• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 한국경영과학회 2006년도 추계학술대회
• /
• pp.373-376
• /
• 2006

In general, there are three kinds of methods in analyzing dynamic robust design experiment: loss model approach, response function approach, and response model approach. In this talk, we review the three modeling approaches in terms of several criteria in comparison. This talk also generalizes the response model approach based on a generalized linear model. We develop a generalized two-step optimization procedure to substantially reduce the process variance by dampening the effect of both explicit and hidden noise variables. The proposed method provides more reliable results through iterative modeling of the residuals from the fitted response model. The method is compared with three existing approaches in practical examples.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.988-991
• /
• 1996

The purpose of this thesis is to develop methods of designing robust LQR/LQG controllers for time-varying systems with real parametric uncertainties. Controller design that meet desired performance and robust specifications is one of the most important unsolved problems in control engineering. We propose a new framework to solve these problems using Linear Matrix Inequalities (LMls) which have gained much attention in recent years, for their computational tractability and usefulness in control engineering. In Robust LQR case, the formulation of LMI based problem is straightforward and we can say that the obtained solution is the global optimum because the transformed problem is convex. In Robust LQG case, the formulation is difficult because the objective function and constraint are all nonlinear, therefore these are not treatable directly by LMI. We propose a sequential solving method which consist of a block-diagonal approach and a full-block approach. Block-diagonal approach gives a conservative solution and it is used as a initial guess for a full-block approach. In full-block approach two LMIs are solved sequentially in iterative manner. Because this algorithm must be solved iteratively, the obtained solution may not be globally optimal.

• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1301-1306
• /
• 2018

Artificial intelligence is one of the key technologies of the Fourth Industrial Revolution. This paper introduces the diverse kinds of approaches to subjects that tackle diverse kinds of research fields such as model-based MS approach, deep neural network model, image edge detection approach, cross-layer optimization model, LSSVM approach, screen design approach, CPU-GPU hybrid approach and so on. The research on Superintelligence and superconnection for IoT and big data is also described such as 'superintelligence-based systems and infrastructures', 'superconnection-based IoT and big data systems', 'analysis of IoT-based data and big data', 'infrastructure design for IoT and big data', 'artificial intelligence applications', and 'superconnection-based IoT devices'.

• Journal of the Korean Institute of Intelligent Systems
• /
• 제11권6호
• /
• pp.477-486
• /
• 2001

In this paper , we study the problem of designing TS(Takagi-Sugeno) fuzzy controllers for the systems that can be approximated or represented by the TS fuzzy model. The main strategy used in this paper is the inverse optimal approach, in which the cost function is determined later than the Lyapunov function and its corresponding control input satisfying the design requirements such as stability, decay rate, and robustness against uncertainty. This approach is useful because it yields controllers satisfying the inherent robustness of optimal controllers as well as the considered design goals. The design procedures established in this paper are all in the from of solving LMIs(Iinear matrix inequalities). Since the LMIs arising in the design procedures can be solved within a given tolerance by the interior point methods. the design method of the paper are efficient in practice. The applicability of the proposed design procedures is demonstrated by design examples.

• Journal of the Ergonomics Society of Korea
• /
• 제32권1호
• /
• pp.59-68
• /
• 2013

Objective: The purpose of this paper is to introduce a task complexity model combining task design aspects and complexity dimensions and to explain an approach to identifying and organizing task complexity factors based on the model. Background: Task complexity is a critical concept in describing and predicting human performance in complex systems such as nuclear power plants(NPPs). In order to understand the nature of task complexity, task complexity factors need to be identified and organized in a systematic manner. Although several methods have been suggested for identifying and organizing task complexity factors, it is rare to find an analytical approach based on a theoretically sound model. Method: This study regarded a task as a system to be designed. Three levels of design ion, which are functional, behavioral, and structural level of a task, characterize the design aspects of a task. The behavioral aspect is further classified into five cognitive processing activity types(information collection, information analysis, decision and action selection, action implementation, and action feedback). The complexity dimensions describe a task complexity from different perspectives that are size, variety, and order/organization. Combining the design aspects and complexity dimensions of a task, we developed a model from which meaningful task complexity factors can be identified and organized in an analytic way. Results: A model consisting of two facets, each of which is respectively concerned with design aspects and complexity dimensions, were proposed. Additionally, twenty-one task complexity factors were identified and organized based on the model. Conclusion: The model and approach introduced in this paper can be effectively used for examining human performance and human-system interface design issues in NPPs. Application: The model and approach introduced in this paper could be used for several human factors problems, including task allocation and design of information aiding, in NPPs and extended to other types of complex systems such as air traffic control systems as well.

