Title :
Multivariable MAC control based on residual error gray feedback correction and its application
Author :
Guo, Peng ; Li, Tengfei
Author_Institution :
Dept. of Autom., North China Electr. Power Univ., Beijing
Abstract :
Unit load system of power plant is a double-input and double-output multivariable system with big inertia. The conventional control method cannot get satisfactory result. In this paper, the system is firstly decomposed into two double-input and single-output subsystems and two MAC controllers are designed accordingly. By solving the two controllerspsila equations, predictive control sequence is obtained. As to systempsilas uncertainty such as time-varying characteristics, residual error gray multi-step feedback correction (REGFC) method takes the place of the conventional single step feedback correction method. The new method can better predict the error sequence between the object output and the MAC model output so that a better feedback correction effect can be achieved. The simulation of 125 MW unit load system proves that the new method has satisfactory performance.
Keywords :
control system synthesis; feedback; load regulation; multivariable control systems; power plants; power system simulation; predictive control; time-varying systems; MAC controller design; double-input multivariable system; multivariable MAC control; power plant; predictive control sequence; residual error gray feedback correction; residual error gray multistep feedback correction method; single step feedback correction method; single-output multivariable system; time-varying characteristics; unit load system; Control systems; Equations; Error correction; Feedback; MIMO; Power generation; Predictive control; Predictive models; Time varying systems; Uncertainty; multivariable MAC; residual error gray multi-step feedback correction; unit load system;
Conference_Titel :
Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-2502-0
Electronic_ISBN :
978-1-4244-2503-7
DOI :
10.1109/ICAL.2008.4636188