• 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.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.4 no.3
• /
• pp.255-263
• /
• 2006

Vol-oxidizer is a device to convert $UO_2$ pellets into $U_3O_8$ powder and to feed a homogeneous powder into a Metal Conversion Reactor in the ACP(Advanced Spent Fuel Conditioning Process). In this paper, we propose a design model of the vol-oxidizer, develop the new vol-oxidizer with a capacity of 20 kg HM/batch in $UO_2$ pellets, and conduct a verification for the device. Design considerations include the internal structure, the capacity, the heating position of the device, and the size. The dimensions of the new vol-oxidizer are decided by the design model. We determine a permeability test of the $U_3O_8$ measuring the temperature distribution, and the volume of $UO_2$ and $U_3O_8$. We manufactured the new vol-oxidizer for a 20 kg HM/batch in $UO_2$ pellets, and then analyzed the characteristics of the $U_3O_8$ powder for the verification. The experimental results show that the permeability of the $U_3O_8$ throughout mesh enhance more than old vol-oxidizer, the oxidation time takes only 8 hours when compared with the 13 hours of the old device, and the average distribution of particle size is $40{\mu}m$. The capacities of new vol-oxidizer for a 20 kg HM/batch in $UO_2$ pellets were agree well with the predictions of design model.

• KEPCO Journal on Electric Power and Energy
• /
• v.6 no.3
• /
• pp.305-313
• /
• 2020

The use of cable tunnels for electric power transmission as well as their construction in difficult conditions such as in subsea terrains and large overburden areas has increased. So, in order to efficiently operate the small diameter shield TBM (Tunnel Boring Machine), the estimation of advance rate and development of a design model is necessary. However, due to limited scope of survey and face mapping, it is very difficult to match the rock mass characteristics and TBM operational data in order to achieve their mutual relationships and to develop an advance rate model. Also, the working mechanism of previously utilized linear cutting machine is slightly different than the real excavation mechanism owing to the penetration of a number of disc cutters taking place at the same time in the rock mass in conjunction with rotation of the cutterhead. So, in order to suggest the advance rate and machine design models for small diameter TBMs, an EPB (Earth Pressure Balance) shield TBM having 3.54 m diameter cutterhead was manufactured and 19 cases of full-scale tunneling tests were performed each in 87.5 ㎥ volume of artificial rock mass. The relationships between advance rate and machine data were effectively analyzed by performing the tests in homogeneous rock mass with 70 MPa uniaxial compressive strength according to the TBM operational parameters such as thrust force and RPM of cutterhead. The utilization of the recorded penetration depth and torque values in the development of models is more accurate and realistic since they were derived through real excavation mechanism. The relationships between normal force on single disc cutter and penetration depth as well as between normal force and rolling force were suggested in this study. The prediction of advance rate and design of TBM can be performed in rock mass having 70 MPa strength using these relationships. An effort was made to improve the application of the developed model by applying the FPI (Field Penetration Index) concept which can overcome the limitation of 100% RQD (Rock Quality Designation) in artificial rock mass.

• The Korean Journal of Nuclear Medicine
• /
• v.34 no.1
• /
• pp.62-73
• /
• 2000

Purpose: Esophageal cancer patients have a difficulty in the intake of meals through the blocked esophageal lumen, which is caused by an ingrowth of cancer cells and largely influences on the prognosis. It is reported that esophageal cancer has a very low survival rate due to the lack of nourishment and immunity as the result of this. In this study a new radioactive stent, which prevents tumor ingrowth and restenosis by additional radiation treatment, has been developed. Materials and Methods: Using ${\ulcorner}HANARO{\lrcorner}$ research reactor, the radioactive stent assembly ($^{166}Ho$-SA) was prepared by covering the metallic stent with a radioactive sleeve by means of a post-irradiation and pre-irradiation methods. Results: Scanning electron microscopy and autoradiography exhibited that the distribution of $^{165/166}Ho\;(NO_3)$ compounds in polyurethane matrix was homogeneous. A geometrical model of the esophagus considering its structural properties, was developed for the computer simulation of energy deposition to the esophageal wall. The dose distributions of $^{166}Ho$-stent were calculated by means of the EGS4 code system. The sources are considered to be distributed uniformly on the surface in the form of a cylinder with a diameter of 20 mm and length of 40 mm. As an animal experiment, when radioactive stent developed in this study was inserted into the esophagus of a Mongrel dog, tissue destruction and widening of the esophageal lumen were observed. Conclusion: We have developed a new radioactive stent comprising of a radioactive tubular sleeve covering the metallic stent, which emits homogeneous radiation. If it is inserted into the blocked or narrowed lumen, it can lead to local destruction of the tumor due to irradiation effect with dilatation resulting from self-expansion of the metallic property. Accordingly, it is expected that restenosis esophageal lumen by the continuous ingrowth and infiltration of cancer after insertion of our radioactive stent will be decreased remarkably.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.1
• /
• pp.81-92
• /
• 2013

