• 한국정보시스템학회지:정보시스템연구
• /
• 제27권4호
• /
• pp.115-128
• /
• 2018

Purpose The purpose of this study is to predict future behaviors of business process. Specifically, this study tried to predict the last activities of process instances. It contributes to overcoming the limitations of existing approaches that they do not accurately reflect the actual behavior of business process and it requires a lot of effort and time every time they are applied to specific processes. Design/methodology/approach This study proposed a novel approach based using deep learning in the form of dynamic recurrent neural networks. To improve the accuracy of our prediction model based on the approach, we tried to adopt the latest techniques including new initialization functions(Xavier and He initializations). The proposed approach has been verified using real-life data of a domestic small and medium-sized business. Findings According to the experiment result, our approach achieves better prediction accuracy than the latest approach based on the static recurrent neural networks. It is also proved that much less effort and time are required to predict the behavior of business processes.

• International Journal of Internet, Broadcasting and Communication
• /
• 제12권2호
• /
• pp.120-126
• /
• 2020

Studies are being conducted regarding falling recognition using sensors on smartphonesto recognize falling in human daily life. These studies use a number of sensors, mostly acceleration sensors, gyro sensors, motion sensors, etc. Falling recognition system processes the values of sensor data by using a falling recognition algorithm and classifies behavior based on thresholds. If the threshold is ambiguous, the accuracy will be reduced. To solve this problem, Deep learning was introduced in the behavioral recognition system. Deep learning is a kind of machine learning technique that computers process and categorize input data rather than processing it by man-made algorithms. Thus, in this paper, we propose a falling recognition application system using deep learning based on smartphones. The proposed system is powered by apps on smartphones. It also consists of three layers and uses DataBase as a Service (DBaaS) to handle big data and address data heterogeneity. The proposed system uses deep learning to recognize the user's behavior, it can expect higher accuracy compared to the system in the general rule base.

• Smart Structures and Systems
• /
• 제14권6호
• /
• pp.1105-1129
• /
• 2014

The unscented Kalman filter (UKF) has been developed for nonlinear model parametric identification, and it assumes that the model parameters are symmetrically distributed about their mean values without any constrains. However, the parameters in many applications are confined within certain ranges to make sense physically. In this paper, a constrained unscented Kalman filter (CUKF) algorithm is proposed to improve accuracy of numerical substructure modeling in hybrid testing. During hybrid testing, the numerical models of numerical substructures which are assumed identical to the physical substructures are updated online with the CUKF approach based on the measurement data from physical substructures. The CUKF method adopts sigma points (i.e., sample points) projecting strategy, with which the positions and weights of sigma points violating constraints are modified. The effectiveness of the proposed hybrid testing method is verified by pure numerical simulation and real-time as well as slower hybrid tests with nonlinear specimens. The results show that the new method has better accuracy compared to conventional hybrid testing with fixed numerical model and hybrid testing based on model updating with UKF.

• 한국정밀공학회지
• /
• 제21권1호
• /
• pp.181-188
• /
• 2004

Structural analysis for materials containing regularly spaced in-homogeneities is usually executed by using averaged material properties. For the homogenization process, a unit cell is defined and loaded somehow, and its response is investigated to evaluate the properties. The imposed loading conditions should accord to the behavior of unit cell immersed in the macroscopic structure in order to guarantee the accuracy of the effective properties. Each unit cell shows periodic variation of strain if the material is loaded uniformly, and in this study, direct implementation of this characteristic behavior is attempted on FE models of unit cell. Conventional finite element analysis tool can be used without any modification, and the boundary of unit cell is constrained in a way that the periodicity is satisfied. The proposed method is applicable to skew arrayed in-homogeneity problems. The flexibility matrix relating tonsorial stress and strain components in skewed rectilinear coordinate system is transformed so that the required engineering constants can be evaluated. Effective properties are computed for the materials with square and skew arrayed circular holes, and its accuracy is examined.

