모드전환 PID제어기를 활용한 EGR 밸브 학습알고리즘 방법연구 = A Study on the Method of EGR Valve Learning Algorithm Using the Mode Switching PID Controller

다국어 초록 (Multilingual Abstract)

This paper describes a design of the position-current mode switching controller to carry out the EGR valve both end limit positions learning algorithm. The purpose of this algorithm is to know the travel range of the valve.
The mode switching controller is the multi-controller system which chooses a controller to use among controllers which are designed by multiple purposes. But the mode switching controller can’t switch mode softly. The output of a controller which is using is different from other controllers which are not using. So, there is a jump on plant input. This phenomenon is called bump transfer. Because of a bump transfer, a controller can’t control a plant for a short time after mode switching and switch other mode softly.
In this paper, the position-current mode switching controllers which use state-succession method and conditioned transfer to avoid the side effect of bump transfer are developed. The state-succession method is that a controller which will be used next mode succeed the state of the controller which is using now. And the conditioned transfer is based on the back calculation scheme which is one of the anti-windup schemes. This method controls the state of the integral controller by using the plant input and the controller output to switch next mode softly.
Using the designed position-current mode switching controller and a position-current mode switching controller which applies for a bump transfer, the learning algorithm is carried out and compared in terms of rise time, settling time, steady state error, bump size. The comparison results show that bump transfer can’t switch mode from position control mode to current control mode softly. On the other hand, the state succession method and conditioned transfer can switch mode softly and show similar performance in terms of rise time, settling time. Especially, the conditioned transfer shows better performance in terms of settling time.

목차 (Table of Contents)

• Ⅰ. 서 론 1
• 1. EGR 밸브 양끝위치 학습알고리즘 1
• 2. 모드전환 제어기 2
• Ⅱ. EGR 밸브 모델링 4
• 1. 모터 전기자회로 모델링 4
• 2. 기구부 모델링 5
• Ⅲ. 위치-전류 모드전환 제어기 설계 8
• 1. 전류제어기와 위치제어기 설계 8
• 1) 전류제어기 설계 8
• 2) 위치제어기 설계 10
• 2. 위치-전류 모드전환 제어기 설계 12
• 1) 상태계승방법을 사용한 위치-전류 모드전환 제어기 설계 12
• (1) 위치제어기에서 전류제어기로 상태계승 12
• (2) 전류제어기에서 위치제어기로 상태계승 13
• 2) 컨디션드 전환구조를 사용한 위치-전류 모드전환 제어기 설계 14
• (1) 컨디션드 전환구조를 사용한 모드전환 PID제어기 설계 14
• (2) 컨디션드 전환구조를 사용한 위치-전류 모드전환 제어기 17
• Ⅳ. 실험 및 검증 18
• 1. 실험 환경 18
• 2. 실험 결과 21
• Ⅴ. 결 론 29
• 참고문헌 30
• ABSTRACT & KEYWORD 32