Search

For the management of rotating machines, machine learning (ML) has been researched with the use of feature parameters that have physical and statistical meanings of vibration signals. Genetic algorithm (GA) and principal component analysis (PCA) are the algorithms used for the selection or extraction process of the features; equipment condition. This study proposes a new method to maximize the advantages of the extraction and selection algorithms, thereby improving the fault classification performance. The proposed method is estimated in a variety of equipment conditions by selecting and extracting the effective features for status classification. To evaluate the performance of the fault classification through feature selection and extraction of the ML, a comparative analysis with the proposed method and the original method is also performed. With Lab-scale gearbox, several types of fault tests are conducted, and seven different fault types of equipment conditions, including the normal status, are simulated. The results of the experiments show that, the performance of classification of GA for feature selection is 85%, while PCA for feature extraction is 53%. The performance result of the proposed method for fault classification is 95%, meaning that the performance of fault diagnosis is more efficient in terms of discriminative learning than the original method. Therefore, the proposed method with feature extraction and selection algorithm can improve the fault classification performance by 10% and more for fault diagnosis through ML.

The stability of intermetallic second-phase particles (SPPs) in coated Zircaloy-2 was studied in 700 °C steam environments up to 20 hours. Hydrogen generated from high-temperature steam oxidation of uncoated Zr-induced δ-hydrides formation in the Zircaloy matrix. Synchrotron XRD demonstrated that longer exposure times increased hydride peak intensity and decreased intermetallic SPPs’ peak intensity. Cross-sectional SEM analysis verified the intermetallic SPPs’ volume fraction reduction. The size distribution of intermetallic SPPs was characterized and larger particles were dissolved at longer oxidation time. A correlation between the hydrogen concentration and the volume fraction of intermetallic SPPs at 700 °C steam environment was found, with the volume fraction of SPPs decreasing as hydrogen concentration increases.

ZnO-graphene quasi core-shell quantum dot (QD) structures in which the inner ZnO QDs are covered with graphene nanoshells have been synthesized via a simple solution process method. The outer graphene nanoshells were identified as a single graphene layer using high resolution transmission electron microscopy (HR-TEM). Zn-O-C (graphene) chemical bonds between the inner ZnO QDs and the oxygen-containing functional groups introduced into the graphene layer are believed to be important in the formation of the consolidated quasi core-shell QD structure. A multilayer structure organic ultraviolet (UV) photovoltaic (PV) device was fabricated using ZnO-graphene core-shell QDs as the absorption layer. A quenching behavior as large as 71% near the UV emission peak for the ZnO-graphene core-shell QDs was observed in the photoluminescence. Density of state (DOS) calculations for the graphene using density functional theory (DFT) revealed that the static quenching can be attributed to a faster charge separation via the direct electron transfer from the conduction band (CB) of the ZnO QDs to the induced lowest unoccupied molecular orbitals (LUMO) of the graphene nanoshell resulting from the Zn-O-C bonding. This charge separation mechanism was confirmed experimentally using time-correlated single photon counting (TCSPC) measurements. The calculated average lifetime of 0.13 ns and 0.165 ns of the 375 and 383 nm UV emissions, respectively, for the ZnO-graphene core-shell QDs were approximately 10 times faster than those of 1.86 ns and 1.83 nm for the reference ZnO QDs; this is indicative of the existence of an additional high efficiency relaxation channel. The observed saturation current density (J sc), open circuit voltage (V oc), fill factor (FF), and power conversion efficiency (η) were 196.4 μA/cm2, 0.99 V, 0.24, and 2.33%, respectively. In this study, it was found that the UV power conversion efficiency of ZnO QDs could be significantly improved by invoking a fast photoinduced charge separation and the subsequent transport of carriers to the collecting electrodes through conjugation with highly conductive graphene nanoshell acceptors to the ZnO QDs donor.

Hybrid nanostructures combining inorganic materials and graphene are being developed for applications such as fuel cells, batteries, photovoltaics and sensors. However, the absence of a bandgap in graphene has restricted the electrical and optical characteristics of these hybrids, particularly their emissive properties. Here, we use a simple solution method to prepare emissive hybrid quantum dots consisting of a ZnO core wrapped in a shell of single-layer graphene. We then use these quantum dots to make a white-light-emitting diode with a brightness of 798 cd m(-2). The strain introduced by curvature opens an electronic bandgap of 250 meV in the graphene, and two additional blue emission peaks are observed in the luminescent spectrum of the quantum dot. Density functional theory calculations reveal that these additional peaks result from a splitting of the lowest unoccupied orbitals of the graphene into three orbitals with distinct energy levels. White emission is achieved by combining the quantum dots with other emissive materials in a multilayer light-emitting diode.

A nonparametric inversion method has been employed to obtain the detailed attenuation characteristics of horizontal shear-wave Fourier spectra at high frequencies based on the large spectral dataset mostly from small earthquakes occurring at shallow depths in the southern part of the Korean Peninsula. Nonparametric expressions of the spectral attenuation were evaluated for 59 frequency bands in the range of 1 Hz ≤ f ≤ 30 Hz and hypocentral distance ranging from 10 to 400 km by applying the nonparametric inversion method to the observed spectral dataset. Based on the nonparametric expressions, three linear regions roughly separated at 60 and 110 km in distance could be identified and the use of the conventional functional form of Q0fη assuming the geometrical spreading of R−0.5 could be validated for 3 Hz ≤ f ≤ 30 Hz and R ≥ 190 km with the best fitting result of 292f0.53. In the near-source region of R < 70 km, the spectral attenuation due to the geometrical spreading ranged from R−1.3 to R−1.5 irrespective of the Q-values. The distance-attenuation trend of the recent observed dataset from a moderate inland earthquake in Korea was also well predicted by the inverted nonparametric path term.

Flexible audible display (FAD) based on PVDF(Poly-vinylidene fluoride) film is researched to combine piezoelectric property with light emitting diode. Indium-Tin-Oxide (ITO) is deposited on PVDF film at low temperature condition with the vacuum web coating. Under the optimized working pressure and the oxygen/argon pressure ratio, ITO on PVDF shows good electrical property of sheet resistance about 120 ohm/sq at 160 nm thickness, good transmittance of 76%, and sound pressure level more than 70 dB. OLED device with Al/LIF/Alq3/ NPB/ 2T-NATA structure is fabricated on ITO/PVDF and shows the turn-on voltage of 3.5 V and the maximum luminescence of 1011.2 cd/cm2 at the driving voltage of 13 V. C-F chemical bondings are measured to decrease in the XPS spectrum of ITO/PVDF interface. This is originated from the bonding scission caused by incident energetic particles, such as sputtered atoms, backscattered Ar ions, and negative O ions during the sputtering and deposition process. These particles have energy, which is simulated by TRIM and is more than 5 eV, enough to cause the scission of C-F bondings with binding energy of 4 eV/atom.