Motomiki Uchida

Optimal regulation for thermal power plants

The design of an advanced digital control system that is used to regulate the steam temperature in the boiler of a thermal power plant is described. An autoregressive model is used to obtain the plant transfer characteristics. To compensate for system nonlinearities, the calculations for system identification and feedback gain are determined for two or three load levels. Experimental results have demonstrated the viability of the method and suggest that the advanced digital control system will replace conventional control in future applications.<<ETX>>

Implementation of Optimal Control at a Supercritical Variable-Pressure Thermal Power Plant

Abstract Optimal steam-temperature control of a supercritical variable-pressure type thermal power plant is presented with emphasis on the actual procedure of controller design. A control system based on Linear Quadratic Regulator Theory is employed. After an introduction on the special features of variable-pressure boiler control, the concept of control system and some results of the preliminary study on a digital simulation model are described. Then, the LQ regulator implemented at a 500MW plant is introduced with some field test results which confirm the advantages of the optimal controller over the conventional PID controller. The results of this investigation revealed that the optimal control system, which has been successfully in routine operation at the constant-pressure plants is also effective for improving variable-pressure boiler control.

Computing-Predictor-Based MRACS for Nonlinear Time-Varying Power Plants Based on a New System Identification Method

Abstract This paper introduces an MRACS (Model Reference Adaptive Control System) developed by the authors, which can be applied to nonlinear time-varying plants. The feature of the proposed MRACS is that it uses a simulation model which has the same structure and parameters as the plant. Control signals, that makes the plant output follow specified reference outputs, are synthesized in a circuit called the “computing-predictor-network” in which prediction of state variables taken from the simulator is utilized. System identification and parameter estimation are also carried out by using a simulator with identical structure to the plant. The method used for system identification is the one developed to suit the proposed MRACS. In the paper, the concept of the proposed MRACS is introduced in the first place. Then the proposed system identification method is explained with an example of its application to the simulation model of the feedwater heater control system of a power plant. Finally, the control performance of the proposed MRACS is demonstrated with some results of simulation study, in which the steam temperature control system of a variable-pressure power plant is used.

A Quick Identification Method of Continuous-Time Nonlinear Systems and Its Application to Power Plant Control

Abstract In this paper, a quick identification method based on the short time record of input-output data is introduced for joint state and parameter estimation of nonlinear continuous-time systems. The method can then be used for on-line monitoring the system with unknown parameters. An application way of the method to main-steam temperature control of a thermal power plant are developed in the framework of model reference adaptive control system (MRACS). Simulation studies on a super-heater model have been carried out to demonstrate the effectiveness of the proposed method.

A Hybrid Control Approach to Nonlinear Plant Stabilization, Suppression of Disturbance, and Compensation of Deviation Errors

This brief proposes a hybrid control approach which can realize desired control performances of multiple objectives for a class of nonlinear plants, such as stability, suppression of disturbance, compensation of deviation errors, and so on. The proposed method is composed of two control schemes. One is a multiple-purpose pole control (MPPC) for the stabilization of an unstable nonlinear plant and the other is simulator-based foresight control (SFC) which aims the suppression of disturbance and the compensation of deviation errors. In MPPC, an unstable nonlinear plant can be stabilized by means of old fashioned pole assignment control method. The procedure of tuning of control parameters is performed with simulator-study assisted design by just observing the responses of state variables. In SFC, the suppression of disturbance can be realized by the insertion of servo filter in line of disturbance which can adjust the relative degree of disturbance. In addition, ideal convergence of errors of state variables can be realized by defining control target equation which describes an ideal convergence dynamics of deviation errors. The proposed SFC can be applied for the full range disturbance, and can predict output behavior of nonlinear plant at any load level. In this brief, the simulation results on the control of a stable nonlinear boiler with 50 width load change of a 14th order model as well as many unstable nonlinear system models are illustrated to demonstrate the effectiveness of the proposed method.