• Journal of Korean Society for Quality Management
• /
• v.47 no.4
• /
• pp.783-793
• /
• 2019

Purpose: The purpose of this study is to apply the machine and deep learning methodology on error terms which are continuously auto-generated on the sensors with specific time period and prove the improvement effects of power generator prediction diagnosis system by comparing detection ability. Methods: The SVM(Support Vector Machine) and MLP(Multi Layer Perception) learning procedures were applied for predicting the target values and sequentially producing the error terms for confirming the detection improvement effects of suggested application. For checking the effectiveness of suggested procedures, several detection methodologies such as Cusum and EWMA were used for the comparison. Results: The statistical analysis result shows that without noticing the sequential trivial changes on current diagnosis system, suggested approach based on the error term diagnosis is sensing the changes in the very early stages. Conclusion: Using pattern of error terms as a diagnosis tool for the safety control process with SVM and MLP learning procedure, unusual symptoms could be detected earlier than current prediction system. By combining the suggested error term management methodology with current process seems to be meaningful for sustainable safety condition by early detecting the symptoms.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.9 no.5
• /
• pp.64-69
• /
• 2010

This study is the machining of fuel supply pipe used in large vessels. The fuel supply pipe of large vessels have effects to reduce engine exhaust because of common rail system and show excellent fuel efficiency so it is in the limelight as a vessel engine of next generation. At present, the shape of fuel supply pipe of common rail used for huge two-stroke & low-speed vessels is like a peanut hole so the second machining is necessary after the first machining. There is high error rate for machining and the materials waste caused by machining error is serious. Also, in this time the request for increasing the length of fuel supply pipe is suggested in the world market, it's judged that current methods will show higher error rate for machining. Therefore, the purpose of this study is to improve the machining process used originally. For that, the system controlling the process was developed as well as surface roughness and straightness which are evaluation items of fuel supply pipe were measured so that improved process can be observed in real time.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.9
• /
• pp.54-63
• /
• 1993

Quantization error, generated by the quantization process of an image, is inherent in computer vision. Because, especially in stereo vision, the quantization error in a 2-D image results in position errors in the reconstructed 3-D scene, it is necessary to analyze it mathematically. In this paper, the analysis of the probability density function (pdf) of quantization error for a line-based stereo matching scheme is presented. We show that the theoretical pdf of quantization error in the reconstructed 3-D position information has more general form than the conventional analysis for pixel-based stereo matching schemes. Computer simulation is observed to surpport the theoretical distribution.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.177-177
• /
• 2000

In this paper, we propose a two-input two-output fuzzy controller to improve the performance of transient response and to eliminate the steady state error. The outputs of this controller are the control input calculated by position-type fuzzy controller and the accumulated error scaling factor. Here, the accumulated error scaling factor is adjusted on-line by fuzzy rules according to the current trend of the controlled process. To show the usefulness of the proposed controller, it is applied to several systems that are difficult to get satisfactory response by conventional PD controllers or PI controllers.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.215-218
• /
• 2000

In this paper, a new algorithm for adaptive IIR filters based on equation error methods using the RLS algorithm is proposed. In the proposed algorithm, the concept of feedback of the scaled output error proposed by tin and Unbehauen is employed and the forgetting factor is varied in adaptation process for avoiding the accumulation of the estimation error for additive noise . The proposed algorithm has the good convergence property without the parameter estimation error under the existence of mea-surement noise.

• Journal of the Korean Mathematical Society
• /
• v.44 no.4
• /
• pp.835-844
• /
• 2007

This paper considers the problem of estimating the error distribution function in nonparametric regression models. Sufficient conditions are given under which the kernel estimator of the error distribution function based on nonparametric residuals satisfies the law of iterated logarithm.

• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.6
• /
• pp.524-532
• /
• 2002

This paper presents a new in-flight alignment method for an SDINS under large initial heading error. To handle large heading error, a new attitude error model is introduced. The attitude errors are divided into heading error and leveling errors using a newly defined horizontal frame. Some navigation error dynamic models are derived from the attitude error model for indirect feedback filtering of the in-flight alignment system. A Kalman filter with Position measurement is designed to estimate navigation errors as the indirect feedback filter Simulation results show that the proposed in-flight alignment method reduces the heading error very quickly from more than 40deg to about 5deg so as to apply a refined navigation filter. The total alignment process including leveling mode and navigation mode in addition to the proposed one allows large initial values not only in heading error but also in leveling errors.

• Transactions of Materials Processing
• /
• v.13 no.7
• /
• pp.594-599
• /
• 2004

In the sheet metal forming operations, the strain measurement of sheet panel is an essential work which provides the formability information needed in die design, process design, and product inspection. To measure efficiently complex geometry strains, the 3-dimensional automative strain measurement system, which theoretically has a high accuracy but practically has about 3～5％ strain error, is often used. For eliminating the strain error resulted in measuring the strains of formed panels using an automated strain measurement system, the position error calibration method is suggested, which computes accurate strains using the grids with accurate nodal coordinates. The accurate nodal coordinates are calculated by adding the nodal coordinates measured by the measurement system and the position error found using the multiple regression method as a function of the main error parameters obtained from the analysis of strain error in a standard cube. For the verification, the strain distributions of square and dome cups obtained from the position error calibration method are compared with those provided by the finite element analysis and ASAME.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.3
• /
• pp.174-183
• /
• 2000

Non-Parallelism the axial direction occurs during grinding process of long slender shafts. The reason for the axial error is due to elastic deformation of the components, accumulation phenomenon of the grinding and wheel wear during the grinding process. The accumulation phenomenon, the size generation mechanism and the wheel wear process during traverse grinding result in complicated process at each step on the wheel surface. The grinding system stiffness obtained from the stiffness of the center on the tailstock and the workpiece varing according to the relative position of the wheel and the workpiece. Further more, the value of wheel wear increases as the grinding process advances. The above mentioned issues make the shape generation process during traverse grinding quite complicated. This research analyzes the shape generation process in the direction of the work spindle. First, the formulation of the grinding system stiffness was conducted and the simulation analysis method of the traverse grinding was established. Also, a measuring system for assessing the dimensinal accuracy of the workpiece has been developed.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.12 no.2
• /
• pp.114-120
• /
• 2013

An EP(Electrolytic Polishing)-MAP(Magnetic Abrasive Polishing) hybrid process was applied to remove burr on the micro pattern. Micro pattern fabrication processes are combined with micro milling and EP-MAP hybrid process for deburring. Depending on the micro milling conditions which are applied, micro burrs are formed around the side and top of the pattern. The EP-MAP deburring is used to remove these burrs effectively. To optimize removal rate and form error in the EP-MAP hybrid process, a design of experiment was performed. The effect of deburring process and form error of micro pattern are evaluated via SEM images and the results of AFM.