Issues of optimality and computation in repetitive control
Başlık çevirisi mevcut değil.
- Tez No: 586409
- Danışmanlar: Prof. DAVID H. OWENS
- Tez Türü: Doktora
- Konular: Matematik, Mathematics
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2009
- Dil: İngilizce
- Üniversite: The University of Sheffield
- Enstitü: Yurtdışı Enstitü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 215
Özet
Özet yok.
Özet (Çeviri)
In this thesis, a few new controller schemes are proposed in the linear optimal Repetitive Control (RC) context. According to the literature, previous research in this field is relatively small in that employing high degree repetitive controller always leads to large computational cost issues for solving the relevant equations (Diaphontine or Riccati equations). The necessity of the research work in this area is prominent and urgent in both continuous and discrete time RC study, and more specifically, for both single and multi-periodic continuous and discrete time repetitive control systems. The control schemes in this thesis, termed as Low Computation Cost Schemes, whilst still require to solve the algebraic Riccati equations (AREs), they bypass the original requirement of solving high order AREs by reducing the order of large systems into their low order counterparts. This is achieved either by a projection method or a low order scheme based on the spectrum analysis of the demand signal. By doing so, the original requirement of solving large sparse AREs is converted into simply solving their low order counterparts, which hugely lessens the original high computation requirement. To demonstrate the effectiveness of two proposed methods, they are applied to two practical examples: a software cam system and a servomotor motion control system, where the experimental results suggest expected tracking performances. The major contribution of this thesis is the simplification of the requirement of solving high order Riccati equations in optimality based repetitive control systems. By using these two proposed methods, desired small tracking errors can be gained even when the period/periods of reference/disturbance is/are large, and the overall closed loop systems highly reduce the computation cost by only solving low order Riccati equations. The analysis shows that the closed loop systems obtained from both methods are globally Bounded Input Bounded Output (BIBO) stable and applications to the practical examples demonstrate the theoretical findings.
Benzer Tezler
- Gezgin satıcı problemi
Traveling salesman problem
VOLKAN M. ÖZALP
Yüksek Lisans
Türkçe
1995
Endüstri ve Endüstri Mühendisliğiİstanbul Teknik ÜniversitesiDOÇ.DR. FÜSUN ÜLENGİN
- General continuous time models for production planning and scheduling of batch processing plants: MILP formulations and computational ıssues
Kesikli üretim merkezlerinin planlanması ve çizelgelenmesi için genel sürekli zaman modelleri: MILP formülasyonları ve hesapsal boyutları
SEZA ORÇUN
Doktora
İngilizce
1999
Kimya MühendisliğiBoğaziçi ÜniversitesiKimya Mühendisliği Ana Bilim Dalı
PROF. DR. ÖNER HORTAÇSU
- Short term electricity load forecasting with deep learning
Derin öğrenme ile kısa dönemli elektrik yük talep tahmini
İBRAHİM YAZICI
Doktora
İngilizce
2022
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiEndüstri Mühendisliği Ana Bilim Dalı
DR. ÖĞR. ÜYESİ ÖMER FARUK BEYCA
- Kuvvetli yer hareketi kayıtlarının GPS ile optimal süzgeçlenmesi
Optimal filtering of strong motion records
MURAT DORUK ŞENTÜRK
Yüksek Lisans
Türkçe
2019
Jeofizik MühendisliğiAnkara ÜniversitesiJeofizik Mühendisliği Ana Bilim Dalı
PROF. DR. BAHADIR AKTUĞ
- Spor salonundaki aletlerin kapasite tahminlemesi için bir uygulama çerçevesi
An application framework for gym equipment capacity forecast
OSMAN ÇAYLI
Yüksek Lisans
Türkçe
2018
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DR. ÖĞR. ÜYESİ GÖKHAN İNCE