• 제목/요약/키워드: robust model

검색결과 2,726건 처리시간 0.032초

Usage of auxiliary variable and neural network in doubly robust estimation

  • Park, Hyeonah;Park, Wonjun
    • Journal of the Korean Data and Information Science Society
    • /
    • 제24권3호
    • /
    • pp.659-667
    • /
    • 2013
  • If the regression model or the propensity model is correct, the unbiasedness of the estimator using doubly robust imputation can be guaranteed. Using a neural network instead of a logistic regression model for the propensity model, the estimators using doubly robust imputation are approximately unbiased even though both assumed models fail. We also propose a doubly robust estimator of ratio form using population information of an auxiliary variable. We prove some properties of proposed theory by restricted simulations.

전차량의 능동 현가 장치 제어를 위한 중복 분산형 견실 고유구조지정 제어기 설계 (Overlapping Decentralized Robust EA Control Design for an Active Suspension System of a Full Car Model)

  • 정용하;최재원;김영호
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 2000년도 제15차 학술회의논문집
    • /
    • pp.217-217
    • /
    • 2000
  • A decentralized robust EA(eigensoucture assignment) controller is designed for an active suspension system of a vehicle based on a full car model with 7-degree of freedom. Using overlapping decomposition, the full car model is decentralized by two half car models. For each half car model, a robust eigenstructure assignment controller can be obtained by using optimization approach. The performance of the decentralized robust EA controller is compared with that of a conventional centralized EA controller through computer simulations.

  • PDF

입방형 영역에서의 G-효율이 높은 Model-Robust 실험설계 (Model-Robust G-Efficient Cuboidal Experimental Designs)

  • 박유진;이윤주
    • 산업공학
    • /
    • 제23권2호
    • /
    • pp.118-125
    • /
    • 2010
  • The determination of a regression model is important in using statistical designs of experiments. Generally, the exact regression model is not known, and experimenters suppose that a certain model form will be fit. Then an experimental design suitable for that predetermined model form is selected and the experiment is conducted. However, the initially chosen regression model may not be correct, and this can result in undesirable statistical properties. We develop model-robust experimental designs that have stable prediction variance for a family of candidate regression models over a cuboidal region by using genetic algorithms and the desirability function method. We then compare the stability of prediction variance of model-robust experimental designs with those of the 3-level face centered cube. These model-robust experimental designs have moderately high G-efficiencies for all candidate models that the experimenter may potentially wish to fit, and outperform the cuboidal design for the second-order model. The G-efficiencies are provided for the model-robust experimental designs and the face centered cube.

감소하는 비용함수를 가진 Robust EOQ 모형 (Robust EOQ Models with Decreasing Cost Functions)

  • 임성묵
    • 한국경영과학회지
    • /
    • 제32권2호
    • /
    • pp.99-107
    • /
    • 2007
  • We consider (worst-case) robust optimization versions of the Economic Order Quantity (EOQ) model with decreasing cost functions. Two variants of the EOQ model are discussed, in which the purchasing costs are decreasing power functions in either the order quantity or demand rate. We develop the corresponding worst-case robust optimization models of the two variants, where the parameters in the purchasing cost function of each model are uncertain but known to lie in an ellipsoid. For the robust EOQ model with the purchasing cost being a decreasing function of the demand rate, we derive the analytical optimal solution. For the robust EOQ model with the purchasing cost being a decreasing function of the order quantity, we prove that it is a convex optimization problem, and thus lends itself to efficient numerical algorithms.

Observer-based Feedback Controller Design for Robust Tracking of Discrete-time Polytopic Uncertain LTI Systems

  • Oh, Sangrok;Kim, Jung-Su;Shim, Hyungbo
    • Journal of Electrical Engineering and Technology
    • /
    • 제10권6호
    • /
    • pp.2427-2433
    • /
    • 2015
  • This paper presents an observer-based robust controller for constant reference tracking of linear time invariant systems with polytopic model uncertainties. To this end, this paper not only designs a robust integral controller gain but also suggests how to determine the robust observer gain and the observer model used in the observer. Since the observer model selection is not obvious due to the polytopic uncertainties, particular attention needs to be paid to that. This paper computes the robust controller and observer gains first. Then, the observer model is selected in a way that the whole closedloop is stable and LMIs are used in the middle of choosing the gains and observer model. Simulation examples show that the proposed observer-based feedback control successfully achieves robust reference tracking.

