Publication: Yapay Sinir Ağları İle Modelleme Ve Aktif Manyetik Yataklamalı Bir Sisteme Uygulanması
Loading...
Files
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Fen Bilimleri Enstitüsü
Institute of Science And Technology
Institute of Science And Technology
Type
Abstract
Bu çalışmada, nonlineer sistemlerin yapay sinir ağları ile modellenmesi incelenmiş ve sinir ağlarının kontrol amaçlı kullanımı hakkında temel bilgiler verilmiştir. Öncelikle, açık çevrimde kararlı nonlineer örnek bir sistemin modellenmesi gerçekleştirilmiş ve elde edilen modelin geçerliliğini test etmek için bazı korelasyon fonksiyonu testleri uygulanmışıtr. Bu testler yardımıyla tatminkar bir model bulunmuştur. Daha sonra aktif yataklamalı manyetik sistemin simülasyonlarda kullanılmak üzere matematiksel bir modeli elde edilmiştir. Aktif manyetik yataklama sistemi açık çevrimde kararsız olduğundan tanıma süreci, kontrol işareti ve buna ek olarak sisteme verilen tanıma amaçlı işaret yardımıyla kapalı çevrimde gerçekleştirilmelidir. Elde edilen verilerden eğiticili öğrenme ile bir yapay sinir ağı modeli elde edilmiştir. Tüm simülasyonlar Matlab© ve Simulink© kullanılarak gerçekleştirilmiştir.
In this study, nonlinear system identification with neural networks has been examined, and some basics about the use of the neural networks in control is mentioned. Firstly, static neural network model of an exemplary open loop stable nonlinear system was obtained, and some correlation function tests were performed to validate the model. Through these tests, a satisfactory model was found. Next, the mathematical model of an active magnetic levitation system was obtained to perform simulations. Since the active magnetic bearing system is open loop unstable, identification of the system must be studied under the stabilizing feedback control signal with an extra dithering signal. Supervised learning has been performed by taking samples from the closed loop system. Good results have been obtained through the analytical model and the neural network based model. The performance has been tested through simulations carried out using Matlab© and Simulink©.
In this study, nonlinear system identification with neural networks has been examined, and some basics about the use of the neural networks in control is mentioned. Firstly, static neural network model of an exemplary open loop stable nonlinear system was obtained, and some correlation function tests were performed to validate the model. Through these tests, a satisfactory model was found. Next, the mathematical model of an active magnetic levitation system was obtained to perform simulations. Since the active magnetic bearing system is open loop unstable, identification of the system must be studied under the stabilizing feedback control signal with an extra dithering signal. Supervised learning has been performed by taking samples from the closed loop system. Good results have been obtained through the analytical model and the neural network based model. The performance has been tested through simulations carried out using Matlab© and Simulink©.
Description
Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 2001
Thesis (M.Sc.) -- İstanbul Technical University, Institute of Science and Technology, 2001
Thesis (M.Sc.) -- İstanbul Technical University, Institute of Science and Technology, 2001
Journal or Series
ISSN
ISBN
Rights
İTÜ tezleri telif hakkı ile korunmaktadır. Bunlar, bu kaynak üzerinden herhangi bir amaçla görüntülenebilir, ancak yazılı izin alınmadan herhangi bir biçimde yeniden oluşturulması veya dağıtılması yasaklanmıştır.
İTÜ theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.
İTÜ theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.
Keywords
Yapay Sinir Ağları, Nonlineer Sistem Tanıma, Aktif Manyetik Yataklama, Artificial Neural Networks, Nonlinear System Identification, Active Magnetic Bearing
Citation
Endorsement
Review
Supplemented By
Referenced By
Google Scholar
Scholar'da Ara ↗ Bu yayında DOI yok — Altmetric/Dimensions/PlumX/BIP! rozetleri DOI gerektirir.