The ultimate flexural strength of HSB I-girders, considering the effect of local bucking, was investigated through a series of nonlinear finite element analysis. The girders were selected such that the inelastic local flange buckling or the plastic yielding of compression flanges governs the flexural strength. Both homogeneous sections fabricated from HSB600 or HSB800 steel and hybrid sections with HSB800 flanges and SM570-TMC web were considered. In the FE analysis, the flanges and web were modeled using thin shell elements and initial imperfections and residual stresses were imposed on the FE model. An elasto-plastic strain hardening material was used for steels. After establishing the validity of present FE analysis by comparing FE results with test results published in the literature, the effects of initial imperfection and residual stress on the inelastic flange local buckling behavior were assessed. The ultimate flexural strengths of 60 I-girders with various compression flange slenderness were obtained by FE analysis and compared with those calculated from the KHBDC, AASHTO LRFD and Eurocode 3 provisions. Based on the comparison, the applicability of design equations in these specifications for the flexural strength of I-girder considering flange local buckling was evaluated.

• The Journal of Engineering Geology
• /
• v.3 no.2
• /
• pp.149-165
• /
• 1993

Nowadays, the field hydraulic test is still an only method to evaluate groundwater characteristics in subsurface. The results of hydraulic test are very important for the concept model of fracture hydrogeology as well as the geometric pattern of fractures. The hydraulic tests performed in Korea are generally analysed under such assumption as steady radial flow in homogeneous aquifer or along simple geometry of fractures. Also the transmissivity measured in a fixed interval length is equivalent to a sum of individual fracture transmissivities in test legth. The boundary effects of weH hydraulics and the geometry of flow paths are hardly obtained from the test results analysed by a steady flow method. To circumvent this problem, the flow dimensional analysis was attempted from the results of constant pressure injection test carried out in a fractured granite area. A comparison of the hydraulic conductivity values from the transient and steady analysis shows that the latter is about a factor of 2~3 higher than the former. However, it was possible to analyse a flow dimension of each test interval from flow rate variation with time. The upper part of the bedrock(<10m deep) indicates an open boundary and the flow dimension shows nearly steady states, while the lower part of the bedrock(>25m deep) is characterized as sublinear flow dimension with a dosed boundary. In one of the test sections(15m deep), the flow dimension was changed from linear flow to spherical flow. From the experience of this study, one of the immediate problems to be solved is to enhance the field testing equipments, i.e., an accurate flowmeter with autorecording and a pressure detecting device to be able to install in the test section.

• The Journal of Korean Academy of Prosthodontics
• /
• v.34 no.1
• /
• pp.101-124
• /
• 1996

In this study, we designed the finite element models of mandible with varying their connecting types between the prosthesis on implant fixture and 2nd premolar, which were free-standing case(Mf), precision attachment case(Mp), semiprecision attachment case(Ms) and telescopic case(Mt). The basic model of the designed finite element models, which contained a canine and the 1st & 2nd premolar, was implanted in the edentulous site of the 1st & 2nd molar by two implant fixtures. We applied the load in all models by two ways. A vertical load of 200N was applied at each central fossa of 2nd premolar and 1st implant. A tilting load of 20N with inclination of $45^{\circ}$ to lingual side was applied to buccal cusp tips of each 2nd premolar and 1st implant. And then we analyzed three-dimensional finite element models, making a comparative study of principal stress and displacement in four cases respectively. Three-dimensional finite element analysis was performed for the stress distribution and the displacement using commercial software(IDEAS program) for SUN-SPARC workstation. The results were as follows : 1 Under vertical load or tilting load, maximum displacement appeared at the 2nd premolar. Semiprecision case showed the largest maximum displacement, and maximum displacement reduced in the order of precision attachment, free-standing and telescopic case. 2. Under vertical load. the pattern of displacement of the 1st implant appeared mesio-inclined because of the 2nd implant splinted together. But displacement pattern of the 2nd premolar varied according to their connection type with prosthesis. The 2nd premolar showed a little mesio-inclined vertical displacement in case of free-standing and disto-inclined vertical displacement due to attachment in case of precision and semiprecision attachment. In telescopic case, the largest mesio-inclined vertical displacement has been shown, so, the 1st premolar leaned mesial side. 3. Under tilting load, The pattern of displacement was similar in all four cases which appeared displaced to lingual side. But, the maximum displacement of 2nd premolar appeared larger than that of the first implant. Therefore, there was large discrepancy in displacement between natural tooth and implant during tilting load. 4. Under vertical load, the maximum compressive stress appeared at the 1st implant's neck. Semiprecision attachment case showed the largest maximum compressive stress, and the maximum compressive stress reduced in the order of precision attachment, telescopic and free-standing case. 5 Under vertical load, the maximum tensile stress appeared at the 2nd implant's distal neck. Semiprecision attachment case showed the largest maximum tensile stress, and the maximum tensile stress reduced in the order of precision attachment, telescopic and free-standing case. 6. Under vertical load or tilting load, principal stress appeared little between natural tooth & implant in free-standing case, but large principal stress was distributed at upper crown and distal contact site of the 2nd premolar in telescopic case. Principal stress appeared large at keyway & around keyway of distal contact site of the 2nd premolar in precision and semiprecision attachment case, appearing more broad and homogeneous in precision attachment case than in semiprecision attachment case.

