Printed circuit board (PCB) defect inspection is a key task in industrial automatic optical inspection (AOI). Its deployment in low-label scenarios remains difficult because PCB defects are often small, weakly contrasted, and unevenly distributed across categories and scales. Semi-supervised object detection can reduce annotation demand by using unlabelled images, but its performance depends strongly on the reliability of pseudo labels. In PCB inspection, confidence scores are shaped not only by defect category but also by object scale, which makes fixed or class-only filtering insufficient for small-defect supervision. This paper proposes dual-granularity pseudo-label calibration (DGPC) for semi-supervised PCB surface defect detection. DGPC first fuses weak-view and strongview teacher predictions into a unified candidate pool. It then estimates pseudo-label thresholds from both classlevel and class–scale statistics, and applies a scale-dependent lower-bound relaxation to retain informative smalldefect candidates. The method is applied only during training and leaves the detector architecture and inference procedure unchanged. Experiments on Deep PCB at 5%, 10%, and 20%labeled ratios show that DGPC improves pseudo-label-based training under low-label settings. The gains are most consistent when the teacher provides a sufficiently informative candidate pool, especially at 10% and 20% labelled ratios. Ablation, stability, crossdetector, pseudo-label statistics, scale-wise, qualitative, and sensitivity analyses show that class–scale-aware calibration improves pseudo-label selection for annotation-efficient PCB inspection.

Conventional amplitude-based indicators may be insufficient to characterise the effects of varying crack elevation in offshore wind turbine towers. To address this limitation, a bidirectional multi-feature sensitivity analysis was conducted using a finite element model based on the NREL 5 MW reference wind turbine. Transient dynamic analyses were performed in ANSYS under combined stochastic wind–wave loading. One intact case and five cracked cases with identical crack dimensions but different elevations were considered. Acceleration responses in the X and Y directions were extracted at eight measurement points. The root-mean-square value, peak absolute acceleration, dominant frequency, and frequency-band energy indices were calculated, and their sensitivities to variations in crack elevation were quantified using the absolute relative change with respect to the intact case. Under the adopted loading conditions, the X-direction response exhibited a larger overall amplitude. However, the RMS and frequency-band energy indices exhibited higher relative sensitivities in the Y direction, whereas the peak-value sensitivity was higher in the X direction. No detectable change in dominant frequency was observed at the adopted frequency resolution. Among the investigated features, the frequency-band energy in the 0–0.05 Hz band exhibited the highest sensitivity in both directions. Among the eight candidate measurement points, P1 showed the highest sensitivity under the considered loading and crack cases, and the sensitivity generally decreased with increasing measurement height. These findings provide a basis for damage-sensitive feature selection and sensor-layout optimisation in crack monitoring of offshore wind turbine towers.

Military tents are crucial for military operations. Existing military tents use insulating and photonically inert textiles, which cannot be used as electric or light-driven heaters, significantly restricting their application in high-altitude combat zones. Considering the simultaneous threats of rain, snow, electromagnetic radiation and bullet impact in the battlefield, a kind of multifunctional protective material is necessary for military tents. In this paper, a multifunctional material including MXene as the functional layer, nylon fabric as the base material, and room temperature vulcanized silicone rubber as the surface waterproof layer is developed. The results show that: Results show that the synergistic effect of MXene and silicone rubber modifies the surface roughness and surface energy of pure nylon fabric, increasing its hydrophobic contact angle from 38.3° to 105.3°. MXene can form continuous conductive networks on fabric surfaces at varying concentrations. The electromagnetic shielding effectiveness of MXene and silicone rubber modified nylon (MS-Nylon) in the X-band spectrum increases to 17.64 dB to 18.54 dB when the mass fraction of MXene reaches 18.87%. Concurrently, the fabric's photothermal performance was significantly enhanced. Comparing with Nylon without photothermal property, the surface temperature of MS-Nylon reached 42.29 °C after exposure to sunlight (illuminance of 88.5 × 103 LUX). Following exposure to incandescent (200 W) and infrared (375 W) lamps, MS-Nylon temperatures reached 49.45 °C and 90.00 °C respectively. Notably, nylon's inherent mechanical and bulletproof properties remained nearly unchanged. This work provides valuable insights for the design of next generation multifunctional protective equipment.

정책 학습(policy learning)에 관한 상당수의 연구는 다양한 영향력 집단의 역학 관계가 정책 입안과 실행을 형성해 온 분권화된 시스템을 바탕으로 그 개념을 정립해 왔다. 이러한 분석은 대개 변화를 도모하기 위한, 특히 정책 개선을 위한 다양한 행위자들 사이의 권력 관계에 초점을 맞춘다. 본 논문은 상대적으로 위계적(hierarchical)이고 중앙집권적인 맥락 에서의 정책 학습 분석을 통해 이러한 기존의 관점에 새로운 시각을 더한다. 이 논문은 국가 및 지역 수준에서 중국 교육 부문의 ‘쌍감(雙減, double reduction)’ 정책의 수립과 실행에 저자가 직접 참여한 경험을 바탕으로 한다. 특히 두 가지 정책 학습 사례에 초점을 맞춘다. 첫째는 초기 정책 및 후속 정책을 고안하 기 위한 하향식(top-down) 학습이었고, 둘째는 기존 정책을 개선하기 위한 상향식(bottom-up) 학습이었다. 이러한 대조는 다양한 정책 행위자들이 학습을 다르게 인식할 수 있으며, 서로 다른 목적을 위해 학습한다는 것을 보여준다. 또한 정책 학습이 항상 선형적이거나 과학적인 것은 아니며, 학습이 언제나 정책 개선으로 직결되지는 않는다는 점을 시사한다. 아 울러 중앙·지방의 정책결정자, 싱크탱크 및 대학의 전문가, 그리고 여러 지역과 변화하는 시대 상황 속 일반 대중 간의 상호작용이 정책학습에 긍정적으로든 부정적으로든 영향을 미칠 수 있음을 덧붙인다.

