• Membrane Journal
• /
• v.33 no.6
• /
• pp.325-343
• /
• 2023

Direct methanol fuel cells (DMFCs) have been attracting attention as energy conversion devices that can directly supply methanol liquid fuel without a fuel reforming process. The commercial polymer electrolyte membranes (PEMs) currently applied to DMFC are perfluorosulfonic acid ionomer-based PEMs, which exhibit high proton conductivity and physicochemical stability during the operation. However, problems such as high methanol permeability and environmental pollutants generated during decomposition require the development of PEMs for DMFCs using novel ionomers. Recently, studies have been reported to develop PEMs using hydrocarbon-based ionomers that exhibit low fuel permeability and high physicochemical stability. This review introduces the following studies on hydrocarbon-based PEMs for DMFC applications: 1) synthesis of grafting copolymers that exhibit distinct hydrophilic/hydrophobic phase-separated structure to improve both proton conductivity and methanol selectivity, 2) introduction of cross-linked structure during PEM fabrication to reduce the methanol permeability and improve dimensional stability, and 3) incorporation of organic/inorganic composites or reinforcing substrates to develop reinforced composite membranes showing improved PEM performances and durability.

• Journal of Animal Science and Technology
• /
• v.65 no.4
• /
• pp.735-747
• /
• 2023

The composition of fatty acids determines the flavor and quality of meat. Flavor compounds are generated during the cooking process by the decomposition of volatile fatty acids via lipid oxidation. A number of research on candidate genes related to fatty acid content in livestock species have been published. The majority of these studies focused on pigs and cattle; the association between fatty acid composition and meat quality in chickens has rarely been reported. Therefore, this study investigated candidate genes associated with fatty acid composition in chickens. A genome-wide association study (GWAS) was performed on 767 individuals from an F2 crossbred population of Yeonsan Ogye and White Leghorn chickens. The Illumina chicken 60K significant single-nucleotide polymorphism (SNP) genotype data and 30 fatty acids (%) in the breast meat of animals slaughtered at 10 weeks of age were analyzed. SNPs were shown to be significant in 15 traits: C10:0, C14:0, C18:0, C18:1n-7, C18:1n-9, C18:2n-6, C20:0, C20:2, C20:3n-6, C20:4n-6, C20:5n-3, C24:0, C24:1n-9, monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA). These SNPs were mostly located on chromosome 10 and around the following genes: ACSS3, BTG1, MCEE, PPARGC1A, ACSL4, ELOVL4, CYB5R4, ME1, and TRPM1. Both oleic acid and arachidonic acid contained the candidate genes: MCEE and TRPM1. These two fatty acids are antagonistic to each other and have been identified as traits that contribute to the production of volatile fatty acids. The results of this study improve our understanding of the genetic mechanisms through which fatty acids in chicken affect the meat flavor.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• Korean Journal of Environmental Agriculture
• /
• v.34 no.2
• /
• pp.91-97
• /
• 2015

BACKGROUND: Earthworms are essential detritus feeders that play a vital role in the process of decomposition of organic matter and soil metabolism. The complex process of partial breakdown of organic matter and mixing with mucous and gut microbial flora in the form of earthworm cast results in the reduction of the toxicity. This study focused on the change of cast amount and pollutant contents before and after the eating of the organic waste and upland soil with the two species of earthworm. METHODS AND RESULTS: The two species of earthworms were compared to the cast production. In the upland soil material, the daily amount of worm's cast was 1.42 g in E. andrei and 0.40 g in A. agrestis. In the organic waste material, the cast of E. andrei was 0.78~0.83 g and the cast of A. agrestis. have not been collected because all earthworms died after the treatment. The heavy metals treated in the upland soil were evaluated the impact of the worm excretion. With the E. andrei, the cast production was decreased 0.1~0.8 times in zinc, 0.2~0.5 times in copper, and 0.1~0.7 times in cadmium compared to the control treatment according to the levels of concentration. With A. agrestis, the cast amount was decreased 0.3~1.1 times in zinc, 0.2~0.3 times in copper, and 0.1~2.1 times in cadmium, respectively. The changes of pollutant contents before and after the eating of the organic wastes with E. andrei were studied. In the treatment of the Alcohol Fermentation Processing Sludge and the Fruit Juice Processing Sludge, heavy metal content of the cast was increased 0.7~53.3% compared to the sludge materials. PAHs contents were decreased 50.1% in the cast of the Alcohol Fermentation Processing Sludge and 36.6% in the cast of the Fruit Juice Processing Sludge, respectively. CONCLUSION: In conclusion, although the A. agrestis was bigger than E. andrei in size and weight, the cast amount of A. agrestis was small. The two species of earthworm was less excretion with high concentration of heavy metals. While the heavy metals such as zinc, copper, and cadmium were considerably accumulated in the cast, the total compounds, PAHs were fairly decomposed. There results would provide us for restoring contaminated soil and cleaning organic wastes.

