Micromachines最新文献

The flexible micro-structured surface found in biological skins exhibits remarkable drag reduction properties, inspiring applications in the aerospace industry, underwater exploration, and pipeline transportation. To address the challenge of efficiently replicating such structures, this study presents a composite flexible polymer film with a bio-inspired micro-dimple array, fabricated via an integrated process of precision milling, polishing, and micro-injection molding using thermoplastic polyurethane (TPU). We systematically investigated the influence of key injection parameters on the shape accuracy and surface quality of the film. The experimental results show that polishing technology can significantly reduce mold core surface roughness, thereby enhancing film replication accuracy. Among the parameters, melt temperature and holding time exerted the most significant effects on shape precision PV and bottom roughness R_a, while injection speed showed the least influence. Under optimized conditions of a melt temperature of 180 °C, injection speed of 60 mm/s, holding pressure of 7 MPa, and holding time of 13 s, the film achieved a micro-structure shape accuracy of 13.502 μm and bottom roughness of 0.033 μm. Numerical simulation predicted a maximum drag reduction rate of 10.26%, attributable to vortex cushion effects within the dimples. This performance was experimentally validated in a flow velocity range of 0.6-2 m/s, with the discrepancy between simulated and measured drag reduction kept within 5%, demonstrating the efficacy of the proposed manufacturing route for flexible bio-inspired drag reduction film.

The widespread adoption of low-power devices and microelectronic systems has intensified the need for efficient energy harvesting solutions. While cantilever-beam piezoelectric energy harvesters (PEHs) are popular for their simplicity, their performance is often limited by conventional mass block designs. This study addresses this by proposing a comprehensive structural optimization framework for a triangular cantilever PEH to significantly enhance its electromechanical conversion efficiency. The methodology involved a multi-stage approach: first, an embedded coupling design was introduced to connect the mass block and cantilever beam, improving space utilization and strain distribution. Subsequently, the mass block's shape was optimized. Furthermore, a negative Poisson's ratio (NPR) honeycomb structure was integrated into the cantilever beam substrate to induce biaxial strain in the piezoelectric layer. Finally, a variable-density mass block was implemented. The synergistic combination of all optimizations-embedded coupling, NPR substrate, and variable-density mass block-culminated in a total performance enhancement of 69.07% (17.76 V) in voltage output and a 44.34% (28.01 Hz) reduction in resonant frequency. Through experimental testing, the output performance of the prototype machine showed good consistency with the simulation results, successfully verifying the effectiveness of the structural optimization method proposed in this study. These findings conclusively show that strategic morphological reconfiguration of key components is highly effective in developing high-performance, low-frequency adaptive piezoelectric energy harvesting systems.

Magnetorheological fluids (MRFs) exhibit dynamic, field-responsive mechanical properties, as they form chain-like and networked microstructures under magnetic stimuli. This study numerically investigates the structural and mechanical behavior of three-dimensional (3D) microbead chain assemblies, focusing on cubic and hexagonal body-centered tetragonal (BCT) configurations formed under compressive and magnetic field-driven aggregation. A finite element-based model simulates magnetostatic and stress evolution in solidified structures composed of up to 20 particle chains. The analysis evaluates magnetic flux distribution, total magnetic force, and time-resolved stress profiles under vertical loading. Results show that increasing chain density significantly enhances magnetic coupling and reduces peak stress, especially in hexagonal lattices, where early stress equilibration and superior lateral load distribution are observed. The hexagonal BCT structure exhibits superior resilience, lower stress concentrations, and faster dissipation under dynamic loads. These findings offer insights into designing energy-absorbing MRF-based materials for impact mitigation, adaptive damping, and protective microfluidic structures.

With the rapid advancement of Transformer-based large language models (LLMs), these models have found widespread applications in industrial domains such as code generation and non-functional requirement (NFR) classification in software engineering. However, recent research has primarily focused on optimizing linear matrix operations, while nonlinear operators remain relatively underexplored. This paper proposes hardware-efficient approximation and acceleration methods for the Softmax and RMSNorm operators to reduce resource cost and accelerate Transformer inference while maintaining model accuracy. For the Softmax operator, an additional range reduction based on the SafeSoftmax technique enables the adoption of a bipartite lookup table (LUT) approximation and acceleration. The bit-width configuration is optimized through Pareto frontier analysis to balance precision and hardware cost, and an error compensation mechanism is further applied to preserve numerical accuracy. The division is reformulated as a logarithmic subtraction implemented with a small LOD-driven lookup table, eliminating expensive dividers. For RMSNorm, LOD is further leveraged to decompose the reciprocal square root into mantissa and exponent parts, enabling parallel table lookup and a single multiplication. Based on these optimizations, an FPGA-based pipelined accelerator is implemented, achieving low operator-level latency and power consumption with significantly reduced hardware resource usage while preserving model accuracy.

With the increasing demand for high throughput and ultra-dense small cell deployment in the next-generation communication networks, spectrum resources are becoming increasingly strained. At the same time, the security risks posed by eavesdropping remain a significant concern, particularly due to the broadcast-access property of optical fronthaul networks. To address these challenges, we propose a high-security, high-spectrum efficiency radio-over-fiber (RoF) system in this paper, which leverages compressive sensing (CS)-based algorithms and chaotic encryption. An 8 Gbit/s RoF system is experimentally demonstrated, with 10 km optical fiber transmission and 20 GHz radio frequency (RF) transmission. In our experiment, spectrum efficiency is enhanced by compressing transmission data and reducing the quantization bit requirements, while security is maintained with minimal degradation in signal quality. The system could recover the signal correctly after dequantization with 6-bit fronthaul quantization, achieving a structural similarity index (SSIM) of 0.952 for the legitimate receiver (Bob) at a compression ratio of 0.75. In contrast, the SSIM for the unauthorized receiver (Eve) is only 0.073, highlighting the effectiveness of the proposed security approach.

