• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.19-27
• /
• 2022

In this study, we propose a language and platform to describe and manage the MLOps(Machine Learning Operations) workflow for time series data anomaly detection. Time series data is collected in many fields, such as IoT sensors, system performance indicators, and user access. In addition, it is used in many applications such as system monitoring and anomaly detection. In order to perform prediction and anomaly detection of time series data, the MLOps platform that can quickly and flexibly apply the analyzed model to the production environment is required. Thus, we developed Python-based AI/ML Modeling Language (AMML) to easily configure and execute MLOps workflows. Python is widely used in data analysis. The proposed MLOps platform can extract and preprocess time series data from various data sources (R-DB, NoSql DB, Log File, etc.) using AMML and predict it through a deep learning model. To verify the applicability of AMML, the workflow for generating a transformer oil temperature prediction deep learning model was configured with AMML and it was confirmed that the training was performed normally.

• Journal of Power System Engineering
• /
• v.9 no.4
• /
• pp.83-89
• /
• 2005

Molding shrinkage is one of the problems to be solved in conventional injection molding. Despite many trying-out has been to solve it, intrinsic cause of shrinkage such as orientation and thermal exchange between melt and mold has not been solved yet. For reducing shrinkage and residual stress on molding, injection compression molding process was invented. In this study, experiments about effects of injection compression molding's parameters on shrinkage of PMMA molding were conducted and compared with conventional injection molding's shrinkage. Before the injection compression molding experiment, molding shrinkage rate was predicted by analyzing pvT diagram and was compared with the results of experiment. The shrinkage rate of injection compression molding was lower than convention injection molding's one which was different from the predicted shrinkage. The reason was observed that the experimental mold was not a proper type for injection compression, flowing backward of melt into nozzle and unreasonable mechanism of injection molding machine.

• Journal of Biosystems Engineering
• /
• v.26 no.2
• /
• pp.105-114
• /
• 2001

Modulus of elasticity, modulus of rigidity, damping ratio, and natural frequency of three varieties of boxthorn (Lycium chinense Mill) (Cheongyang #2, Cheongyang gugija, and Cheongyang native) branches were analyzed. Modulus of elasticity and modulus of elasticity and modulus of rigidity of the boxthorn branch was determined using standard formula after simple beam bending and torsion test, respectively, using an universal testing machine. Damping ratio and natural frequency of branches were determined using a system consisted of an accelerometer, a PC equipped with A/D converter, and a software for data analysis. Relationship between the elastic modulus and branch diameter in overall varieties and branch types showed a good correlation (r -0.81). There was, however, no correlation between torsional rigidity and branch diameter. The internal damping results were highly variable and the overall range of the damping ratio of the boxthorn branch was 0.014-0.087, which indicated that the branch was a lightly damped structure. The natural frequency of the boxthorn branch was in the range of 89-363 rad/s for the overall varieties and branch types. A good correlation (r 0.82) existed between the natural frequency and branch diameter in overall varieties and branch type.

• Restorative Dentistry and Endodontics
• /
• v.21 no.1
• /
• pp.280-288
• /
• 1996

The purpose of this study was to assess comparatively the shear bond strength on dentin of four dentin bonding agents used in conjunction with light-curing composite resins. Clearfil New Bond, Scotchbond Multipurpose Dentin Adhesive, All-Bond 2 and X-R Bond were applicated on labial dentin surfaces just below dentin - enamel juction of bovine incisor teeth. After shear bond strength testing with the universal testing machine, the bonding interface of the specimens were observed under light stereomicroscope. Following results were obtained. 1. The shear bond strength was high in the order of B,C,D,A and group B Scotchbond Multipurpose Dentine Adhesive revealed greater bond strength than Clearfil New Bond and X-R Bond. (p<.001) 2. When using ANOVA and Duncan's multiple range test, there were statistical differences among the four groups, except between group Band C,group D and A. 3. There was no relationship between mode of failure and shear bond strength.

• Advances in Energy Research
• /
• v.8 no.3
• /
• pp.175-184
• /
• 2022

The dual-fuel technology, which uses gaseous fuel as the main fuel and liquid as the pilot fuel, is an appealing technology for reducing the exhaust emissions. The current study proposes emission models based on ANFIS for a dual-fuel using producer gas (PG)-diesel engine. Emissions measurements were taken at different engine load levels and fuel injection timings. The proposed model predictions were examined using statistical methods. With R2 values in the range of 0.9903 to 0.9951, the established ANFIS model was found to be consistently robust in predicting emission characteristics. The mean absolute percentage deviate in range 1.9 to 4.6%, and mean squared error varies in range 0.0018 to 13.9%. The evaluation of the ANFIS model developed shows a reliable claim of intrinsic sensitivity, strength, and outstanding generalization. The presented meta-model can be used to simulate the engine's operation in order to create an efficient control tool.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.953-964
• /
• 2020

