Search

Anomaly detection plays an essential role in monitoring dependable systems and networks such as computer clusters, water treatment systems, sensor networks, etc. However, anomaly detection nowadays remains a big challenge since previous researches suffer from inaccessible anomaly labels and inconsistent data types. Therefore, we propose a robust unsupervised anomaly detection framework (RUAD) to tackle the above problems. RUAD combines a deep AutoEncoder and a robust layer to extract the latent representations of data and separate normal data from abnormal data respectively, then utilizes Gaussian Mixture Model (GMM) to learn the distribution of normal data. In addition, our model can adapt to different types of data by simply modifying the structure of the deep AutoEncoder. Extensive experiments show that RUAD outperforms state-of-art anomaly detection techniques. [ABSTRACT FROM AUTHOR]

Glass fiber (GF) reinforced polycarbonate (PC) composites were prepared for laser direct structuring (LDS) applications. A small amount of styrene–maleic anhydride (SMA) was introduced to enhance GF/PC interfacial interactions. The PC/SMA/GF LDS composites using metal complex as LDS additive were investigated for applicability in copper circuit development. Rough surface patterns by laser irradiation under different laser parameters were measured by microscopy and Fourier infrared spectrometry. Copper particles and plating layer by subsequent electroless copper plating procedure was observed using microscopy. Thickness of plating layer and adhesion between layer and matrix were also evaluated. The results showed that laser repetition and scanning speed led to different resolutions and ablated surfaces without structural changes of composites. Copper particles gradually deposited, grew, and interconnected during metallization procedure. Plating layer was successfully formed at an optimum LDS additive loading, and appropriate scanning speed and repetition of laser. In addition, the plating layer displayed disparate thickness and distribution, owing to different activated surfaces by irradiation. Severe ablation or unetched parallel regions resulted in leaky or discontinuous plating layer. Better plating microstructure and higher adhesion were obtained for the composite material with 1.0 wt% LDS additive loading, supporting its extensive development and practical application in LDS technology. POLYM. ENG. SCI., 60:860–871, 2020. © 2020 Society of Plastics Engineers [ABSTRACT FROM AUTHOR]

Abstract: Chopped carbon fibers (CCFs) coated by poly(ethylene terephthalate) (PET) were treated using 3‐aminopropyltriethoxysilane (Si), and then used for fabricating thermally conductive polycarbonate/CCF (PC/(Si + PET)@CCFi) composites with various filler contents. Optical microscopy measurements indicated that CCF fillers in the composites had a wide length distribution and, with CCF content increasing, the mean fiber length and mean fiber aspect ratio steadily decreased. Compared to PC/(Si + PET)@CCF0, the molten flowability of composites declined. Thermogravimetric analysis and pyrolysis–gas chromatography–mass spectrometry characterizations showed that introducing CCFs could enhance the thermal stability of the composites and restrict the pyrolytic degradation of the polymer matrix. Moreover, both in‐plane and through‐thickness thermal conductivities increased with increasing CCF content. These conductivities at 50% CCF content reached 2.45 and 0.59 W m−1 K−1, and were improved by 11.25 and 1.95, respectively. Furthermore, the tensile and flexural properties of the composites continuously enhanced, accompanying decreased impact strength and elongation at break. Fractography revealed that fillers dispersed in matrix uniformly and a great many fiber cross‐sections and circular cavities coexisted in fracture surface, besides a clear interface and reduced long fibers. And the fracture behavior was mainly the breaking and peeling of fibers from matrix. © 2018 Society of Chemical Industry [ABSTRACT FROM AUTHOR]

The Belt and Road Initiative (BRI) is designed to intensify reciprocal trade preferentiality between China and the Belt-Road countries. However, there has been little research empirically examining the policy effects on the trade links between China and the involved countries. This paper attempts to evaluate the BRI effects quantitatively by constructing a new bilateral revealed trade preference index to measure the bilateral trade preferentiality between China and its 114 trading partners. Using a difference in differences model, we show that the trade of China with the Belt-Road countries has become more preferentially linked since the implementation of the BRI. In particular, the bilateral revealed trade preference index between China and the Belt-Road countries has grown approximately 8% faster than has that with the non-Belt-Road countries. We further show that the BRI effects are heterogeneous across different regions. The bilateral trade links have been more significantly intensified in the regions of the China–Indochina Peninsula Economic Corridor, the China–Pakistan Economic Corridor, the China–Central Asia–West Asia Economic Corridor and the Bangladesh–China–India–Myanmar Economic Corridor. The findings strongly indicate that BRI has been acting as a catalyst for intensifying bilateral trade preferentiality between China and the Belt-Road countries. [ABSTRACT FROM AUTHOR]