Green cloud computing is pivotal for reducing energy consumption and minimizing environmental impact within the rapidly expanding IT industry. Despite its importance, efficiently allocating significant resources and scheduling tasks in green cloud computing systems face significant challenges, including unpredictable task arrivals, heterogeneity of resources, network congestion, and balancing energy efficiency. Considering these complexities is crucial for achieving sustainable cloud operations. So, this paper proposes a cutting-edge hybrid optimization framework combining the Elephant Herd Algorithm and Red Piranha Optimizer named Hybridized Elephant Herd with Red Piranha Optimizer (HEH-RPO) to improve task allocation in green cloud computing systems. The primary contribution is the development of a robust and scalable algorithm capable of effectively managing large-scale, heterogeneous workloads while mitigating the duration of task execution and completion. Additionally, the scheme innovatively incorporates agile response mechanisms to address task queue overflow probabilities and network congestion proactively. Wide validations involving several conventional scenarios prove that HEH-RPO significantly outperforms conventional frameworks. The energy consumption of the proposed HEH-RPO is 0.034 kWh, whereas existing algorithms such as PSO, GA, DE, and HHO attained 0.024 kWh, 0.026 kWh, 0.028 kWh, and 0.032 kWh, respectively. The outcomes estimate the effectiveness, stability, and robustness of the designed approach, illustrating its capability for fostering sustainable and efficient cloud computing operations.
Ultra-Low Wear Polyethylene (ULWPE) is a promising alternative to Ultra-High Molecular Weight Polyethylene (UHMWPE) for joint prostheses due to its wear resistance, but wear particles are inevitable and their biological effects unclear. In addition, it is difficult to obtain sufficient clinically relevant particles. In this study, we explored to use cryosectioning to generate micron-sized particles with morphology, size, and distribution comparable to those of clinical retrievals. ULWPE particles significantly inhibited the proliferation of RAW264.7 and MC3T3-E1 cells, promoted pro-inflammatory polarization of RAW264.7, and suppressed osteogenic differentiation and mineralization of MC3T3-E1, respectively. However, Quantitative Reverse Transcription (qRT-PCR) analysis showed that ULWPE induced significantly weaker upregulation of TNF‑α and IL‑6 than UHMWPE (p < 0.01). The suppression of ALP activity was also milder in the ULWPE group (p < 0.01). In air pouch model mice, ULWPE also elicited local inflammatory responses similar to those of UHMWPE. Thus, these findings would be benefit for the further clinical translation of ULWPE.
The daily activities of bionic prosthetic hand users usually involve variable dynamic arm movements. Surface Electromyography (sEMG) signals are affected by changes in arm position, which can reduce the accuracy and robustness of gesture recognition. This paper proposes a Dynamic Convolutional Neural Network-Gated Recurrent Unit (DCNN-GRU) model with parallel global-channel weighting and location-resolved channel-spatial gating. The two branches learn complementary weighting patterns from the same window-level feature map, while the GRU models the ordered sequence of window-level representations. Experiments used sEMG data from eight able-bodied participants and two amputees. With training on all six conditions (ALL), DCNN-GRU achieved an aggregate test-window accuracy of 96.76
Biomimetic micro/nano composite structures provide an effective strategy for developing high-performance fog collection surfaces. Here, inspired by Alchemilla diplophylla, crown-shaped microstructures were fabricated via picosecond laser processing, followed by the atomic layer deposition of a nanoscale ZnO coating. The resulting micro/nano composite surface exhibited excellent hydrophobicity, with a water contact angle of 146.2 ± 1.1°, and showed minimal wettability variation across different structural geometries. Further inspired by the desert beetle strategy, patterned surfaces with alternating hydrophilic and hydrophobic regions were constructed via mask-assisted UV irradiation. During the fog collection, hydrophilic spots promoted droplet nucleation and coalescence within hexagonal pits, while the hydrophobic background reduced adhesion and facilitated rapid droplet detachment, significantly enhancing the fog collection performance. Moreover, by optimizing the matching between the micro/nanostructure size and the hydrophilic spot diameter, efficient capillary-driven transport was enabled, achieving a maximum fog collection efficiency of 1188.03 ± 60.1 mg/cm²·30 min, 3.95 times that of the bare stainless steel. In addition, the surface exhibited reversible wettability, and the contact angle remained as high as 138° after six cycles, demonstrating excellent stability and robustness, with strong potential for practical fog collection applications.
Cartilage defect repair is challenging due to the distinct composition and biological functions of cartilage and subchondral bone. Current tissue-engineered solutions for large defects often fall short in both regenerative capacity and mechanical stability, necessitating advanced strategies that integrate structural strength with osteogenic and chondrogenic potential. Biomimetic Biphasic Scaffolds (BBS) address this need by designing separate cartilage and bone layers, which are then integrated into osteochondral grafts. Innovations in 3D printing have further enhanced their regenerative potential by enabling precise incorporation of growth factors and cells, advancing osteochondral repair. This review highlights recent progress in biphasic scaffold design, fabrication, and validation in both in vitro and in vivo models. It also examines their benefits, challenges, and future prospects in osteochondral tissue engineering.
