Abstract
This paper presents a direct generalized minimum-variance self tuning controller with a PID structure using neural network which adapts to the changing parameters of the nonlinear system with nonminimum phase behavior, noises and time delays. The self-tuning controller with a PID structure is a combination of the simple structure of a PID controller and the characteristics of a self-tuning controller that can adapt to changes in the environment. The self-tuning control effect is achieved through the RLS (recursive least square) algorithm at the parameter estimation stage as well as through the Robbins-Monro algorithm at the stage of optimizing the design parameter of the controller. The neural network control effect which compensates for nonlinear factor is obtained from the learning algorithm which the learning error between the filtered reference and the auxiliary output of plant becomes zero. Computer simulation has shown that the proposed method works effectively on the nonlinear nonminimum phase system with time delays and changed system parameter.
본 논문에서는 잡음과 시간지연이 존재하며 시스템 파라미터가 변하는 비선형 비최소위상 시스템에 적응하는 신경회로망이 결합된 PID구조를 갖는 일반화 최소분산 자기동조제어기를 제안한다. PID구조를 갖는 자기동조는 PID제어기처럼 구조가 간단하고 계통을 정밀하게 제어하는 자기동조 제어기의 특성을 그대로 유지할 수 있다. 일반화 최소분산 자기동조 제어기 파라미터는 비선형 시스템을 선형시스템으로 간주하고 순환최소자승법으로 추정하며 설계계수의 값은 확률근사법인 Robbins-Monro 알고리듬을 이용하여 자동조정하였다. 역전파 학습 알고리듬을 사용하는 신경회로망 제어기는 비선형 부분의 제어를 보상하기 위해 필터된 기준입력과 필터된 플랜트 출력이 같도록 제어값을 출력한다. 컴퓨터 시뮬레이션을 통해 제안한 방법이 시스템의 파라미터가 변하는 비최소위상 시스템에 잘 적응함을 보였다.