• Journal of Korea Water Resources Association
• /
• 제51권9호
• /
• pp.769-782
• /
• 2018

Estimation of design floods is typically required for hydrologic design purpose. Design floods are routinely estimated for water resources planning, safety and risk of the existing water-related structures. However, the hydrologic data, especially streamflow data for the design purposes in South Korea are still very limited, and additionally the length of streamflow data is relatively short compared to the rainfall data. Therefore, this study collected a large number design flood data and watershed characteristics (e.g. area, slope and altitude) from the national river database. We further explored to formulate a scaling approach for the estimation of design flood, which is a function of the watershed characteristics. Then, this study adopted a Hierarchical Bayesian model for evaluating both parameters and their uncertainties in the regionalization approach, which models the hydrologic response of ungauged basins using regression relationships between watershed structure and model. The proposed modeling framework was validated through ungauged watersheds. The proposed approach have better performance in terms of correlation coefficient than the existing approach which is solely based on area as a predictor. Moreover, the proposed approach can provide uncertainty associated with the model parameters to better characterize design floods at ungauged watersheds.

• Journal of Power Electronics
• /
• 제9권5호
• /
• pp.757-771
• /
• 2009

This paper proposes a systematic approach to the modeling of the small-signal characteristics of three-phase bridge boost rectifiers under non-sinusoidal conditions. The main obstacle to the conventional synchronous d-q frame modeling approach is that it is unable to identify a steady-state under non-sinusoidal conditions. However, for most applications under non-sinusoidal conditions, the current loops of boost rectifiers are designed to have a bandwidth that is much higher than typical harmonics frequencies in order to achieve good current control for these harmonic components. Therefore a quasi-static method is applied to the proposed modeling approach. The converter small-signal characteristics developed from conventional synchronous frame modeling under different operating points are investigated and a worst case point is then located for the current loop design. Both qualitative and quantitative analyses are presented. It is observed that operating points influence the converter low frequency characteristics but hardly affect the dominant poles. The relationship between power stage parameters, system poles and zeroes is also presented which offers good support for the system design. Both the simulation and experimental results verified the analysis and proposed modeling approach. Finally, the practical case of a parallel active power filter is studied to present the modeling approach and the resultant regulator design procedure. The system performance further verifies the whole analysis.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• 제19권5호
• /
• pp.589-595
• /
• 2010

This paper presents a reliability-based shape optimization (RBSO) using the evolutionary structural optimization (ESO). An actual design involves uncertain conditions such as material property, operational load, poisson's ratio and dimensional variation. The deterministic optimization (DO) is obtained without considering of uncertainties related to the uncertainty parameters. However, the RBSO can consider the uncertainty variables because it has the probabilistic constraints. In order to determine whether the probabilistic constraint is satisfied or not, simulation techniques and approximation methods are developed. In this paper, the reliability-based shape design optimization method is proposed by utilization the reliability index approach (RIA), performance measure approach (PMA), single-loop single-vector (SLSV), adaptive-loop (ADL) are adopted to evaluate the probabilistic constraint. In order to apply the ESO method to the RBSO, a sensitivity number is defined as the change of strain energy in the displacement constraint. Numerical examples are presented to compare the DO with the RBSO. The results of design example show that the RBSO model is more reliable than deterministic optimization.

• Journal of Korean Institute of Industrial Engineers
• /
• 제39권5호
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• Korean Management Science Review
• /
• 제29권3호
• /
• pp.81-89
• /
• 2012

Since introduced by Vining and Myers in 1990, the concept of dual response approach based on response surface methodology has widely been investigated and adopted for the purpose of robust design. Separately estimating mean and variance responses, dual response approach may take advantages of optimization modeling for finding optimum settings of input factors. Explicitly assuming functional relationship between responses and input factors, however, it may not work well enough especially when the behavior of responses are poorly represented. A sufficient number of experimentations are required to improve the precision of estimations. This study proposes an alternative to dual response approach in which additional experiments are not required. An artificial neural network has been applied to model relationships between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Training, validating, and testing a neural network with empirical process data, an artificial data based on the neural network may be generated and used to estimate response functions without performing real experimentations. A drug formulation example from pharmaceutical industry has been investigated to demonstrate the procedures and applicability of the proposed approach.