• The Journal of Fisheries Business Administration
• /
• v.29 no.2
• /
• pp.77-96
• /
• 1998

Fisheries products in Korea generally go through three markets, namely the wholesale market at production site (Market A), the wholesale market at consumption site (Market B), and the retail market (Market C), from producers to end consumers. As the products move from Market A through Market B to Market C, the marginal gap of prices asked in these markets demonstrates an apparent relationship. The producers, middlemen, consumers, and governmental departments concerned may influence the marketing prices of fisheries products. This study employing the cointegration theory tries to investigate whether causality of the price-setting among these markets exists and, if any, what it is. The authors have focused their attention on fisheries markets in Pusan, analyzing the long-run equilibrium relationship and causality between the prices of hairtail and squid among markets at different levels. Data used in this study cover the period f개m August 1984 to December 1997 fer hairtail, and the period from May 1989 to December 1997 for squid. The main findings of the study may be summarized as follows: First, regardless of the price time-series of hairtail and squid in individual market, the first difference is necessary fur satisfying the stationary conditions since each time-series is a first integration. This means homogeneous integration of time-series, which is a requirement of the long-run equilibrium of prices at different markets, is satisfied. Second, the study of the long-run equilibrium relationship between the prices at Market A and Market B shows that a long-run equilibrium relationship does exist for selling prices of the two species at Market A and Market B. Third, the ECM (error correction model ) used here to describe the long- and short-run dynamics of price change demonstrates that, in the case of squid, the price change in Market A will lead to a corresponding price change in Market B in the long-run period. In the short-run, however, the price at Market H is not only influenced by the price change in Market A but influence the price at Market A as well, that is, the Prices between Market A and Market B have a feedback effect. It should be stressed that the limitation in data collection, which cover only two species of hairtail and squid, is likely to cause a sampling bias. Nonetheless, we may conclude that a dynamic relation in the formation of prices does exist in view of the transaction amount of species at different markets. It is believed that the conclusion drawn from this study would not only contribute to a long-lasted debate on the direction of causality of price-setting among academic circle and fishing community, but would provide a useful standard for the policy makers in charge of the price-setting of fisheries products as well.

• KSBB Journal
• /
• v.23 no.3
• /
• pp.245-250
• /
• 2008

The treatment of a model wastewater containing quinoline in an integrated wet oxidation-aerobic biological treatment was investigated. Partial wet oxidation under mild operating conditions was capable of converting the original quinoline to biodegradable organic acids such as nicotinic, formic and acetic acid, the solution of which was subjected to the subsequent aerobic biological treatment. The wet oxidation was carried out at 250$^{\circ}C$ and the initial pH of 7.0, and led to effluents of which nicotinic acid was oxidized through 6-hydroxynicotinic acid by a Bacillus species in the subsequent aerobic biological treatment. Either homogeneous catalyst of $CuSO_4$ or phenol, which is more degradable in the wet oxidation compared to quinoline, was also used for increasing the oxidation rate in the wet oxidation of quinoline at 200$^{\circ}C$. The oxidation of quinoline in the catalytic wet oxidation and the wet co-oxidation with phenol resulted in effluents of which nicotinic acid was biodegradable earlier in the aerobic biological treatment compared to those out of the non-catalytic wet oxidation at 250$^{\circ}C$. However, the lag phase in the biodegradation of nicotinic acid formed out of the wet oxidation at 250$^{\circ}C$ was considerably shortened after the adaptation of Bacillus species used in the aerobic biological treatment with the effluents of the quinoline wet oxidation.

• The Journal of the Korea Contents Association
• /
• v.17 no.10
• /
• pp.385-393
• /
• 2017

The purpose of this study was to develop and apply the participation-centered lecture model and practice principle developed by the Change Laboratory Method to improve the lecture interaction of nursing education and verify the effect. The experimental group was 91, and control group was 83 students who studied in the 3rd and 4th grade at one College of Nursing, This study developed the detail participatory lecture method on January and February and applied it to the students in March, and April, 2017, In the pre-survey, general characteristics, class attitude, and learning strategies were examined and homogeneity test was conducted. Post-survey was conducted with participation, satisfaction, and performance, The results showed that the experimental group and the control group were homogeneous. The average participation levels of the experimental and control groups were 3.98 and 3.77 respectively(t=2.38, p<.05). However, there was no statistically significant difference in the satisfaction and achievement. Therefore, it was expected to improve the participation and performance of students through the application of the participation-centered lecture. In the future, it will be necessary to prepare and implement effective methods for improvement of teaching methods in each subject.