• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 추계학술대회 논문집 학회본부
• /
• pp.58-61
• /
• 1996

For a class of single-input single-output continuous-time nonlinear systems, a multilayer neural network-based controller that feedback-linearizes the system is presented. Control action is used to achieve tracking performance for a state-feedback linearizable but unknown nonlinear system. The multilayer neural network(NN) is used to approximate nonlinear continuous function to any desired degree of accuracy. The weight-update rule of multilayer neural network is derived to satisfy Lyapunov stability. It is shown that all the signals in the closed-loop system are uniformly bounded. Initialization of the network weights is straightforward.

• 전기학회논문지
• /
• 제66권1호
• /
• pp.72-81
• /
• 2017

We propose a switching control scheme for the control of a ball and beam system. It was reported in [4] that a ball and beam system is a nonlinear system which has an ill-defined relative degree. So, the traditional control approaches have been mostly either Jacobian-based control or approximate input-output linearized control in nature. In this paper, motivated by [7], we combine these two traditional control approaches and operate each controller via a pre-designed switching logic so that the improved control result can be obtained without any excessive use of control input. Switching algorithm is developed based on both analysis and actual experimental observation. We verify the effectiveness of our proposed controller via actual experimental results.

• Communications for Statistical Applications and Methods
• /
• 제30권4호
• /
• pp.403-410
• /
• 2023

Two-Phase sampling, which was first introduced by Neyman (1938), has various applications in different forms. Variance estimation for two-phase sampling has been an important research topic because conventional variance estimators used in most softwares are not working. In this paper, we considered a variance estimation for two-phase sampling in which stratified two-stage cluster sampling designs are used in both phases. By defining a conditionally unbiased estimator of an approximate variance estimator, which is calculable when all elements in the first phase sample are observed, we propose an explicit form of variance estimator of the double expanded estimator for a two-phase sample. A small simulation study shows the proposed variance estimator has a negligible bias with small variance. The suggested variance estimator is also applicable to other linear estimators of the population total or mean if appropriate residuals are defined.

• 대한토목학회논문집
• /
• 제32권1B호
• /
• pp.21-27
• /
• 2012

T. Linde가 제안한 HLLL 기법에서는 일반화된 엔트로피 함수의 도입으로 중앙파가 평가되므로 모든 파속이 초기 상태로부터 결정된다. HLLE 기법과 달리 Roe의 선형화 기법과 완전히 결별되고 HLLC 기법과 달리 정확해에 의존되지 않으므로 모태인 HLL 기법의 온전한 계승으로 볼 수 있다. 이 연구에서는 생성항이 없는 1차원 천수방정식에 농도와 관련된 보존변수를 추가한 지배방정식에 대해 총 에너지를 일반화된 엔트로피 함수로 두고 HLLL 기법을 적용하여 모형을 구성하였다. 정확해가 알려진 세 경우에 대해 모의한 결과, 1차 정확도 수치해의 한계에도 불구하고, 대체로 정확해와 잘 일치하였다. HLLL 기법은 그 외 HLL 형 기법에 비해 우수한 것으로 나타났다. 특히, 물이 빠져 바닥이 드러나는 경우에서 그 전선이 비교적 정확하게 포착되었다. 다만, 그 외 기법에 비해 계산 시간이 더 오래 걸리는 단점이 드러났다.

• Wind and Structures
• /
• 제10권4호
• /
• pp.347-366
• /
• 2007

Guyed masts subjected to turbulent winds exhibit complex vibrations featuring many vibration modes, each of which contributes to various structural responses in differing degrees. This dynamic behaviour is further complicated by nonlinear guy cable properties. While previous studies have indicated that conventional frequency domain methods can reliably reproduce load effects within the mast, the system linearization required to perform such an analysis makes it difficult to relate these results directly to corresponding guy forces. As a result, the estimation of peak load effects arising jointly from the structural action of the mast and guys, such as leg loads produced as a result of guy reactions and mast bending moments, is uncertain. A numerical study was therefore undertaken to study peak load effects in a 295 m tall guyed mast acted on by simulated turbulent wind. Responses calculated explicitly from nonlinear time domain finite element analyses were compared with approximate methods in the frequency domain for estimating peak values of selected responses, including guy tension, mast axial loads and mast leg loads. It was found that these peak dynamic load effects could be accurately estimated from frequency domain analysis results by employing simple, slightly conservative assumptions regarding the correlation of related effects.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 추계학술대회 논문집 학회본부 D
• /
• pp.765-768
• /
• 2000

