across both setups, with the dual pack displaying superior uniformity and a 0.45% lower average temperature, adequately averting thermal gradients and enhancing safety. Mathematical validation of SoC dynamics confirmed the dual configuration’s theoretical advantage in charging efficiency, tempered by heat losses. Findings reveal that dual battery packs integrated with active cooling render optimal equilibrium of faster charging, improved thermal management, and extended battery life, addressing limitations associated with conventional single-pack designs.

Solid-particle erosion at pipeline elbows threatens the integrity of oil-and-gas transport systems. This numerical study used the Euler–Lagrange discrete phase model and the Finnie erosion model in ANSYS Fluent to compare a single 90° elbow, two 45° elbows, and three 30° elbows for water–sand flow at inlet velocities of 10–40 m/s and particle diameters of 0.0002–0.0005 m. Two output measures are reported: contour plots show the local cellwise maximum wall erosion rate, whereas line graphs show the area-weighted mean wall erosion rate. At the reference condition of 40 m/s and a particle diameter of 0.0005 m, the area-weighted mean rates were 5.18 $\times$ 10$^{-5}$, 2.89 $\times$ 10$^{-5}$, and 3.59 $\times$ 10$^{-5}$ kg m$^{-2}$ s$^{-1}$ for the single 90° elbow, two 45° elbows, and three 30° elbows, respectively. Relative to the single elbow under the same simulation conditions, the mean erosion rate decreased by 44.2% with two 45° elbows and by 30.7% with three 30° elbows. The corresponding local contour maxima were 3.17 $\times$ 10$^{-3}$, 2.82 $\times$ 10$^{-3}$, and 2.59 $\times$ 10$^{-3}$ kg m$^{-2}$ s$^{-1}$. These results show that distributing the change in flow direction across multiple elbows reduces severe particle–wall impacts, with two 45° elbows providing the lowest area-weighted mean erosion rate.

Zero-day attacks–exploiting unknown vulnerabilities before patches exist–pose a critical threat to modern network infrastructure that signature-based intrusion detection systems cannot address. This paper proposes a hybrid computational framework combining a Convolutional Neural Network (CNN)-Gated Recurrent Unit (GRU)-Attention classifier with a skip-connection convolutional autoencoder (AE) for simultaneous known-attack classification and zero-day anomaly detection. The framework introduces three key computational contributions: (1) deterministic reshaping of 64 Random Forest-selected network flow features into 8 $\times$ 8 spatial images, enabling end-to-end CNN processing without feature engineering; (2) a strict Score-based Label Separation and Ordering (SLSO) data partition enforcing complete information isolation between training, validation, and zero-day evaluation sets; and (3) an OR-fusion hybrid decision rule combining anomaly score and reconstruction error signals. Experimental evaluation on Canadian Institute for Cybersecurity Intrusion Detection System (CICIDS)2017 demonstrates 97.48% zero-day detection rate (Z-DR) (95% confidence interval (CI) [97.1%, 97.9%]) at 4.2% false positive rate (FPR) and Area Under the Receiver Operating Characteristic curve (AUROC) of 0.956 across three held-out zero-day attack families–substantially outperforming all classical baselines (best: Stochastic Gradient Descent-optimized One-Class Support Vector Machine (SGD-OCSVM) at 85.45%). SHapley Additive exPlanations (SHAP) explainability analysis reveals mechanistic complementarity: the CNN captures temporal flow signatures while the AE contributes 1,012 exclusive detections via backward inter-arrival time anomalies. The system operates at 14,201 samples/second on Graphics Processing Unit (GPU), satisfying real-time deployment requirements. These results demonstrate that hybrid supervised-unsupervised fusion with rigorous experimental methodology substantially advances zero-day detection capability for computational network security systems.

Small-scale coffee producers require drying systems that reduce dependence on weather, improve moisture uniformity and remain technically feasible at laboratory or farm scale. This work presents the design and performance analysis of a conical fluidized-bed dryer (FBD) for parchment coffee, focusing on the reactor geometry, air-distribution strategy, heating-chamber configuration and electric-resistance arrangement. The experimental stage included blower characterization from 10 to 60 Hz, with measured outlet air velocities from 4.8 to 35.5 m/s and volumetric flow rates from 0.0098 to 0.0725 m$^3$/s. Fluidization observations were performed with parchment-coffee masses of 0.5, 3.0, 4.0 and 5.0 kg. The numerical stage, developed in SolidWorks Flow Simulation, compared cylindrical and conical reactor configurations, distributor alternatives, conical opening angles of 20°, 25° and 30°, heating-chamber geometries, baffle arrangements and electrical-resistance layouts. The cylindrical prototype showed non-uniform fluidization because the incoming jet expanded poorly and remained attached to one side of the reactor, producing localized high-velocity regions and stagnant zones. In contrast, the conical reactor promoted grain recirculation toward the lower high-velocity region and was therefore selected for the final design. The proposed dryer uses a 25° conical reactor, a 50.8 mm inlet section, a 152.4 mm heating chamber with conical inlet and outlet sections, a flat baffle and four 900 W spiral electrical resistors. The ideal heating demand was estimated as 3.40 kW for increasing the air temperature from approximately 30 ℃ to 70 ℃, while the installed 3.60 kW heater capacity provides a 5.9% margin relative to the ideal requirement. The results demonstrate the design feasibility of a 5 kg parchment-coffee fluidization capacity. However, drying kinetics, pressure drop, energy consumption, final moisture uniformity and calibrated uncertainty must still be validated experimentally in the constructed prototype.

