• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.1
• /
• pp.93-100
• /
• 2021

Forward osmosis (FO) process is a chemical potential driven process, where highly concentrated draw solution (DS) is used to take water through semi-permeable membrane from feed solution (FS) with lower concentration. Recently, commercial FO membrane modules have been developed so that full-scale FO process can be applied to seawater desalination or water reuse. In order to design a real-scale FO plant, the performance prediction of FO membrane modules installed in the plant is essential. Especially, the flux prediction is the most important task because the amount of diluted draw solution and concentrate solution flowing out of FO modules can be expected from the flux. Through a previous study, a theoretical based FO module model to predict flux was developed. However it needs an intensive numerical calculation work and a fitting process to reflect a complex module geometry. The idea of this work is to introduce deep learning to predict flux of FO membrane modules using 116 experimental data set, which include six input variables (flow rate, pressure, and ion concentration of DS and FS) and one output variable (flux). The procedure of optimizing a deep learning model to minimize prediction error and overfitting problem was developed and tested. The optimized deep learning model (error of 3.87%) was found to predict flux better than the theoretical based FO module model (error of 10.13%) in the data set which were not used in machine learning.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.6
• /
• pp.131-137
• /
• 2007

In metal casting process, it is very difficult to predict the form accuracy of cast part and reduce repeatability error. In this study, the variations of form accuracy were measured in the process of metal cast prototyping, where RP part is manufactured from CAD model in the first, and then, wax part is cast in the vacuum environment using the RP part as master model, and finally metal prototype is cast using ceramic mold and the wax part as pattern. To investigate the variations of form accuracy, the averages and standard deviations of error distribution of the parts measured by 3D scanner were compared. It was observed that the biggest shrinkage is generated during the extraction of wax part in the second step and the biggest deterioration of form accuracy is generated during the metal part casting in the last step.

• Korean Journal of Construction Engineering and Management
• /
• v.18 no.6
• /
• pp.98-108
• /
• 2017

Recently, the development of middle and high-rise building technology using modular construction method has emphasized the importance of site construction error management. The modular construction method is very limited to adjust the construction error in the field because of the factory production characteristics. Therefore, in order to prevent the construction error in advance, a management plan reflecting the characteristics of the modular construction method is required, and it is important to make the decision of the construction participant at each stage. This study analyzed the factor of construction error of modular construction and suggested a decision support model for construction error management based on construction process. The result of this study is expected to be a guideline for the modular construction participant to derive the construction error management plan.

• Journal of Korean Society for Quality Management
• /
• v.22 no.2
• /
• pp.41-50
• /
• 1994

Consider a canning process where cans are filled with an expensive ingredient. Cans weighting above the specified limit are sold in a regular market for a fixed price, and underfilled cans are emptied and refilled at the expense of a reprocessing cost. In this paper, the effect of measurement error on the determination of the optimal process mean for a canning process is examined. It is assumed that the quantity X of ingredient in a can is normally distributed with unknown mean and known variance, and the observed value Y of X is also normally distributed with known mean and variance. A profit model is constructed which involves selling price. cost of ingredients, reprocessing cost. and cost from an accepted nonconforming can, and methods of finding the optimal process mean and the cutoff value on Y are presented. It is shown that the optimal process mean increases. and the expected profit decreases when the measurement error is relatively large in comparison to the process variance.

• Bulletin of the Korean Chemical Society
• /
• v.22 no.3
• /
• pp.251-252
• /
• 2001

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.8
• /
• pp.680-686
• /
• 2003

Digital waveguide model is a general method that is used in physical modeling of musical instruments. Wave motion is analyzed by time or by space in digital waveguide model. Because sampling is made via time, it is general that musical instrument model is described by wave motion of time. In this paper, we synthesized the musical instrument sound by adding instrument body model to the spatial based string model. In this way, we could improve sound quality and process musical instrument model's tone control variables effectively. We explained about delay error that happens in string and body in space based sampling and showed method to process fractional delay using FD (Fractional Delay)filter. Finally, we explained the relation between tone quality and number of delays. And we also compared the result with time base digital waveguide model.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.23 no.11
• /
• pp.831-836
• /
• 2010

In this paper, Plasma Enhanced Chemical Vapor Deposition (PECVD) $SiO_2$ film properties are modeled using statistical analysis and neural networks. For systemic analysis, Box-Behnken's 3 factor design of experiments (DOE) with response surface method are used. For characterization, deposited film thickness and film stress are considered as film properties and three process input factors including plasma RF power, flow rate of $N_2O$ gas, and flow rate of 5% $SiH_4$ gas contained at $N_2$ gas are considered for modeling. For film thickness characterization, regression based model showed only 0.71% of root mean squared (RMS) error. Also, for film stress model case, both regression model and neural prediction model showed acceptable RMS error. For sensitivity analysis, compare to conventional fixed mid point based analysis, proposed sensitivity analysis for entire range of interest support more process information to optimize process recipes to satisfy specific film characteristic requirements.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.12 no.5
• /
• pp.9-15
• /
• 2013

Roll-to-roll printing technology has lately become a subject of special interests in the field of printed electronics. Since this technology has the advantage that continuous and mass production is possible. And for high precision register control is required in multi-layer printing to produce the electronic devices, this is one of the most important technologies in roll-to-roll printing technology. Register error could be generated by various reasons like eccentricity of roll and thermal deformation due to temperature variation in drying section. In this study, the effect of tension variation on the register was analyzed. The results of these analyses show that it is essential to consider the tension disturbance which is generated by the change of temperature in drying section, and conventional register model has limitation to estimate the register error. In order to overcome the limitation of the register model, advanced register model based on the SI process was developed. Also, the performance of the developed model was verified experimentally.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2013.04a
• /
• pp.454-459
• /
• 2013

For the purpose of rapid development and superior design quality assurance, sophisticated finite element model for SOM(Spatial Optical Modulator) piezo actuator of MOEMS device has been developed and evaluated for the accuracy of dynamics and residual stress analysis. Parametric finite element model is constructed using ANSYS APDL language to increase the design and analysis performance. Geometric dimensions, mechanical material properties for each thin film layer are input parameters of FE model and residual stresses in all thin film layers are simulated by thermal expansion method with psedu process temperature. $6^{th}$ mask design samples are manufactured and $1^{st}$ natural frequency and 10V PZT driving displacement are measured with LDV. The results of experiment are compared with those of the simulation and validate the good agreement in $1^{st}$ natural frequency within 5% error. But large error over 30% occurred in 10V PZT driving displacement because of insufficient PZT constant $d_{31}$ measurement technology.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.327-329
• /
• 1998

Up-Propagation is an algorithm for inverting and learning neural network generative models. Sensory input is processed by inverting a model that generates patterns from hidden variables using top-down connections. The inversion process is iterative, utilizing a negative feedback loop that depends on an error signal propagated by bottom-up connections. The error signal is also used to learn the generative model from examples. the algorithm is benchmarked against principal component analysis in experiments on images of handwritten digits.