• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.64-69
• /
• 1993

In this paper we present the performance bounds of the optimal FIR filter in continuous time systems with modeling uncertainty. The performance measure bounds are calculated from the estimation error covariance bounds of the optimal FIR filter and the suboptimal FIR filter. Performance error bounds range are expressed by the upper bounds on the estimation error covariance difference between the real and nominal values in case of the systems with noise uncertainty or model uncertainty. The performance bounds of the systems are derived on the assumption that the system uncertainty and the estimation error covariance are imperfectly known a priori. The estimation error bounds of the optimal FIR filter is compared with those of the Kalman filter via a numerical example applied to the estimation of the motion of an aircraft carrier at sea, which shows the former has better performances than the latter.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2008년도 학술대회
• /
• pp.79-83
• /
• 2008

An approach to CFD code validation is developed that gives proper consideration to experimental and simulation uncertainties. The comparison errors include the difference between the data, simulation values and represents the combination of all errors. The uncertainties of modeling and numerical analysis in the CFD prediction were estimated by a Coleman's theory. In this paper, the numerical solutions are calculated by A-type standard uncertainty and Richardson extrapolation Method.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2008년 추계학술대회논문집
• /
• pp.79-83
• /
• 2008

An approach to CFD code validation is developed that gives proper consideration to experimental and simulation uncertainties. The comparison errors include the difference between the data, simulation values and represents the combination of all errors. The uncertainties of modeling and numerical analysis in the CFD prediction were estimated by a Coleman's theory. In this paper, the numerical solutions are calculated by A-type standard uncertainty and Richardson extrapolation Method.

• 스마트미디어저널
• /
• 제11권9호
• /
• pp.9-20
• /
• 2022

심층강화학습은 자율주행 도메인에서 널리 사용되는 end-to-end 데이터 기반 제어 방법이다. 그러나 기존의 강화학습 접근 방식은 자율주행 과제에 적용하기에는 비효율성, 불안정성, 불확실성 등의 문제로 어려움이 존재한다. 이러한 문제들은 자율주행 도메인에서 중요하게 작용한다. 최근의 연구들은 이런 문제를 해결하고자 많은 시도가 이루어지고 있지만 계산 비용이 많고 특별한 가정에 의존한다. 본 논문에서는 자율주행 도메인에 불확실성 순차 모델링이라는 방법을 도입하여 비효율성, 불안정성, 불확실성을 모두 고려한 새로운 알고리즘 MCDT를 제안한다. 강화학습을 높은 보상을 얻기 위한 의사 결정 생성 문제로 바라보는 순차 모델링 방식은 기존 연구의 단점을 회피하고 효율성과 안정성을 보장하며, 여기에 불확실성 추정 기법을 융합해 안전성까지 고려한다. 제안 방법은 OpenAI Gym CarRacing 환경을 통해 실험하였고 실험 결과는 MCDT 알고리즘이 기존의 강화학습 방법에 비해 효율적이고 안정적이며 안전한 성능을 내는 것을 보인다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.251-257
• /
• 2007

For the modern aircraft, uncertainty has bee an important issue to its aeroelastic stability. Therefore, many researches have been conducted regarding this topic. The uncertainties in the aeroelastic system amy consist of the structural and aerodynamic uncertainty. In this paper, we suggest a parametric uncertainty modeling and conduct the aeroelastic stability analysis of a typical wing including the uncertainty.

• 한국화재소방학회논문지
• /
• 제26권2호
• /
• pp.40-52
• /
• 2012

본 논문에서는 원자력발전소의 모터제어반 스위치기어실 화재 모델링에 대한 입력변수 불확실성 분석을 수행하였다. 화재모델링은 FDS 5.5를 사용하였고 FDS 입력변수 램던 샘플링은 라틴하이퍼쿠브 몬테칼로 방법을 이용하였다. 본 연구에서 수행한 입력변수 불확실성 분석 결과를 비교하기 위해 NUREG-1934의 화재모델링 결정론적 불확실성 분석과 민감도 분석 방법을 이용한 분석도 수행하였다. 분석결과, 본 연구의 모터제어반 스위치 기어룸 화재 모델링에 대한 입력변수 불확실성 분석방법이 NUREG-1934의 방법보다 보수적인 결과를 얻을 수 있음을 확인하였다.