Injury prevention in military and industrial environments requires reliable systems capable of detecting biomechanical stress under practical monitoring conditions. Conventional assessment approaches are often limited to post-event analysis and lack the ability to provide immediate feedback under dynamic loading conditions. This study aimed to develop an embedded-oriented smart-skin system for injury-risk detection using a biomechanical mannequin equipped with multisensory technology. The proposed system consisted of a flexible smart-skin layer embedded with pressure and temperature sensors and connected to an embedded data acquisition unit for continuous monitoring. Sensor data were processed through a structured pipeline comprising signal filtering, window-based segmentation, and statistical feature extraction. Several lightweight machine-learning models, including Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and eXtreme Gradient Boosting (XGBoost), were employed to classify the biomechanical conditions as either safe or injury risk. This classification supported rapid decision-making under controlled experimental conditions. Experimental evaluation was conducted under controlled static, dynamic, and combined loading scenarios. The results showed that the proposed system achieved high classification performance, with a maximum accuracy of 99.66% and consistently high F1-scores across all evaluated models. The high performance can be attributed to the use of discriminative statistical features and the controlled experimental setup, which enabled clear separation between the two classes. These findings indicate that the integration of multi-sensor smart-skin technology with efficient data processing and lightweight machine-learning models provides a feasible framework for injury-risk detection under controlled mannequin-based experimental conditions. This study contributes an integrated experimental measurement and computational analysis pipeline for embedded-oriented injury-risk monitoring, with potential applications in ergonomic assessment, military training, and occupational safety.

Face aging simulation plays a critical role in generating age-progressed facial images, which is vital in applications such as facial recognition, security surveillance, and medical imaging. Traditional image enhancement techniques often suffer from limitations such as age blurring and loss of fine-grained details. To address these challenges, this research proposes an improved Detail-Enhancement Cycle-Consistent Generative Adversarial Network (CycleGAN) framework using deep learning for efficient and accurate face aging. The primary objective is to generate high-quality, age-preserved facial images from low-resolution or blurred inputs. The proposed method uses a refined CycleGAN architecture, with the generator comprising advanced downsampling and upsampling modules to effectively capture and reconstruct facial features. The discriminator network evaluates the authenticity of the generated images, distinguishing between real and fake outputs to improve generation quality. This adversarial learning approach ensures the structural consistency and sharpness of the enhanced images. The model is trained and evaluated on a comprehensive face dataset containing a wide range of facial expressions, lighting conditions, and angles. To ensure a robust evaluation, four distinct loss functions adversarial loss, cycle-consistency loss, identity loss, and age loss are calculated and analyzed during training. The losses collectively improve the structural realism and age preservation in the generated images. Experimental results demonstrate that the improved CycleGAN model outperforms conventional Generative Adversarial Network (GAN)-based enhancement techniques in terms of age clarity, texture preservation, and overall visual quality. Quantitative metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) further confirm the superiority of the proposed method. Finally, the enhanced CycleGAN provides a reliable solution for real-world facial image applications that require precise age detail.

The rapid expansion of autonomous Large Language Model (LLM) agents introduces critical security risks, particularly the confused deputy problem caused by orchestrator compromise or indirect prompt injection (IPI). Traditional defenses rely on in-band filtering, which fails to prevent a subverted orchestrator from bypassing security checks. We propose Intent-Execution Integrity Binding (I-EIB), a reference architecture designed to enforce complete mediation by isolating execution capabilities within a Trusted Execution Environment (TEE). By leveraging Amazon Web Services (AWS) Nitro Enclaves and Key Management Service (KMS)-based capability sealing, I-EIB ensures that sensitive API keys and session tokens remain inaccessible to the host. These credentials are only released when a TEE-based verifier confirms that the processing provenance matches a user-signed domain-specific language (DSL) policy and a predefined chain layout. We formalize the system’s security invariants and provide a structured proof sketch demonstrating how I-EIB maintains intent-execution integrity even under full host compromise. Compared with a host-mediated baseline that performs policy checking without enclave isolation or attestation-gated capability release, the proposed architecture introduces a mean additional overhead of 14.2 ms. Experimental results show a 100% detection rate for bounded step-omission attacks and a false positive rate below 0.1% after normalization. A workflow-oriented case study and computational-path analysis further indicate that the architecture is feasible for high-stakes domains such as finance and healthcare, while still carrying deployment constraints associated with TEE availability, policy precision, and wrapper management.

This paper presents a new four-node strain-based finite element (SBQ12) formulated within the framework of Reissner–Mindlin plate theory for the dynamic and stability analysis of isotropic plates resting on elastic foundations. The independent approximation of the bending and transverse shear strains in the proposed element is efficient in eliminating the shear locking and enhancing the numerical accuracy and stability for thin and thick plates. In this study, the strain-based finite element is used to investigate the free vibration and linear buckling of plates on Winkler, Pasternak and Kerr elastic foundations. The SBQ12 element is extensively validated for square and rectangular plates with different boundary conditions (all edges simply supported (SSSS), all edges clamped (CCCC), two opposite edges simply supported, two opposite edges clamped (SCSC), and two opposite edges simply supported, two opposite edges free (SFSF), etc.), thickness ratios ($a/h$ ranging from 5 to 1000), and aspect ratios. The numerical results show excellent agreement with the analytical and reference solutions, with maximum relative errors generally less than 3.5% for free vibration analyses and 4% for buckling analyses in the majority of cases tested. Particular attention is given to the influence of foundation stiffness parameters on natural frequencies and critical buckling loads. The obtained results confirm the accuracy, reliability, and computational efficiency of the proposed element. Overall, the developed SBQ12 element proves to be a robust and highly accurate tool for the analysis of isotropic plate structures resting on elastic foundations, offering a valuable contribution to computational mechanics.