• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.848-850
• /
• 1997

A control method of UPFC (Unified Power Flow Controller) to reduce low frequency oscillation is proposed. UPFC is modelled by voltage source, which magnitude and phase angle can be controlled. Because there needs some time to change to desired value, d-axis voltage and Q-axis voltage is modeled by 1st order delay. LQG(Linear Quadratic Gaussian) is used. It is shown that low frequency can be damped by control of UPFC effectively.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.7
• /
• pp.1309-1313
• /
• 2010

In this paper, we propose a novel guidance law for controlling an UAV(Unmanned Air-Vehicle) to follow a reference line in vertical plane. A kinematics model representing the relative motion of the UAV to the reference line is derived. And then LQR(Linear Quadratic Regulator) theory is applied to the model to derive the VLFG(Vertical Line Following Guidance) law. The resultant guidance law forms a gain-scheduling controller scheduled by a simple parameter $\sigma$ which is a function of the UAV's velocity, axial acceleration, gravity, and the slope of the reference line. Also derived is a stability condition for the $\sigma$ variation based on Lyapunov theory. Simulation results show that the proposed guidance law can be applied effectively to UAV guidance algorithm design.

• Proceedings of the KIEE Conference
• /
• 1998.07c
• /
• pp.1024-1026
• /
• 1998

In this paper we present a Neural network approach to select weighting matrices of Linear-Quadratic-Gaussian (LQG) controller for TCSC control. The selection of weighting matrices is usually carried out by trial and error. A weighting matrices of LQG control selected effectively using Neural network. It is shown that simulation results in application of this method to one machine infinite bus system are satisfactory.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.3
• /
• pp.212-219
• /
• 1999

In this paper we present a neural network approach to select weighting matrices of Linear-Quadratic-Gaussian(LQG) controller for TCSC control. The selection of weighting matrices is usually carried out by trial and error. A weighting matrices of LQG control are selected effectively using Kohonen network. It is shown that simulation results in application of this method to three machine nine bus system are satisfactory.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2240-2242
• /
• 2001

A rotary double inverted pendulum, the nonlinear system has a regulation problem. In this paper, we linearize the nonlinear system at the upright equilibrium position. The linearized system can be expressed in state space. To maintain the upright position, we design a feedback controller using LQR(Linear Quadratic Regulator) algorithm. Then we simulate the system with third-order Adams Bashforth Moulton Method. The simulated result shows that the applied algorithm is effective for the regulation problem.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.11a
• /
• pp.115-120
• /
• 2004

This paper concerns the active aeroelastic control of flapped wing systems exposed to blast and/or the sonic boom in an incompressible flow field. This is achieved via implementation of a robust estimation capability (sliding mode observer: SMO), and of the use of the deflected flap as to suppress the flutter instability or enhance the subcritical aeroelastic response to blast loads. To this end, a control methodology using LQG(Linear Quadratic Gaussian) in conjunction with SMO is implemented, and its performance toward suppressing flutter and reducing the vibrational level in the subcritical flight speed range is demonstrated. Moreover, its performances are compared to the ones provided via implementation of conventional LQG with Kalman filter.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2006.03a
• /
• pp.511-518
• /
• 2006

In this study, seismic response mitigation effect of an MR damper generating response-dependent frictional force is investigated. It has been reported in previous studies that passively operated MR damper with constant input current doesn't show better control performance than semi-active MR damper with varying input current calculated by control algorithms such as linear quadratic regulator and sliding mode control. However, in order to operate the MR damper semi-actively, other control systems besides the damper itself such as sensors for measuring structural responses and controller for calculating optimal input current are necessary, which deteriorate the economical efficiency. This study presents a MR damper generating frictional force of which magnitude is controlled in accordance to the displacement and velocity transferred to the damper. Numerical analyses results indicate that the performance of the response dependent MR damper is closely related with the range of the friction force and it can be designed to short better control performance than the passive MR damper.

• Earthquakes and Structures
• /
• v.21 no.6
• /
• pp.577-585
• /
• 2021

In this research, a neural network (NN) method was developed, which combines H-infinity and fuzzy control for the purpose of stabilization and stability analysis of nonlinear systems. The H-infinity criterion is derived from the Lyapunov fuzzy method, and it is defined as a fuzzy combination of quadratic Lyapunov functions. Based on the stability criterion, the nonlinear system is guaranteed to be stable, so it is transformed to be a linear matrix inequality (LMI) problem. Since the demo active vibration control system to the tuning of the algorithm sequence developed a controller in a manner, it could effectively improve the control performance, by reducing the wind's excitation configuration in response to increase in the cost efficiency, and the control actuator.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.6
• /
• pp.915-927
• /
• 2008

Networked control systems(NCSs) have gained increasing attention in recent years due to their advantages and potential applications. The network Quality-of-Service(QoS) in NCSs always fluctuates due to changes of the traffic load and available network resources. To handle the network QoS variations problem, this paper presents an intelligent scheduling control method for NCSs, where the sampling period and the control parameters are simultaneously scheduled to compensate the effect of QoS variation on NCSs performance. For NCSs with network-induced delays and packet dropouts, a discrete-time switch model is proposed. By defining a sampling-period-dependent Lyapunov function and a common quadratic Lyapunov function, the stability conditions are derived for NCSs in terms of linear matrix inequalities(LMIs). Based on the obtained stability conditions, the corresponding controller design problem is solved and the performance optimization problem is also investigated. Simulation results are given to demonstrate the effectiveness of the proposed approaches.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.2
• /
• pp.944-949
• /
• 2017

Recently, the demand of maglev systems in the manufacturing industry for LCD and OLED display panels, which are required to be very clean and possess vacuum systems, has been increasing due to their characteristics such as being non-contact, noise free and eco-friendly. However, it is still a challenge to simultaneously control both the propulsion and levitation for their interactive effect difficult to be exactly measured. In this paper, we proposed a new tuning method for controlling the magnetic levitation force robustly against the levitation disturbance caused by a propulsion system, based on LQ servo optimal control. The disturbance torque of the LSM propulsion system is calculated through FEM analysis in such a way that the LQ servo controller is determined in order to minimize the effect of the disturbance. The robust performance of the proposed LQ servo control method for the in-track type magnetic levitation systems is demonstrated via simulations and experiments.