• 제어로봇시스템학회논문지
• /
• 제19권6호
• /
• pp.489-494
• /
• 2013

This paper tackles the problem of designing a dynamic output-feedback control for linear discrete-time norm-bounded uncertain systems with input saturation. By employing a LPV (Linear Parameter Varying) instead of LTI (Linear Time-Invariant) control, the useful information on interpolation parameters appearing in the procedure of representing saturation nonlinearity as a convex polytope is additionally applied in the control design procedure. By solving the addressed problem that can be recast into a convex optimization problem characterized by LMIs (Linear Matrix Inequalities) with one prescribed scalar, the vertices of convex set containing an LPV output-feedback control gain and the associated maximal invariant set of initial states are simultaneously obtained.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1995년도 하계학술대회 논문집 B
• /
• pp.710-713
• /
• 1995

Hopfield networks have been applied to the problem of linear system identification. In this paper, Hopfield network based parameter identification scheme of non-linear dynamic systems is proposed. Simulation results demonstrate that Hopfield network can be used effectively for the identification of non-linear systems assuming that the system states and their time derivatives are available. Therefore, the proposed scheme can be applied in fault detection and isolation(FDI) and adaptive control of non-linear systems where the Hopfield networks perform on-line identification of system parameters.

• 한국지능시스템학회논문지
• /
• 제14권5호
• /
• pp.571-576
• /
• 2004

본 논문은 시간지연을 갖는 비선형 시스템에 대한 지연 종속 $H_{\infty}$ 제어기 설계 방법을 제안한다. 지연 종속 Lyapunov 함수를 이용하여 지수 함수적 안정화와 $H_{\infty}$ 성능 문제를 고려한다. 제어기의 존재성에 대한 충분조건을 유도하고 선형행렬부등식(LMI: linear matrix inequality)으로 나타낸다. 제어기 설계는 병렬 분산 보상의 개념을 이용하고, 필터는 LMI 해를 구함으로써 바로 구할 수 있다. 제안한 방법의 설계과정 및 타당성을 시뮬레이션 예제를 통하여 나타낸다. 지연종속 제어기는 제어기 존재 조건을 나타내는 선형 행렬 부등식에 시간지연항의 크기를 포함하고 있으므로 시간지연항의 크기에 상관없이 시스템을 안정화시키는 기존의 지연 독립적인 제어기보다 더 효과적인 설계 방법이다.

• 제어로봇시스템학회논문지
• /
• 제9권6호
• /
• pp.413-419
• /
• 2003

The objective of this paper is to survey and put in perspective the existing methods of dynamic filter development. This includes theories and practices for linear and nonlinear filters, multiple model filters, and data association methods for tracking in multitarget environment. The presentation of this paper is motivated by recent surge of interest in the area of designing feedback control systems with reduced number of sensors, detection and identification of abrupt changes, and multitarget tracking in clutter. It is hoped to be useful in view of the need to take a grasp of existing techniques before using them in practice and developing new techniques.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1989년도 하계종합학술대회 논문집
• /
• pp.48-53
• /
• 1989

We attempt to achieve both high dynamic performance and maximal power efficiency by means of linear decoupling of rotor speed (or motor torque) and rotor flux. The induction motor with our controller possesses the input-output dynamic characteristics of a linear system such that the rotor speed (or motor torque) and the rotor flux are decoupled. The rotor speed (or motor torque) responses are not affected by abrupt changes in the rotor flux and vice versa. The rotor flux need not be measured but is estimated by the well-known flux simulator. The effect of large variation in the rotor resistance on the control performances is minimized by employing a parameter adaptation method. To illuminate the significance of our work. we present simulation and experimental results as well as mathematical performance analyses.

• Computers and Concrete
• /
• 제23권2호
• /
• pp.81-95
• /
• 2019

The unexpected seismic interaction of dry-assembled precast concrete frame structures typical of the European heritage with their precast cladding panels brought to extensive failures of the panels during recent earthquakes due to the inadequateness of their connection systems. Following this recognition, an experimental campaign of cyclic and pseudo-dynamic tests has been performed at ELSA laboratory of the Joint Research Centre of the European Commission on a full-scale prototype of precast structure with vertical and horizontal cladding panels within the framework of the Safecladding project. The panels were connected to the frame structure by means of innovative arrangements of fastening systems including isostatic, integrated and dissipative. Many of the investigated configurations involved a strong frame-cladding interaction, modifying the structural behaviour of the frame turning it into highly non-linear since small deformation. In such cases, properly modelling the connections becomes fundamental in the framework of a design by non-linear dynamic analysis. This paper presents the peculiarities of the numerical models of precast frame structures equipped with the various cladding connection systems which have been set to predict and simulate the experimental results from pseudo-dynamic tests. The comparison allows to validate the structural models and to derive recommendations for a proper modelling of the different types of existing and innovative cladding connection systems.

• Journal of Mechanical Science and Technology
• /
• 제18권11호
• /
• pp.1891-1899
• /
• 2004

The characteristics of dynamic systems subjected to multiple linear constraints are determined by considering the constrained effects. Although there have been many researches to investigate the dynamic characteristics of constrained systems, most of them depend on numerical analysis like Lagrange multipliers method. In 1992, Udwadia and Kalaba presented an explicit form to describe the motion for constrained discrete systems. Starting from the method, this study determines the dynamic characteristics of the systems to have positive semidefinite mass matrix and the continuous systems. And this study presents a closed form to calculate frequency response matrix for constrained systems subjected to harmonic forces. The proposed methods that do not depend on any numerical schemes take more generalized forms than other research results.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.215-218
• /
• 1997

A dynamic back-propagation neural network is addressed for adaptive neural control system to approximate non-linear control system rather than static networks. It has the capability to represent the approximation of nonlinear system without mathematical analysis and to carry out the on-line learning algorithm for real time application. The simulated results show fast tracking capability and adaptive response by using dynamic back-propagation neurons.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.47.5-47
• /
• 2002

Some integration methods for the linear subsystems are examined and which algorithm is optimal for real-time simulation is considered. First, a number of typical integration methods for a linear time-invariant system are given. Then, the dynamic errors are shown both from the point of view of characteristic root errors and transfer function errors. After that we compute the dynamic errors of integration and choose the appropriate method for each system.

• 말소리와 음성과학
• /
• 제1권4호
• /
• pp.127-132
• /
• 2009

The feature vectors which are used in conventional speaker recognition (SR) systems may have many correlations between their neighbors. To improve the performance of the SR, many researchers adopted linear transformation method like principal component analysis (PCA). In general, the linear transformation of the feature vectors is based on concatenated form of the static features and their dynamic features. However, the linear transformation which based on both the static features and their dynamic features is more complex than that based on the static features alone due to the high order of the features. To overcome these problems, we propose an efficient method that applies linear transformation and temporal information of the features to reduce complexity and improve the performance in speaker verification (SV). The proposed method first performs a linear transformation by PCA coefficients. The delta parameters for temporal information are then obtained from the transformed features. The proposed method only requires 1/4 in the size of the covariance matrix compared with adding the static and their dynamic features for PCA coefficients. Also, the delta parameters are extracted from the linearly transformed features after the reduction of dimension in the static features. Compared with the PCA and conventional methods in terms of equal error rate (EER) in SV, the proposed method shows better performance while requiring less storage space and complexity.