• Fire Science and Engineering
• /
• v.26 no.6
• /
• pp.99-108
• /
• 2012

These days, spontaneous ignition phenomena by oxidizing heat frequently occur in the circumstances of processing and storing waste tires. Therefore, to examine the phenomena, in this work, this researcher conducted the tests of fires of fragmented waste tires (shredded tire), closely investigated components of the fire residual materials collected in the processing and storing place, and analyzed the temperature of the starting of the ignition, weight loss, and heat of reaction. For the study, this researcher conducted fire tests with fragmented waste tires in the range of 2.5 mm to 15 mm, whose heat could be easily accumulated, and performed heat analysis through DSC and TGA, DTA, DTG, and GC/MS to give scientific probability to the possibility of spontaneous ignition. According to the tests, at the 48-hour storage, rapid increase in temperature ($178^{\circ}C$), Graphite phenomenon, smoking were observed. And the result from the DTA and DTG analysis showed that at $166.15^{\circ}C$, the minimum weight loss occurred. And, the result from the test on the waste tire analysis material 1 (Unburnt) through DSC and TGA analysis revealed that at $180^{\circ}C$ or so, thermal decomposition started. As a result, the starting temperature of ignition was considered to be $160^{\circ}C$ to $180^{\circ}C$. And, at $305^{\circ}C$, 10 % of the initial weight of the material reduced, and at $416.12^{\circ}C$, 50 % of the intial weight of the material decreased. The result from the test on oxidation and self-reaction through GC/MS and DSC analysis presented that oxidized components like 1,3 cyclopentnadiene were detected a lot. But according to the result from the heat analysis test on standard materials and fragmented waste tires, their heat value was lower than the basis value so that self-reaction was not found. Therefore, to prevent spontaneous ignition by oxidizing heat of waste tires, it is necessary to convert the conventional process into Cryogenic Process that has no or few heat accumulation at the time of fragmentation. And the current storing method in which broken and fragmented materials are stored into large burlap bags (500 kg) should be changed to the method in which they are stored into small burlap bags in order to prevent heat accumulation.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.24 no.3
• /
• pp.63-74
• /
• 2016

The objective of this study was to investigate the variations of physico-chemical properties during the swine manure composting, sawdust as the bulking agent was composted at different points (Top layer, Side of middle layer, Bottom layer). Air suction system with constant bottom aeration in bench scale reactors (30 L). The highest temperature was reached in the range of $58^{\circ}C$ to $62^{\circ}C$ on $3^{rd}$ day and this thermophilic phase (> $50^{\circ}C$) was continued for 3 days in all the treatment mixtures. However, the temperature was gradually decreased to room temperature at the end of 60 day composting process. Except control, the discharged ammonia ($NH_3$) was a maximum in the treatment order of Top layer>Bottom layer>Side of middle layer as 500 ppm, 162 ppm and 120 ppm, respectively, on the $4^{th}$ day and showing that Top layer point Air suction produce much more ammonia content than the other point. During the composting process, the total Kjeldahl nitrogen (TKN) was gradually increased due to the mass loss in the composting mixtures. At the same time, C/N ratio was decreased to Top layer, 13; Side of middle layer, 12 and Bottom layer, 13 at Air suction points. The significant reduction of C/N ratio in all different air suction system when manure was matured. The $NH_4-N$ to $NO_3-N$ ratio was recorded as 10.52 at the initial stage of the compost mixtures and reduced to 0.97 (Top layer), 0.70 (Side of middle layer), 3.2 (Bottom layer) because of manure decomposition. The overall results revealed that Top layer and Side of middle layer Air suction is a suitable option when compared other point for high quality composts.