Ever since media contents took over the world, the importance of typography has increased, and the influence of fonts has be n recognized. Nevertheless, the current Hangul font system is very poor and is provided passively, so it is practically impossible to understand and utilize all the shape characteristics of more than six thousand Hangul fonts. In this paper, the characteristics of Hangul font shapes were selected based on the Hangul structure of similar fonts. The stroke element detection training was performed by fine tuning Faster R-CNN Inception v2, one of the deep learning object detection models. We also propose a system that automatically extracts the stroke element characteristics from characters by introducing an automatic extraction algorithm. In comparison to the previous research which showed poor accuracy while using SVM(Support Vector Machine) and Sliding Window Algorithm, the proposed system in this paper has shown the result of 10 % accuracy to properly detect and extract stroke elements from various fonts. In conclusion, if the stroke element characteristics based on the Hangul structural information extracted through the system are used for similar classification, problems such as copyright will be solved in an era when typography's competitiveness becomes stronger, and an automated process will be provided to users for more convenience.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.13 no.22
• /
• pp.51-57
• /
• 1990

A new interactive multiobjective decision making technique, which is called the fuzzy sequential proxy optimization technique, has been proposed. This technique is the revised version the sequential proxy optimization technique that the decision-maker's marginal rates of substitution is interpreted as type of L-R fuzzy numbers. It used to the square of normalized scalar product as the doptimalilry condition. However, this technique ignores the imprecise nature of a decision-maker's judgement of marginal rates of substitution. Also, it have a shortcoming that can be only applied over three objective functions. In this paper, considering the imprecise nature of a decision-maker's judgement, we presents an interactive fuzzy decision-making method on the basis of the decision-maker's MRS presented through the use of five types of membership functions including non-linear functions. FORTRAN programs that run in conversational mode are developed to implement man-machine interactive procedure.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.23 no.4
• /
• pp.269-281
• /
• 2007

Fluoric acid etching is an essential procedure in cementation of reinforced ceramics to tooth surface. But there have been few studies about the changes of surface structure and flexural strength of IPS $Empress^{(R)}$ 2 ceramic according to the etching time. The objectives of this study were to examine the surface structure changes and the difference in biaxial flexural strength of IPS $Empress^{(R)}$ 2 ceramic according to various etching times. Sixty one disk-shaped specimens of IPS $Empress^{(R)}$ 2 ceramic($14mm{\times}1.2mm$) were fabricated for the biaxial flexural strength test and SEM analysis according to the manufacturer's recommendations. Sixty specimens were divided into 6 groups(n=10) according to the time of HF acid etching(0, 20, 180 and 300s)and silane/resin cement application. Each disk was loaded using a piston-on-3 ball biaxial configuration in a universal testing machine. The failure loads(N) were recorded, and the biaxial flexural strength for each disk was calculated. A one-way analysis of variance and independent t-test on transformed fracture strength data were used to determine significant differences between groups. The groups of no cementation showed a trend toward progressive weakening with increasing the etching time. However, this was not statistically significant at p=0.05 level. The groups of resin cementation exhibited no apparent trend in their mean strength values. SEM photomicrographs showed very different results of etching. Within the conditions of this study, alteration of surface topography by acid etching does not have a deleterious effect on the biaxial flexural strength of IPS $Empress^{(R)}$ 2 ceramic.

• Structural Engineering and Mechanics
• /
• v.49 no.5
• /
• pp.645-660
• /
• 2014

This paper presents the design and project of an innovative concept for a Single Point Incremental Forming (SPIF) Machine. Nowadays, equipment currently available for conducting SPIF result mostly from the adaptation of conventional CNC machine tools that results in a limited range of applications in terms of materials and geometries. There is also a limited market supply of equipment dedicated to Incremental Sheet Forming (ISF), that are costly considering low batches, making it unattractive for industry. Other factors impairing a quicker spread of SPIF are large forming times and poor geometrical accuracy of parts. The following sections will depict the development of a new equipment, designed to overcome some of the limitations of machines currently used, allowing the development of a sounding basis for further studies on the particular features of this process. The equipment here described possesses six-degrees-of freedom for the tool, for the sake of improved flexibility in terms of achievable tool-paths and an extra stiffness provided by a parallel kinematics scheme. A brief state of the art about the existing SPIF machines is provided to support the project's guidelines.

• Journal of Digital Convergence
• /
• v.18 no.8
• /
• pp.231-242
• /
• 2020

The thesis presents a system that continuously collects the human body's physiological vital information at rest with sensors and ICT information technology and predicts diabetes using the collected information. it shows the artificial neural network machine learning method and essential basic variable values. The study method analyzed the correlation between heart rate measurements of BCG and ECG sensors in 20 DM- and 15 DM+ subjects. Artificial Neural Network (ANN) machine learning program was used to predictability of diabetes. The input variables are time domain information of HRV, heart rate, heart rate variability, respiration rate, stroke volume, minimum blood pressure, highest blood pressure, age, and sex. ANN machine learning prediction accuracy is 99.53%. Thesis needs continuous research such as diabetic prediction model by BMI information, predicting cardiac dysfunction, and sleep disorder analysis model using ANN machine learning.