• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.331-331
• /
• 2000

This paper presents the mixed $H_2/H_{\infty}$ output feedback controIler design method for linear systems with delayed state. The objective is to design the output feedback controller which minimizes the H$_2$-norm of one transfer function while ensuring the H$_{\infty}$-norm of the other is held below a chosen level. When objective is tormulated in terms of a common Lyapunov function, the sufficient conditions of existence of mixed $H_2/H_{\infty}$ controller are given in terms of LMIs. terms of LMIs.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.11 no.5
• /
• pp.163-169
• /
• 2003

Modeling of engine-generator system and its control responses are investigated using high performance generator controller. The nonlinear engine is modeled using mean torque production model based on experimental engine map. In case of diesel engine. the amount of injected fief is decided by engine controller depending on the APS(Acceleration Position Sensor) value. An electromechanical generator model contains electrical circuits and moment of inertia. The generator controller maximizes the performance of generator using decoupling and linearized current feedback control. The generator control system consists of 3-phase IGBT inverter and controller board based on 32 bit floating point DSP. Field oriented control algorithm with digital current feedback control at 10kHz sampling enabled high performance torque and speed control of induction machine. Not only the steady state but also the transient state responses can be evaluated through a batch test of the engine generator system. Developed engine and generator modeling and control can be utilized in various applications such as Series Hybrid Electric Vehicle(SHEV), engine-generator for emergency, and other hybrid generation systems.

• Proceedings of the KIPE Conference
• /
• 2004.07b
• /
• pp.799-802
• /
• 2004

There are many methods in controlling inverter's voltage and currents. most of all, PI control method is a general method. PI control has some merits. But, PI control has zero effect. So, steady-state response errors always exist by the zero effect. For removing the steady-state error, This paper presents the modeling, design and analysis of the double loop feedback control scheme. and computing the value of parameters and applying In the single-phase full bridge inverter for comparison and analysis between the PI control and PDFF control. The system model is employed to examine the dynamics of power circuit and select appropriate feedback variables for stable operation of the closed-loop UPS inverter system. It analyzes and proves the output characteristic of inverter system with the PDFF control.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.3
• /
• pp.591-597
• /
• 2009

The problem of controlling a finite-state asynchronous sequential machine is examined. The considered machine is governed by input/output control, where access to the state of the machine is not available. In particular, disturbance inputs can infiltrate into the asynchronous machine and provoke unauthorized state transitions. The control objective is to use output feedback to compensate the machine so that the closed-loop system drive the faulty asynchronous machine from a failed state to the original one. Necessary and sufficient condition for the existence of appropriate controllers are presented in a theoretical framework. As a case study, the closed-loop system of an asynchronous machine with the proposed control scheme is implemented in VHDL code.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.04a
• /
• pp.372-377
• /
• 1996

A time-optimal control law for quick, strongly nonlinear systems like revolute robots has been developed and demonstrated. This procedure involves the utilization of neural networks as state feedback controllers that learn the time-optimal control actions by means of an iterative minimization of both the final time and the final state error for the known and unknown systems with constrained inputs and/or states. The nature of neural networks as a parallel processor would circumvent the problem of "curse of dimensionality".ity".uot;.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.151.2-151
• /
• 2001

In this paper, the design of a two-degree-of-freedom(TDF) controller is proposed to track the reference model, as well as to reject an influence of measurable disturbance for descrpitor system. The TDF controllers based on the Youla parametrization reveals that the design of the feedforward controller and the regulator can be done independently. First, to solve this problem, we will change descriptor system into regular state space system using a state feedback. And then, the feedforward controller is determined by solving a full information approach for augmented system with a nominal control constraint, and the regulator is designed by means of the loop-Shaping method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.6
• /
• pp.487-493
• /
• 2001

In this paper, a new reduced-order dynamic observer method is presented. Two types of observers are pronounced, namely, the model based reduced-order dynamic observer and the Luenburger type reduced-order dynamic observer. Useful design algorithms are also provided for each structure. The essential features of the proposed observed design methods are addressed to be qualified ad effective observers. The proposed method clarifies the duality between the controller and observer designs.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1078-1081
• /
• 1993

The Objective of this paper is to provide fuzzy control designers with a design tool for stable fuzzy logic controllers. Given multiple sets of data disturbed by vagueness uncertainty, we generate the implicative rules that guarantee stability and robustness of closed-loop fuzzy dynamic systems. We propose the cell-state transition method which utilizes Hsu's cell-to-cell mapping concept [1]. As a result, a generic and implementable design methodology for obtaining a fuzzy feedback gain K, a fuzzy hypercube [2], is provided and illustrated with simple examples.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.6
• /
• pp.641-646
• /
• 2014

In this paper, we propose a controller for a ball and beam system which reduces the measurement error effect under AC and DC noise. The ball and beam system measures data through a sensor. If sensor noise is included in a controller via the feedback channel, the signal is distorted and the entire system cannot work normally. Therefore, some appropriate action for the measurement error effect is essential in the controller design. Our controller is equipped with a gain-scaling factor and a compensator to reduce the effect of measurement error in the feedback signal. Effectively, our proposed controller can reduce the AC and DC noise of a feedback sensor. We analyze the proposed controller by Laplace transform technique and illustrate the improved control performance via an experiment for a ball and beam system.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.12
• /
• pp.2292-2297
• /
• 2010

To improve the performances of Sawyer motors and to regulate yaw rotation, various feedback control methods have been developed. Almost all of these methods require information on the position, velocity or full state of the motor. Therefore, in this paper, a nonlinear observer is designed to estimate the full state of the four forcers in a Sawyer motor. The proposed method estimates the full state using only positional feedback. Generally, Sawyer motors are operated within a yaw magnitude of several degrees; outside of this range, Sawyer motors step out. Therefore, this observer design assumes that the yaw is within ${\pm}90^\b{o}$. The convergence of the estimation error is proven using the Lyapunov method. The proposed observer guarantees that the estimation error globally exponentially converges to zero for all arbitrary initial conditions. Furthermore, since the proposed observer does not require any transformation, it may result in a reduction in the commutation delay. The simulation results show the performance of the proposed observer.