Yong Ho Kim

A scheduling method for network-based control systems

This paper presents a scheduling method for network-based control systems with three types of data (periodic data, sporadic data, and messages). As a basic parameter for the scheduling method for network-based control systems, a maximum allowable delay bound is used,which guarantees stability of network-based control systems and is derived from characteristics of the given plant using the presented theorems. The presented scheduling method for network-based control systems can adjust the sampling period as small as possible, allocate the bandwidth of the network for three types of data, and exchange the transmission orders of data for sensors and actuators. In addition, the presented scheduling method guarantees real-time transmission of sporadic and periodic data, and minimum utilization for nonreal-time messages. The proposed method is shown to be useful by examples.

An application of min–max generalized predictive control to sintering processes

Abstract This paper presents an application of min-max generalized predictive control (MMGPC) to burnthrough point control in industrial sintering processes. A simple model is used as a plant model for the burnthrough point control, which is derived for a complicated sintering process by introducing events into a continuous variable system. As a control algorithm, the MMGPC is used. It is robust to disturbances and has guaranteed stability. Simulations of the burnthrough point control using the MMGPC are performed using a derived model of a POSCO (Pohang Steel Company, Korea) sintering plant and real burnthrough time data of sinter packages. The simulation results of the MMGPC are compared with those of PID control, and are satisfactory.

Stability and a scheduling method for network-based control systems

This paper presents network-induced maximum allowable delay bounds for stability of network-based control systems and a scheduling method for network-based control systems by using these bounds. The maximum allowable delay bounds for stability of network-based control systems are obtained using the Lyapunov theorem. Also a scheduling method considering three kinds of data (periodic, real-time asynchronous, and nonreal-time asynchronous data) is presented, which allocates the bandwidth of a network for each node and determines the sampling time of each sensor and controller. This method also guarantees the real-time transmission of the real-time synchronous data and periodic data, and the transmission of the nonreal-time asynchronous data. To show usefulness of the proposed method, the proposed method is tested in two types of network protocols such as the token control and the central control.

Event-based modeling and control for the burnthrough point in sintering processes

This paper treats modeling and control for the burnthrough point in industrial sintering processes. First, a simple state-space model for event-time dynamics is derived for a complicated system by introducing events into a continuous variable system. For the control of the sintering process, a two-stage control policy is used. First, optimal interevent time is obtained from an output-constrained receding horizon control with a least-square prediction algorithm. Then the average strand speeds for the interevent times are obtained and applied as a command signal to a motor control system. It is proved that the proposed output-constrained receding horizon control law with a short horizon stabilizes the closed-loop system and the steady-state error for a set-point becomes zero. The real-time experiments are carried out in a POSCO (Pohang Steel Company, Korea) sintering plant and satisfactory results are presented in this paper. Also, computer simulations are carried out and compared with the real-time experiments.

Computational complexity of general fuzzy logic control and its simplification for a loop controller

This paper analyzes the number of operations and parameters of general fuzzy logic control algorithms. And limitations of loop controllers to implement the fuzzy logic control are investigated in terms of computation time and required memory. Using analysis of general fuzzy logic control algorithms, it is shown that general fuzzy logic control algorithms are not suitable for loop controllers. A simplified fuzzy logic control for loop controllers is presented using a fixed control table with mapping parameters. This simplified fuzzy logic control is shown to be suitable for a loop controller with regard to the computation time and the required memory. For easy programming of the simplified fuzzy logic control, a fuzzy logic control block is suggested as a specific block for general graphical block diagram editors. Experiments show that the simplified fuzzy logic control overcomes the limitations of loop controllers and provides good performance.

Min–max generalized predictive control with stability

Abstract This paper presents a min–max generalized predictive control (MMGPC) which is robust to disturbances and has guaranteed stability. The MMGPC is derived from the min–max problem. It has non-recursive forms which do not use the Riccati equations. Stability conditions of the proposed control law are presented, which can be met by adjustment of some parameters such as input–output weightings. This paper presents a systematic way to obtain appropriate parameters for these stability conditions by using the linear matrix inequality (LMI) method. It is also shown that the suggested control guarantees that induced norm from disturbances to system outputs is bounded by a constant value under the same stability conditions.

Design of a fuzzy logic controller module for a loop controller

In this paper a design of a fuzzy logic controller module for a loop controller is proposed. The limitations in implementing the fuzzy logic controller in a loop controller are the speed of a fuzzy logic controller, and the number of parameters to define a fuzzy logic controller. By analyzing a fuzzy logic controller program, a fuzzy logic controller is shown to be hard to be implemented in a loop controller because of the inference speed and the number of parameters. As a solution of the problem, a method using a fixed control table with mapping parameters is shown to be implementable to a loop controller. The fixed control table increases the inference speed and reduces the number of defining parameters, while the mapping parameters provides flexibility of the designed fuzzy logic controller. By implementation the proposed fuzzy logic controller is shown to meet the speed limitation.

Robust generalised predictive control with terminal output weightings

Abstract A GPC (generalised predictive control) algorithm which is robust to disturbances is proposed. This control algorithm is derived with constrained terminal output weightings using a standard cost function of linear quadratic discrete game theory. The solution is obtained from the min-max problem which can be solved by discrete game theory and has nonrecursive forms which do not use the Riccati equations. It is derived for both finite and infinite terminal output weightings. To improve the possibility of the existence of a solution for robust constrained GPCs, a modified robust GPC algorithm which uses a different cost function from the standard one is also presented. The usefulness of the proposed controllers is demonstrated by computer simulations.

Weighted receding horizon predictive control and its related GPC with guaranteed stability

Develops a receding horizon predictive control (RHPC) with terminal output weighting for stochastic state space models, which consists of receding horizon tracking control (RHTC) and receding horizon filter (RHF). A modified GPC is suggested for I/O models, which has the same transfer function as RHPC with RBF. It is shown that the standard GPC is a special case of this modified GPC. It is also shown that RHPC with infinite terminal output weighting and its related modified GPC guarantee the closed-loop stability of linear systems.<<ETX>>

Applications of Min-Max Generalized Predictive Control to Sintering Processes

Abstract This paper presents applications of min-max generalized predictive control (MMGPC) to burnthrough point control in industrial sintering processes. A simple model is used as a plant model for the burnthrough point control. It is derived for a complicated sintering process system by introducing events into a continuous variable system. As a control algorithm, the MMGPC is used. It is robust to disturbances and has guaranteed stability. Simulations of the burnthrough point control using the MMGPC are performed using a derived POSCO (Pohang Steel Company, Korea) sintering plant model and real burnthrough time data of sinter packages. The simulation results of the MMGPC are compared with those of PID control and are satisfactory.