As an alternative fuel that can be used in SI engines, LPG is a clean fuel with larger H/C ratio compared to gasoline, low CO 2 emission, and small amount of pollutants such as sulfur compounds. In the Spark-ignition (SI) engine, Direct injection (DI) technology can significantly increase the engine volumetric efficiency and decrease the need for a throttle valve. DI allows engine operation with the stratified charge, which enables a relatively higher combustion efficiency. Stratified charge can be supplied to nearby spark plugs to allow for overall lean combustion, which improves thermal efficiency and can cope with problems regarding emission regulations. In this study, a visualization experiment system that consists of visualization combustion chamber, air supply control system, emission control system, LPG fuel supply system, electronic control system and image data acquisition system was designed and manufactured. For all cases for which ignition was successful, flame propagation image was digitally recorded using ICCD camera, and the recorded flame propagation characteristics were examined. This study, in its results, is expected to make a contribution in terms of important data for the design and optimization of a Spark-ignited direct injection (SIDI) LPG engine.

In this study, the exhaust characteristics of the diesel engine for the change of the mixing ratio of biodiesel fuel were quantitatively analyzed by using the numerical analysis method. As the fuel in the experiment, the diesel and biodiesel (waste cooking oil, soybean oil), mixture BD2 (diesel only), BD3, BD5, BD20, BD50 and BD100 were used. The injection pressure (p inj ) was set at 400 bar, 600 bar, 800 bar, 1000 bar and 1200 bar as the experimental variables. The concept of the standard deviation, Pearson's correlation coefficient and Spearman rank-order correlation coefficient based on the statistics were introduced in order to analyze the exhaust characteristics of the quantitative NOx and Soot according to the injection pressure and the mixing ratio of biodiesel blended fuel. The regression method was introduced in order to obtain the increasing and decreasing aspects of NOx and Soot that can not be known from the correlation coefficients alone. From the study it is inferred that, for the waste cooking oil, NOx and Soot can be simultaneously reduced through control of the mixing ratio in the region of p inj =400 bar and p inj =600 bar, and the Soot can be reduced without affecting the emission of NOx for p inj more than 800 bar. For the soybean oil, NOx and Soot can be simultaneously reduced at p inj =400 bar and the Soot can be reduced without affecting the emission of NOx at p inj =600 bar.

JPEG is a widely used image compression standard that shows a reasonable image quality for a wide range of compression rates. However, when compressed with a low compression quality factor to increase the compression rate, it brings a large loss in the frequency domain, which turns into visible artifacts in the image domain. Accordingly, removing artifacts in JPEG-compressed images has been an essential image restoration task. While most previous methods use the information on compression quality factors available in the header of the JPEG file, we note this approach is not practical in the real-world scenario because many compressed images' metadata are not exist. To deal with this issue, we propose a new method based on a Deformable Offset Gating Network (DOGNet) and a Variational Autoencoder (VAE). We train the overall network in an end-to-end manner, where the role of the VAE is to guide the offset of the deformable convolution to flexibly deal with images compressed with diverse and unknown quality factors. Extensive experiments validate that our method achieves better or comparable results to the state-of-the-art methods in JPEG Artifact Removal.