• Korean Journal of Environmental Agriculture
• /
• v.23 no.4
• /
• pp.220-227
• /
• 2004

This study showed that thermophilic anaerobic acid fermentation of food wastes had an enhanced hydrolysis capability and improvement of acidification efficiency. Influence of pH on the anaerobic hydrolysis and acidogenesis was investigated to determine the proper alkalinity in the thermophilic fermentation of food wastes. The results of putting NaOH as alkali to evaluate hydrolysis and acid fermentation efficiency In acid fermentation process of food wastes showed that the food wastes pretreated with 0.05 g NaOH/g TS had the maximum 12,600 mg/L of VFAs concentration during HRT 3 days in $55^{\circ}C$ thermophilic condition and the maximum 9,700 mg/L of VFAs concentration during HRT 5 days in $35^{\circ}C$ mesophilic condition. The accomplishment of high VFAs concentration resulted from that the main component of food wastes such as cellulose, lignin and etc. is performed active chemical decomposition by alkali in thermophilic condition. The major components of VFAs produced from the thermophilic acid fermentation process of food wastes were the short chain fatty acids such as acetic acid, butyric acid, and propionic acid.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Economic and Environmental Geology
• /
• v.38 no.6 s.175
• /
• pp.675-687
• /
• 2005

The rocks of the five storied stone pagoda in the Jeongrimsaji temple site are 149 materials in total with porphyritic biotite granodiorite. They include pegmatite veinlet, basic xenolith and evenly developed plagioclase porphyry. This stone pagoda has comparably small fracture and cracks which are farmed in the times of rock properties, but surface exfoliation and granular decomposition are in process actively since the rocks are generally weakened from the influence of air contaminants and acid rain. Structural instability of constituting rocks in the 4th roof materials are observed to occur from distortion and tilt. Such instability is judged to threat stability of the upper part of the stone pagoda. Also, chemical weathering is operating even more as the contaminants, ferro-manganese hydroxides eluted from water-rock interaction on the rock surface. Most of the rock surface is covered with yellowish brown, dark black and light gray contaminants, and especially occur in the lower part of the roof rocks on each floor. The roof underpinning rocks are severe in surface pigmentation from manganese hydroxides and light gray contaminants. The surface of rocks lives bacteria. algae, lichen, or moss and diverse productions in colors of light gray, dark Bray and dark green. Grayish white crustose lichen grows thick on the surface with darkly discolored by fungi and algae in the first stage on basement rocks, and weeds grows wild on the upper part of each roof rocks. This stone pagoda must closely observe the movements of the upper part rock materials through minute safety diagnosis and long term monitoring for structural stability. Especially since the surface discoloration of rocks and pigmentation of secondary contaminants are severe, establishment of general restoration and scientific conservation treatment are necessary through more detailed study for this stone pagoda.

• Korean Journal of Environmental Agriculture
• /
• v.32 no.3
• /
• pp.179-184
• /
• 2013

BACKGROUND: Accurate estimates of total direct $CH_4$ emissions from croplands on a country scale are important for global budgets of anthropogenic sources of $CH_4$ emissions and for the development of effective mitigation strategies. Methane production resulted by the anaerobic decomposition of organic compounds where $CO_2$ acts as inorganic electron acceptor. This process could be affected by the addition of rice straw, water management and rice variety itself. METHODS AND RESULTS: Rice (Oryza sativa L. Japonica type, var Samkwangbyeo) was cultivated in four plots: (1) Nitrogen-Phosphorus-Potassium (NPK) ($N-P_2O_5-K_2O$:90-45-57 kg/ha); (2) NPK plus 3 Mg/ha rice straw (RS3); (3) NPK plus 5 Mg/ha rice straw (RS5); (4) NPK plus 7 Mg/ha rice straw (RS7) for 3 years (2010-2012) and the rice straw incorporated in fall (Nov.) in Gyeonggi-do Hwaseong-si. Gas samples were collected using the closed static chamber which were installed in each treated plot of $152.9m^2$. According to application of 3, 5, 7 Mg/ha of rice straw, methane emission increased by 46, 101, 190%, respectively, compared to that of the NPK plot. CONCLUSION(S): We obtained a quantitative relationship between $CH_4$ emission and the amount of rice straw applied from rice fields which could be described by polynomial regression of order 2. The emission scaling factor estimated by the relationship were in the range of IPCC GPG (2000).