관측기 모델 선정을 통한 모델 불확실성을 갖는 선형 시불변 시스템 강인 안정화 (Robust Stabilization of Uncertain LTI Systems via Observer Model Selection)

  • 오상록;김정수;심형보
    • 제어로봇시스템학회논문지
    • /
    • 제20권8호
    • /
    • pp.822-827
    • /
    • 2014
  • This paper presents a robust observer-based output feedback control for stabilization of linear time invariant systems with polytopic uncertainties. To this end, this paper not only finds a robust observer gain but also suggests how to determine the model used in the observer, which is not obvious due to model uncertainties in the conventional observer design method. The robust observer gain and the observer model are selected in a way that the whole closed-loop is stable by solving LMIs and BMIs (Linear Matrix Inequalities and Bilinear Matrix Inequalities). A simulation example shows that the proposed robust observer-based output feedback control successfully leads to closed-loop stability.

이상 트래픽 탐지를 위한 로버스트 추정 방법 비교 연구 (A Comparative Study of a Robust Estimate Method for Abnormal Traffic Detection)

  • 정재윤;김삼용
    • Communications for Statistical Applications and Methods
    • /
    • 제18권4호
    • /
    • pp.517-525
    • /
    • 2011
  • 본 연구는 이상치가 존재하는 자료에 적용될 수 있는 방법을 비교한 연구로서, 이분산 시계열 모형 하에서 로버스트 추정 방법의 효용성을 보이고자 한다. GARCH 모형하에서 이상치 탐지 기법과 GARCH 모형을기반한 로버스트 추정방법의 성능을 비교하였다. 실제 인터넷 트래픽 자료에 두 방법을 적용했을때, 로버스트 추정방법이 이상치 탐지 기법에 비해 덜 복잡하고 성능이 우수함을 입증하였다.

CONFIGYRATION OF A ROBUST MODEL FOLLOWING SYSTEM WITH AN ADAPTIVE IDENTFIER

  • Saito, Tomoaki;Kimura, Mitsuyoshi;Kikuta, Akira;Kamiya, Yuji
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 1994년도 Proceedings of the Korea Automatic Control Conference, 9th (KACC) ; Taejeon, Korea; 17-20 Oct. 1994
    • /
    • pp.548-552
    • /
    • 1994
  • The robust compensation controller, which has been proposed by one of the authors and is based on the fundamental principle of making the plant follow the reference model, consists of the reference model and the robust compensator. The reference model is constructed by using the nominal model of the plant and determines the input-output properties of the resultant system. The robust compensator is obtained as a solution of the mixed sensitivity problem in H infinity control theory. Therefore the resultant system is of low sensitivity and robust stability. In the case where uncertainty does not occur in the plant, the plant follows perfectly the reference model. Therefore, in the case where uncertainty occurs in the plant, we propose the system configuration which improves the following accuracy without replacing the 개bust compensator but by identifying, the plant and reconstructing the reference model.

  • PDF

강인 H₂필터를 이용한 속도정합 알고리즘 (Velocity Matching Algorithm Using Robust H₂Filter)

  • 양철관;심덕선;박찬국
    • 제어로봇시스템학회논문지
    • /
    • 제7권4호
    • /
    • pp.363-363
    • /
    • 2001
  • We study on the velocity matching algorithm for transfer alignment of inertial navigation system(INS) using a robust H₂ filter. We suggest an uncertainty model and a discrete robust H₂filter for INS and apply the suggested robust H₂ filter to the uncertainty model. The discrete robust H₂filter is shown by simulation to have better performance time and accuracy than Kalman filter.

혼합 $H_{\infty}$ 최적화 기법을 이용한 견실 $H_{\infty}$ 증기발생기 수위제어기 설계 (Robust $H_{\infty}$ Controller Design for Steam Generator Water Level Control using Mixed $H_{\infty}$ Optimization Method)

  • 서성환;조희수;박홍배
    • 제어로봇시스템학회논문지
    • /
    • 제5권3호
    • /
    • pp.363-369
    • /
    • 1999
  • In this paper, we design the robust $H_{\infty}$ controller for water level control of steam generator using a mixed $H_{\infty}$ optimization with model-matching method. Firstly we choose the desired model which has good disturbance rejection performance. Secondly we design a stabilizing controller to keep the model-matching error small and also provide sufficiently large stability margin against additive perturbations of the nominal plant. Simulation results show that proposed robust $H_{\infty}$ controller at specific power operation has satisfactory performances against the variations of load power, steam flow rate, primary circuit coolant temperature, and feedwater temperature. It can be also observed that the proposed robust $H_{\infty}$ controller exhibits better robust stability than conventional PI controller.

  • PDF