The rapid development of precise diagnosis and treatment of diabetes has imposed higher requirements for the sensitivity, selectivity, and stability of glucose sensors. Given the bottlenecks of traditional carbon nanotubes in electrochemical sensing applications, such as low purity, numerous structural defects, and poor biocompatibility, this paper systematically reviews the mechanism of glucose detection, preparation and purification of high-purity carbon nanotubes, and the preparation methods and advantages of carbon nanotube-metal nanoparticle composite electrodes. To address these critical limitations, this review focuses on three interconnected aspects of CNT-based glucose sensing technology. First, the catalyst regeneration, dynamic process control and green carbon source substitution have effectively overcome the problems of high energy consumption, low purity and environmental burden of traditional methods. Second, the purification and innovative functionalization of carbon nanotubes have significantly improved their purity and electrochemical performance. Finally, the preparation method of a carbon nanotube-metal nanoparticle composite electrode is described. It not only achieves the precise spatial positioning of the catalytic active center, but also significantly enhances the long-term stability of the electrode through the synergistic regulation of chemical bonding strength and interface electronic structure. These advancements lay a theoretical foundation for the development of a new generation of wearable sensors with antibiofouling properties and resistance to complex physiological interferences.

Poor bonding occurs with resin due to surface inertness of carbon fiber (CF), so CF surfaces were often treated. In some common surface treatments, sizing was a simple and effective modification method. Polyurethane (PU) was used as the main component of sizing agents due to its similar structure to polyamide 6 (PA6). The CF/PA6 composites’ interfacial properties were improved using PU as a sizing agent. Meanwhile, in this paper, glycidol (GLD) was introduced into the PU emulsion so that the epoxy group reacted with the carboxyl group on the acidified CF. After testing, when the content of glycidyl in the sizing agent is 2%, the CF/PA6 composites showed an important improvement in tensile, impact, and flexural strengths, which increased by 49.4%, 94.6%, and 53.2%, respectively. In addition, the effect of modified WPU sizing agents with different GLD contents on the properties of CF/PA6 composites was investigated.

The ZnCl2 chemical activation method is widely employed for the preparation of biomass-derived porous carbons. In most of the related studies, the emphasis lies on investigating how experimental preparation conditions impact the performance of the final products. However, the performance of the porous carbon also depends on the chemical structure of the carbon source. In this study, we used alkali lignin, ammoxidized lignin and sodium lignosulfonate as carbon sources to prepare porous carbon through ZnCl2 activation. The influence of the chemical structures of lignin on the activation process is explored. The porous carbons prepared from alkali lignin (ALC) and ammoxidized lignin (AOLC) both exhibit similar and relatively high specific surface areas (ALC: 1164 m2 g− 1, AOLC: 1156 m2 g− 1) and capacitance contribution ratios (ALC: 80.6%, AOLC: 79.4%). The porous carbon prepared from sodium lignosulfonate has a specific surface area of 890 m2 g− 1 and a mesopore ratio of 26.1%, with the capacitance contribution accounting for only 75.1%. ZnS and NaCl generated during the activation process involving sodium lignosulfonate can partially enable mesopores by template effect, which in turn results in lower electrochemical properties. This study explores the reasons for the differences in ZnCl2 activation on different lignins, providing data to support research on the mechanism of how lignin structure influences ZnCl2 activation.

본 연구는 뉴미디어 환경이 문화예술 창의 표현의 방식과 시 각화 구조를 어떻게 변화시키는지 분석하기 위해 문헌 검토와 사례 분석을 병행하여 수행되었다. 본 연구는 텍스트·이미지·영 상·아이콘·그래픽·모션과 같은 시각 요소들이 디지털 플랫폼에서 융합되면서 문화예술 창의 표현의 범위가 확장되고, 이용자의 인지적·정서적·행위적 반응이 다층적으로 형성되는 현상을 확인 하였다. 또한 연구는 뉴미디어 환경에서 시각화 전파의 특징이 실시간 상호작용성, 알고리즘 기반 노출, 이용자 참여도 확장으 로 구조화되며, 이는 창의 콘텐츠가 확산되는 경로와 속도, 해석 방식에 유의미한 차이를 발생시킨다는 점을 밝혔다. 특히 연구 는 예술·디자인·광고 영역에서의 문화예술 창의가 단순한 이미지 표현을 넘어 참여 기반의 ‘확장형 창의(extended creativity)’로 재구성되고, 이는 디지털 정체성 형성, 감성 기반 소비, 참여적 문화의 확산 등 사회문화적 변화와 긴밀히 연결되어 있음을 규 명하였다. 본 연구는 시각화 전파가 창의 산업의 구조적 변화를 견인하고, 뉴미디어 기반 창의 전략이 향후 문화예술 및 디자인 산업 전반의 혁신 방향을 제시할 수 있다는 함의를 제공한다. 본 연구는 향후 뉴미디어 창의 연구가 이용자 경험·디지털 감정· 시각 알고리즘 분석 등으로 확장될 필요가 있음을 제안하며, 문 화예술 기반 창의의 시각화 전파를 이해하기 위한 이론적·방법 론적 기초를 마련하였다.