The escalating heat flux density and temperature in highly integrated microelectronic devices adversely affect their reliability and service life, making efficient thermal management crucial for stable operation. This study utilizes Galinstan liquid metal as the coolant to investigate the flow and heat transfer performance in microchannel heat sinks incorporating various turbulator configurations. It is revealed that for microchannels featuring expanded regions, turbulators that create highly symmetric flow fields are preferable due to improved flow distribution. The long teardrop-shaped turbulator provides the best heat transfer performance among all the investigated heat transfer enhancement structures. And this turbulator yields a 13.8-25.9% higher enhancement effectiveness compared to other configurations, at the expense of a 28-41% increase in pressure loss. However, the sudden cross-sectional expansion in the expanded region causes a significant reduction in fluid velocity. Consequently, microchannels with expanded regions and turbulators exhibit a higher bottom surface temperature than the original, straight microchannels, leading to an overall deterioration in heat transfer performance.

This study presents a method for enriching oil-suspended particles within a rectangular microfluidic channel using acoustic standing waves. A modified Helmholtz equation is solved to establish the acoustic field model, and the equilibrium between acoustic radiation forces and viscous drag is described by combining Gor'kov potential theory with the Stokes drag model. Based on this force balance, the particle motion equation is derived, enabling the determination of the critical particle size necessary for efficient enrichment in oil-filled microchannels. A two-dimensional standing-wave microchannel model is subsequently developed, and the influences of acoustic, fluidic, and particle parameters on particle migration and aggregation are systematically investigated through theoretical analysis and numerical simulations. The results indicate that when the channel dimension and acoustic wavelength satisfy the half-wavelength resonance condition, a stable standing-wave field forms, effectively focusing suspended particles at the acoustic pressure nodes. Enrichment efficiency is found to be strongly dependent on inlet flow velocity, particle diameter, acoustic frequency, temperature, and particle density. Lower flow velocities and larger particle sizes result in higher enrichment efficiencies, with the most uniform and stable pressure distribution achieved when the acoustic frequency matches the resonant channel width. Increases in temperature and particle density enhance the acoustic radiation force, thereby accelerating the aggregation of particles. These findings offer theoretical foundations and practical insights for acoustically assisted online monitoring of wear particles in lubricating oils, contributing to advanced condition assessment and fault diagnosis in mechanical systems.

Focused Ion Beam (FIB) systems are increasingly utilized in nanotechnology for nanostructuring, surface modification, doping, and rapid prototyping. Recently, their potential for quantum applications has been explored, leveraging FIB's direct-write capabilities for in situ single ion implantation, which is crucial for fabricating single photon emitters. Color centers in diamond can function as qubits and are of particular interest due to their capacity to store and transmit quantum information. While Group-IV color centers exhibit high brightness, they require low temperatures to retain coherence. However, lead-vacancy in diamond (PbV) operates at the higher end (4 K) of this temperature spectrum due to larger ground-state splitting, making them particularly interesting. In this context, our study presents results for lead (Pb)-containing alloys with eutectic points below 600 °C and results on using tantalum (Ta) and titanium (Ti) as emitter materials for a Pb liquid metal alloy ion source. We show that a standard FIB system is able to resolve the different Pb isotopes and achieve nanoscale spot sizes, as required for quantum information science applications.

Flexible sensors have emerged as critical interfaces for information exchange between soft biological tissues and machines. Here, we present a dual-mode stretchable sensor system capable of synchronous strain and electromyography (EMG) signal detection, integrated with wireless WIFI transmission for real-time joint movement monitoring. The system consists of two key components: (1) A multi-channel gel electrode array for high-fidelity EMG signal acquisition from target muscle groups, and (2) a novel capacitive strain sensor made of stretchable micro-cracked gold film based on Styrene Ethylene Butylene Styrene (SEBS) that exhibits exceptional performance, including >80% stretchability, >4000-cycle durability, and fast response time (<100 ms). The strain sensor demonstrates position-independent measurement accuracy, enabling robust joint angle detection regardless of placement variations. Through synchronized mechanical deformation and electrophysiological monitoring, this platform provides comprehensive movement quantification, with data visualization interfaces compatible with mobile and desktop applications. The proposed technology establishes a generalizable framework for multimodal biosensing in human motion analysis, robotics, and human-machine interaction systems.

In this paper, a dual-band bandpass filter with a wide upper stopband is proposed and designed by integrating stepped-impedance resonators (SIRs) and a low-pass filter. The operating center frequencies of the designed dual-band filter are targeted at 2.5 GHz and 5.35 GHz, respectively, to meet the frequency requirements of typical wireless communication scenarios. Notably, the filter achieves a wide upper stopband ranging from 6.1 GHz to 25 GHz, which can effectively suppress unwanted high-frequency interference signals within this frequency range and avoid mutual interference with other high-frequency communication systems. And it exhibits insertion losses of 0.12 dB (2.5 GHz) and 0.6 dB (5.35 GHz) in its two passbands to ensure minimal useful signal attenuation. The simulation results agree well with the measured results.