Nowadays, blood is often considered as a base fluid to examine the chemical interactions among its components, and many modern medications are designed in the form of tri-hybrid nanofluids. In this study, the objective is to develop the mathematical framework of tri-hybrid nanofluid dynamics, which can serve as a model for drug delivery through a permeable channel, using pure blood as the base fluid combined with MgO, ZnO, and Au nanoparticles. Since biological fluid transport through arteries is an important subject in current research, the effects of heat generation due to absorption and strong magnetic fields are also incorporated. The governing equations are transformed into a system of nonlinear ordinary differential equations using similarity transformations, and the built-in bvp4c function in MATLAB is employed for numerical solutions. The analysis reveals that the tri-hybrid nanofluid composed of MgO, ZnO, and Au enhances both drug delivery in contracting and expanding channels and thermal performance. Moreover, the heat source-sink parameter increases heat generation in the fluid, thereby elevating the temperature profile as its values rise.
This work proposes a comprehensive, multi-fidelity modeling framework integrating computational fluid dynamics, machine learning, and statistical regression techniques to analyze the effective thermal conductivity of nano-encapsulated phase change materials subjected to natural convection within a cylindrical annular domain. Two key parameter spaces are explored: thermal (0.1 < theta(f) < 0.9 and 0.005 < delta < 0.5) and energy strength (10(4) < Ra-Lc < 10(6) and 10 < chi < 200), using an optimal space-filling design of experiments. Predictive models, including genetic aggregation, kriging, neural networks, and non-parametric regression, are evaluated for accuracy across these domains. Results show that genetic aggregation and kriging offer robust global predictions (R-2 = 1), while neural networks excel in local accuracy, with approximation errors below 2% at verification points. The predicted effective conductivity ratio k(eff)/k varies between about 3.4 -9.8 across the studied ranges, increasing with Rayleigh number (e.g., from approximate to 3.5 at Ra-Lc = 10(4) up to approximate to 9.8 at Ra-Lc = 10(6) for chi = 50) and showing sensitivity to melting strength (with a local minimum near chi approximate to 120 - 130). In the thermal space, optimal performance occurs when the melt front (theta(f)) aligns with the central convective zone, yielding peak ratios of approximate to 5.37, while variations in mushy-region thickness (delta) alter the magnitude and robustness of this enhancement. Additionally, the system transitions from convection-dominated to phase-change-buffered regimes as Ra-Lc and chi increase, highlighting complex interactions between latent heat and flow structures. This framework provides a generalizable approach for analyzing and optimizing NEPCM-based thermal systems under natural convection.
This study introduces and systematically investigates a novel Couette-Poiseuille backward-facing step configuration, in which the conventional stationary upper wall is replaced by a uniformly moving belt. The flow domain is filled with a trihybrid nanofluid composed of (Al2O3 - Cu - MWCNT) nanoparticles, enabling the interplay of pressure-driven and shear-driven forces within a sudden-expansion geometry with ER = 2. Numerical analysis explores the effects of varying Reynolds numbers (100 <= ReH <= 500), Rayleigh numbers (6.99 x 105 <= RaH <= 1.75 x 107), Grashof numbers (1.0 x 104 <= GrH <= 2.5 x 105), and top wall terminal velocity ratios ( - 3 <= Ubelt,r <= + 3) on flow behavior, heat transfer, and entropy generation, with the constant Prandtl number (Pr) of 69.93. At low ReH, aiding wall motion stabilizes the flow and minimizes entropy generation, while opposing wall motion induces strong recirculation, vortex shedding, and Kelvin-Helmholtz instabilities at higher ReH. Thermal field analysis reveals that counter-flow enhances convective mixing and thermal dispersion, whereas co-flow confines heated fluid near the lower wall. The moving wall improves heat transfer performance by up to 14% in optimal opposing-flow conditions and reduces the Irreversibility-to-Heat Transfer Index (IHTI) by over 90% at low ReH, where values remain below 5 x 10-7, indicating high thermodynamic efficiency. However, at high ReH = 500, entropy generation increases by more than 10 times compared to ReH = 100, and IHTI rises sharply, peaking at 4.2 x 10-6, highlighting the cost of intensified shear and recirculation. Entropy generation analysis, including Bejan number (Be) and IHTI, identifies transitions from thermal to frictional irreversibility as ReH increases and wall motion intensifies. The study defines four distinct hydrothermal regimes based on the Cumulative Heat Transfer Enhancement Ratio (CHTER) and Thermo-Hydraulic Performance Factor (THPF), highlighting trade-offs between enhanced mixing and viscous dissipation. Three thermodynamic regimes, DiffusionDominated, Transitional Coupled, and Inertia-Dominated, are characterized by evolving entropy profiles and thermodynamic efficiency. The results offer new insights into optimizing mixed convection flows using walldriven mechanisms and advanced trihybrid nanofluids for superior heat transfer and energy performance in compact thermal systems.
This paper investigates a low Earth orbit (LEO) satellite communication system enhanced by an active stacked intelligent metasurface (ASIM), mounted on the backplate of the satellite's solar panels to efficiently utilize limited onboard space and reduce the main satellite power amplifier requirements. The system serves multiple ground users via rate-splitting multiple access (RSMA) and IoT devices through a symbiotic radio network. Multi-layer sequential processing in the ASIM improves effective channel gains and suppresses inter-user interference, outperforming active RIS and beyond-diagonal RIS designs. Three optimization approaches are evaluated: block coordinate descent with successive convex approximation (BCD-SCA), model-assisted multi-agent constraint soft actor-critic (MA-CSAC), and multi-constraint proximal policy optimization (MCPPO). Simulation results show that BCD-SCA converges fast and stably in convex scenarios without learning, MCPPO achieves rapid initial convergence with moderate stability, and MA-CSAC attains the highest long-term spectral and energy efficiency in large-scale networks. Energy-spectral efficiency trade-offs are analyzed for different ASIM elements, satellite antennas, and transmit power. Overall, the study demonstrates that integrating multi-layer ASIM with suitable optimization algorithms offers a scalable, energy-efficient, and high-performance solution for next-generation LEO satellite communications.
The reliability and efficiency of photovoltaic (PV) systems are critical to ensuring the stability and sustainability of modern power grids, particularly with the increasing integration of renewable energy sources. However, PV installations are exposed to diverse environmental and operational conditions that can lead to various faults, reducing power output and system lifespan. Traditional fault detection and diagnosis (FDD) methods often rely on single-source measurements and handcrafted features, which limit their adaptability and diagnostic precision in complex, real-world conditions. In this context, multimodal learning has emerged as a promising paradigm that leverages heterogeneous data sources such as electrical, thermal, visual, and environmental information to enhance fault detection accuracy and robustness. This paper surveys recent advancements in multimodal-based FDD for PV systems, emphasizing fusion strategies and attention mechanisms that enable effective cross-domain feature integration. We review key approaches at the data, feature, and decision levels, along with hybrid architectures that exploit attention to adaptively weight informative modalities. Critical research challenges, including data heterogeneity, sensor synchronization, imbalance in fault samples, and real-time implementation, are thoroughly discussed. Furthermore, emerging directions such as transformer-based architectures, self-supervised representation learning, edge-intelligent diagnosis, and privacy-preserving federated learning are explored as enablers for scalable and interpretable PV fault diagnosis. This review provides a comprehensive roadmap toward the development of intelligent, adaptive, and resilient FDD frameworks for next-generation photovoltaic systems.