• Earthquakes and Structures
• /
• 제12권2호
• /
• pp.165-175
• /
• 2017

This paper studies soil properties uncertainty and its implementation in the seismic response evaluation of structures. For this, response sensitivity of two 4- and 12-story RC shear walls to the soil properties uncertainty by considering soil structure interaction (SSI) effects is investigated. Beam on Nonlinear Winkler Foundation (BNWF) model is used for shallow foundation modeling and the uncertainty of soil properties is expanded to the foundation stiffness and strength parameters variability. Monte Carlo (MC) simulation technique is employed for probabilistic evaluations. By investigating the probabilistic evaluation results it's observed that as the soil and foundation become stiffer, the soil uncertainty is found to be less important in influencing the response variability. On the other hand, the soil uncertainty becomes more important as the foundation-structure system is expected to experience nonlinear behavior to more sever degree. Since full This paper studies soil properties uncertainty and its implementation in the seismic response evaluation of structures. For this, response sensitivity of two 4- and 12-story RC shear walls to the soil properties uncertainty by considering soil structure interaction (SSI) effects is investigated. Beam on Nonlinear Winkler Foundation (BNWF) model is used for shallow foundation modeling and the uncertainty of soil properties is expanded to the foundation stiffness and strength parameters variability. Monte Carlo (MC) simulation technique is employed for probabilistic evaluations. By investigating the probabilistic evaluation results it's observed that as the soil and foundation become stiffer, the soil uncertainty is found to be less important in influencing the response variability. On the other hand, the soil uncertainty becomes more important as the foundation-structure system is expected to experience nonlinear behavior to more sever degree. Since full probabilistic analysis methods like MC commonly are very time consuming, the feasibility of simple approximate methods' application including First Order Second Moment (FOSM) method and ASCE41 proposed approach for the soil uncertainty considerations is investigated. By comparing the results of the approximate methods with the results obtained from MC, it's observed that the results of both FOSM and ASCE41 methods are in good agreement with the results of MC simulation technique and they show acceptable accuracy in predicting the response variability.

• 한국복합재료학회:학술대회논문집
• /
• 한국복합재료학회 2001년도 추계학술발표대회 논문집
• /
• pp.70-73
• /
• 2001

The layup optimization by genetic algorithm (GA) for the interlaminar strength of laminated composites with free edge is presented. For the calculation of interlaminar stresses of composite laminates with free edges, extended Kantorovich method is applied. In the formulation of GA, repair strategy is adopted for the satisfaction of given constraints. In order to consider the bounded uncertainty of material properties, convex modeling is used. Results of GA optimization with scattered properties are compared with those of optimization with nominal properties. The GA combined with convex modeling can work as a practical tool for maximum interlaminar strength design of laminated composite structures, since uncertainties are always encountered in composite materials and the optimal results can be changed.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.976-979
• /
• 1996

Teleoperation is the extension of a person's sensing and manipulation capability to a remote location. Teleoperators generally can be modeled as linear transfer function indecently including modeling uncertainty. Modeling uncertainties can make the system unstable and its performance poor. Thus I'm studying about a design framework for a bilateral controller of teleoperator systems with modeling uncertainties. In this paper, a method based on the H$_{\infty}$-optimal control and .mu.-synthesis frameworks are introduced to design a controller for the teleoperator that achieves stability and performance in the presence of the modeling uncertainties..ties.inties.

• 대한전기학회논문지
• /
• 제43권5호
• /
• pp.838-847
• /
• 1994

Although there is an analytical proof of modeling capability of the neural network, the convergency error in nonlinearity modeling is inevitable, since the steepest descent based practical larning algorithms do not guarantee the convergency of modeling error. Therefore, it is difficult to apply the neural network to control system in critical environments under an on-line learning scheme. Although the convergency of modeling error of a neural network is not guatranteed in the practical learning algorithms, the convergency, or boundedness of tracking error of the control system can be achieved if a proper feedback control law is combined with the neural network model to solve the problem of modeling error. In this paper, the neural network is introduced for compensating a system uncertainty to control a nonlinear dynamic system. And for suppressing inevitable modeling error of the neural network, an iterative neural network learning control algorithm is proposed as a virtual on-line realization of the Adaptive Variable Structure Controller. The efficiency of the proposed control scheme is verified from computer simulation on dynamics control of a 2 link robot manipulator.