Unsupervised anomaly detection for earthquake detection on Korea high-speed trains using autoencoder-based deep learning models.

We propose a method for detecting earthquakes for high-speed trains based on unsupervised anomaly-detection techniques. In particular, we utilized autoencoder-based deep learning models for unsupervised learning using only normal training vibration data. Datasets were generated from South Korean high-speed train data, and seismic data were measured using seismometers nationwide. The proposed method is compared with the conventional Short Time Average over Long Time Average (STA/LTA) model, considering earthquake detection capabilities, focusing on a Peak Ground Acceleration (PGA) threshold of 0.07, a criterion for track derailment. The results show that the proposed model exhibit improved earthquake detection capabilities than STA/LTA for PGA of 0.07 or higher. Furthermore, the proposed model reduced false earthquake detections under normal operating conditions and accurately identified normal states. In contrast, the STA/LTA method demonstrated a high rate of false earthquake detection under normal operating conditions, underscoring its propensity for inaccurate detection in many instances. The proposed approach shows promising performance even in situations with limited seismic data and offers a viable solution for earthquake detection in regions with relatively few seismic events.

Vivid Colored Cooling Structure Managing Full Solar Spectrum via Near-Infrared Reflection and Photoluminescence.

Colored radiative cooling (CRC) offers an attractive alternative for surface and space cooling, while preserving the aesthetics of an object. However, there has been no study on the CRC using phosphors in regard to vivid coloration, sophisticated performance investigation, retention of properties, functionality, and structural flexibility all at once. Thus, to manage the entire solar spectrum, a colored cooling structure comprising a near-infrared (NIR)-reflective bottom layer and a top colored layer with a phosphor-embedded polymer matrix is proposed. The structure is paintable, vividly colored, hydrophobic, and ultraviolet (UV) and water resistant. In the daytime outdoor measurement, the structure with red, orange, and yellow colors exhibited lower temperature than a control group using commercial white paint by 4.7 °C, 7.2 °C, and 7.4 °C, respectively. After precise theoretical and experimental time-tracing temperature validation, the CRC performance enhancement from NIR reflection and photoluminescence effects was thoroughly analyzed, and a temperature reduction of up to 16.1 °C was achieved for the orange-colored structure. Furthermore, experiments of hydrophobicity infusion and exposure to UV and deionized water verified the durability of the colored cooling structure. In addition, flexible-film-type colored cooling structures were demonstrated using different bottom reflective layers, such as a silver thin film and porous aluminum oxide particle-embedded poly(vinylidene fluoride-co-hexafluoropropylene), suggesting the potential applicability of these colored cooling structures for vivid-colored, functional, and durable CRC.

Hierarchical Co-Pi Clusters/Fe2O3 Nanorods/FTO Micropillars 3D Branched Photoanode for High-Performance Photoelectrochemical Water Splitting.

In this study, an efficient hierarchical Co-Pi cluster/Fe2O3 nanorod/fluorine-doped tin oxide (FTO) micropillar three-dimensional (3D) branched photoanode was designed for enhanced photoelectrochemical performance. A periodic array of FTO micropillars, which acts as a highly conductive "host" framework for uniform light scattering and provides an extremely enlarged active area, was fabricated by direct printing and mist-chemical vapor deposition (CVD). Fe2O3 nanorods that act as light absorber "guest" materials and Co-Pi clusters that give rise to random light scattering were synthesized via a hydrothermal reaction and photoassisted electrodeposition, respectively. The hierarchical 3D branched photoanode exhibited enhanced light absorption efficiency because of multiple light scattering, which was a combination of uniform light scattering from the periodic FTO micropillars and random light scattering from the Fe2O3 nanorods. Additionally, the large surface area of the 3D FTO micropillar, together with the surface area provided by the one-dimensional Fe2O3 nanorods, contributed to a remarkable increase in the specific area of the photoanode. Because of these enhancements and further improvements facilitated by decoration with a Co-Pi catalyst that enhanced water oxidation, the 3D branched Fe2O3 photoanode achieved a photocurrent density of 1.51 mA cm-2 at 1.23 VRHE, which was 5.2 times higher than that generated by the non-decorated flat Fe2O3 photoanode.