Bu kaydın yasal hükümlere uygun olmadığını düşünüyorsanız lütfen sayfa sonundaki Hata Bildir bağlantısını takip ederek bildirimde bulununuz. Kayıtlar ilgili üniversite yöneticileri tarafından eklenmektedir. Nadiren de olsa kayıtlarla ilgili hatalar oluşabilmektedir. MİTOS internet üzerindeki herhangi bir ödev sitesi değildir!

Design Of Robust Power System Stabilizer For Single Machine Infinite Bus System Using Modern Control Approaches

Oluşturulma Tarihi: 04-04-2019

Niteleme Bilgileri

Tür: Tez

Alt Tür: Doktora (Mühendislik)

Yayınlanma Durumu: Yayınlanmamış

Dosya Biçimi: PDF

Dil: İngilizce

Konu(lar): TEKNOLOJİ, Makina mühendisliği ve makinalar,

Yazar(lar): Ali, İssa Yousf Said (Yazar),

Emeği Geçen(ler): İrfanoğlu, Bülent (Danışman),

Anahtar Kelimeler

Active Disturbance Rejection Control (ADRC), Artificial Neural Network (ANN), Feedback Error Learning (FEL), Low Frequency Oscillations (LFO), Power system stabilizer (PSS).


Özet

Due to the rapid growing demand for electricity, power systems nowadays have become operating nearer to their stability limits which cause a lot of instability problems and could potentially result in serious technical challenges. Since the Conventional Power System Stabilizer (CPSS) is the most commonly used controller in power systems, many techniques have been proposed in the last few years aimed to improve the performance of conventional power system stabilizer using some intelligent optimization algorithms such as Genetic Algorithm, Fuzzy Logic, Particle Swarm and others. However, although local optimization can be achieved to a satisfactory degree by setting the stabilizer parameters in optimal way, the robustness of the stabilizer is still in doubt and it may not guarantee good performance when the operating point changes or some unexpected disturbance occurs. This dissertation presents an application of two types of modern robust control strategies on power system in order to improve system dynamic stability. Those two control strategies are Active Disturbance Rejection Control (ADRC), and Feedback ErrorviLearning Control (FEL). The first proposed controller which is ADRC algorithm has an advantage that makes the power system more robust against wide range of troublesome disturbances that commonly encountered in such systems. The most important feature of ADRC approach is that it requires little information from the plant model forasmuch ‘under certain circumstances’ the relative order of open loop transfer function information is quite sufficient to design a robust controller. The second is FEL controller which employs the Artificial Neural Network (ANN) technology in framework of feedback error learning control strategy to enhance dynamic stability of the system. The nature of structural configuration of FEL controller, which combines a conventional power system stabilizer and a neural network, makes it powerful controller includes the well-known advantages of CPSS with the additional features of artificial neural networks like adaptation and nonlinearity. The proposed ADRC and FEL controllers have been developed in this study for a power system consists of Synchronous Machine connected to an Infinite Bus (SMIB) through external reactance under small disturbance. The effectiveness of both ADRC and FEL have been verified by comparing both of them with an optimally tuned conventional power system stabilizer. Moreover, the comparison has been done between the proposed ADRC and FEL control strategies under wide range of operating conditions. All tests and case studies have been conducted under Matlab Simulink environment. The simulation results showed that the proposed control schemes ensured high performance and system stability with presence of different types of loading conditions especially at some critical operating points where the conventional stabilizer had failed.


İçindekiler



Açıklamalar



Haklar



Notlar



Kaynakça


Atıf Yapanlar

Gözat Sayfasına Dön

 

Sosyal Medya ve Araçlar

İstatistikler

  • Kayıt
    • Bu ay: 21
    • Toplam: 2405
  • Online
    • Ziyaretçi: 18
    • Üye: 0
    • Toplam: 18

Detaylı İstatistikler