Linearization of the biped dynamic equations and design of linear controller for the linearized equations are studied in this paper. The biped robot with inverted pendulum type trunk, used to stabilize the dynamic balancing of the biped robot during dynamic walking period, is modelled with 14 DOF and simulated. Despite of well defined linear control theories so far, the linear control methods was limited to the applications for a walking robot, because they have been inherently strong nonlinear properties, such as a modeling parameter uncertainties, external forces as noise, inertial and Coriolis terms by three dimensional modeling and so on. To linearize the nonlinear equations of motion of biped robot on MIMO and time varying linear equations of motion, 1st order Taylor series is used to formulate the linear equation. And a 2nd order numerical perturbation method Is used to approximate partial differential equations. Using the linearized equations of motion, a linear controller is designed by pole placement method with feed forward compensation. Using the obtained linearized equations and linear controller, the continuous walking simulation is performed.

• International Journal of Aeronautical and Space Sciences
• /
• 제11권4호
• /
• pp.326-337
• /
• 2010

This work addresses problems regarding trajectory planning for unmanned aerial vehicle sensors. Such sensors are used for taking measurements of large nonlinear systems. The sensor investigations presented here entails methods for improving estimations and predictions of large nonlinear systems. Thoroughly understanding the global system state typically requires probabilistic state estimation. Thus, in order to meet this requirement, the goal is to find trajectories such that the measurements along each trajectory minimize the expected error of the predicted state of the system. The considerable nonlinearity of the dynamics governing these systems necessitates the use of computationally costly Monte-Carlo estimation techniques, which are needed to update the state distribution over time. This computational burden renders planning to be infeasible since the search process must calculate the covariance of the posterior state estimate for each candidate path. To resolve this challenge, this work proposes to replace the computationally intensive numerical prediction process with an approximate covariance dynamics model learned using a nonlinear time-series regression. The use of autoregressive time-series featuring a regularized least squares algorithm facilitates the learning of accurate and efficient parametric models. The learned covariance dynamics are demonstrated to outperform other approximation strategies, such as linearization and partial ensemble propagation, when used for trajectory optimization, in terms of accuracy and speed, with examples of simplified weather forecasting.

• 지구물리와물리탐사
• /
• 제4권3호
• /
• pp.70-79
• /
• 2001

적분방정식법은 매우 강력한 3차원 전자탐사 모델링 기법이다. 그러나 이 방법은 이상체내의 전기장 계산시 대형 선형방정식의 해를 구해야 하므로 계산시간이 많이 소요된다는 단점이 있다. 특히 3차원 역산의 경우에는 이러한 적분방정식의 단점은 치명적이 될 수밖에 없다. 이상체내의 전기장을 1차장으로 가정하는 통상적인 Born 근사법은 계산이 용이하고 속도가 빠르다는 장점이 있다. 그러나 이 방법은 이상체와 모암간의 전기전도도비가 너무 클 경우에는 정확성에 문제가 있다. 준선형, 준해석 및 확장된 Born 근사는 이상체내의 전기장 계산을 위한 적분방정식을 선형화한 방법으로 적분방정식법에 비하여 계산시간이 빠르고 통상의 Born 근사에 비해서는 정확성이 높은 매우 훌릉한 3차원 전자탐사 모델링 기법이다. 그러나 이들 또한 근본적으로 근사법에 해당되므로 정확성을 향상시킬 필요가 있다. 근사법의 정확성을 높이기 위한 방법으로 반복적 방법을 사용하는 급수 전개법이 동원되며, 이 방법에는 수정 Born 급수, 준선형 급수 및 준해석 급수 등이 있다. 이들 급수 전개법은 적분방정식법 및 여러 근사법과 비교해 볼 때 매우 정확하고 비교적 빠르며, 항상 수렴하여 그 효율성이 높은 것으로 나타났다. 또한 급수 전개법은 전산프로그램의 작성이 용이하다는 장점도 있다. 본 연구에서는 이를 확장된 Born 급수 전개법으로 화장하여 보다 정확한 결과를 얻을 수 있었다. 따라서 확장된 Born 급수법을 포함하는 각종 급수 전개법은 향후 3차원 전자탐사 모델링 및 역산에 적용 가능한 빠르고 정확한 모델링 기법으로 기대된다.