Search

The effect of the internal angle on downward flame spread over a thick folded plate is experimentally investigated in this study. The lateral side of a thermoplastic plate is cut at desired angles and two of these plates are welded to form a single folded sample. All the faces other than the front and two 5-mm areas of the front face from each side were coated to inhibit the combustion. Experiments are conducted for various folded plates of different angles, 60°, 90°, 120°, and 180°. Following ignition at the upper edge of the sample, the opposed flame spread on the front face is observed and the flame spread rate at various positions is quantified by image analysis. The flame spreads more rapidly at the folding edge, while the other parts then accelerate to catch up with the corner spread and, eventually, all the parts across the width achieve the same flame spread rate as the folding edge. This result indicates that the flame spread rate at the folding edge is the characteristic value which represents the steady-state of the system of interest. A simple model is developed considering change of heated volume against the flame and change of the induced flow velocity due to the geometry. The former effect is formulated from geometrical consideration in the vicinity of the edge, while the latter is formulated based on an experimental fact that the flame height is inversely proportional to the internal angle. A prediction formula of the flame spread rate at the folding edge is established by modifying the conventional one for the flat plate. The calculated flame spread rate shows reasonably good agreement with the experimental data. This study helps fundamental understanding of the flame spread behavior of practical combustibles such as pillars or rods with various cross-sectional shapes. [ABSTRACT FROM AUTHOR]

This paper presents a machine learning approach to automatically classifying post-harvest vegetal species. Color images of vegetal species were applied to convolutional neural networks (CNNs) and support vector machine (SVM) classifiers. We focused on okra as the target vegetal species and classified it into two quality types. However, our approach could also be applied to other species. The machine learning solution consists of several components, and each design process and its combinations are essential for classification quality. Therefore, we carefully investigated their effects on classification accuracy. Through our experimental evaluation, we confirmed the following: (1) in color space selection, HLG (hue, lightness, and green) and HSL (hue, saturation, and lightness) are essential for vegetal species; (2) suitable preprocessing techniques are required owing to the complexity of the data and noise load; and (3) the diversity extension of learning image data by mixing different datasets obtained under different conditions is quite effective in reducing the overfitting possibility. The results of this study will assist AI practitioners in the design and development of post-harvest classifications based on machine learning. [ABSTRACT FROM AUTHOR]

A radial correlation ECE radiometer diagnostic has been developed for electron temperature fluctuation measurements in the helical-axis heliotron device, Heliotron J. The radiometer consists of two heterodyne detection systems. One system scans the frequency of a local oscillator from 52 to 64 GHz with a single intermediate frequency filter, and the second system has a fixed frequency, 56 GHz local oscillator with four intermediate frequency filters. This frequency range covers measurement positions spanning from the plasma core to the half radius. Laboratory tests indicate that each system has narrow intermediate frequency bandwidth and high-sensitivity over a large dynamic range. During plasma experiments with NBI heating, radiation temperature fluctuation measured by the CECE radiometer decrease with increasing ECCD commensurate with previous measurements of energetic particle driven modes on Heliotron J.