• 한국정밀공학회지
• /
• 제34권4호
• /
• pp.273-279
• /
• 2017

In sheet metal forming numerical analysis, the strain hardening equation has a significant effect on calculation results, especially in the field of spring-back. This study introduces the Kim-Tuan strain hardening model. This model represents sheet material behavior over the entire strain hardening range. The proposed model is compared to other well known strain hardening models using a series of uniaxial tensile tests. These tests are performed to determine the stress-strain relationship for Al6016-T4, DP980, and CP Ti sheets. In addition, the Kim-Tuan model is used to integrate the CP Ti sheet strain hardening equation in ABAQUS analysis to predict spring-back amount in a bending test. These tests highlight the improved accuracy of the proposed equation in the numerical field. Bending tests to evaluate prediction accuracy are also performed and compared with numerical analysis results.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2004년도 가을 학술발표회 논문집
• /
• pp.262-269
• /
• 2004

Ship structures are basically an assembly of plate elements and the load-carrying capacity or the ultimate strength is one of the most important criteria for safety assessment and economic design. Also, Structural elements making up ship plated structures do not work separately, resulting in high degree of redundancy and complexity, in contrast to those of steel framed structures. To enable the behavior of such structures to be analyzed, simplifications or idealizations must essentially be made considering the accuracy needed and the degree of complexity of the analysis to be used. On this study, to investigate effect of analysis range, the finite element method are used and their results are compared varying the analysis ranges. The model has been selected from bottom panels of large merchant ship structures. For FEA, three types of structural modeling are adopted in terms of the extent of the analysis. The purpose of the present study is to numerically calculate the characteristics of ultimate strength behavior according to the analysis ranges of stiffened panels subject to uniaxial compressive loads.

• The Journal of the Acoustical Society of Korea
• /
• 제15권3E호
• /
• pp.66-73
• /
• 1996

In this paper, we propose the filtered-x least mean fourth (LMF) algorithm where the error raised to the power of four is minimized and analyze its convergence behavior for a multiple sinusoidal acoustic noise and Gaussian measurement noise. Application of the filtered-x LMF adaptive filter to active noise control(ANC) requires estimating of the transfer characteristic of the acoustic path between the output and error signal of the adaptive controller. The results of the convergence analysis of the filtered-x LMF algorithm indicates that the effects of the parameter estimation inaccuracy on the convergence behavior of the algorithm are characterized by two distinct components : Phase estimation error and estimated gain. In particular, the convergence is shown to be strongly affected by the accuracy of the phase response estimate. Also, we newly show that convergence behavior can differ depending on the relative sizes of the Gaussian measurement noise and convergence constant.

• The Journal of the Acoustical Society of Korea
• /
• 제17권2E호
• /
• pp.10-15
• /
• 1998

Application of the Filtered-X LMS adaptive filter to active noise cancellation requires to estimate the transfer characteristics between the output and the error signal of the adaptive canceler. In this paper, we derive an adaptive cancellation algorithm and analyze is convergence behavior when the acoustic noise is assumed to consist of multiple sinusoids. The results of the convergence analysis of the Filtered-X LMS algorithm indicate that the effects of parameter estimation inaccuracy on the convergence behavior of the algorithm are characterize by two distinct components : Phase estimation error and estimated magnitude. In particular, the convergence of the Filtered-X LMS algorithm is show to be strongly affected by the accuracy of the phase response estimate. Simulation results of the algorithm are presented which support the theoretical convergence analysis.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.778-783
• /
• 1994

In operating automated manufacturing system, the long term stability and reliability of NC machine tools become most critical issues. Especially the machining accuracy is dominated by the thermal deformation of machine tools which remains still unsolved and causes troubles in manufacturing operations. Although researches have been carried out on the thermal behavior of a machine tools to minimize or control the thermal deformation of machine tools, the computer models for an analysis of the thermal behacior in machine tools has yet to appear in the open literature. The object of the paper is to present a method of modeling the thermal behavior of a machine tool. The method will make use of finite elements ad be capable of modeling whole machine structures as well as of heat generation processes in the kinematic system components. And temperature distributions and thermal deformations of a CNC lathe are analyzed using the finite element method and are compared with those measured in practice.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1998년도 학술발표대회 논문집 제5권
• /
• pp.80-83
• /
• 1998

In this paper, we propose a new filtered-x least mean fouth (LMF) algorithm where the error raised to the power of four is minimized and analyze its convergence behavior or a multiple sinusoidal acoustic noise and Gaussian measurement noise. Application of the filtered-x LMF adaptive filter to active noise cancellation (ANC) requires estimating of the transfer characteristic of the acoustic path between the ouput and error signal of the adaptive canceller. The results of the convergence analysis of the filtered-x LMF algorithm indicates that the effects of the parameter estimation inaccuracy on the convergence behavior of the algorithm are characterized by two distinct component . Phase estimation error and estimated again. In particular , the convergence is shown to be strongly affected by the accuracy of the phase response estimate. Also, we newly show that convergence behavior can differ depending on the relative sizes of the Gaussian measurement noise and convergence constant.