img
img
Adaptive Input Voltage Prediction Method Based on ANN for Bidirectional DC DC Converter     
Yazarlar (3)
Rahmi İlker Kayaalp
Ahi Evran Üniversitesi, Türkiye
Tuğçe Demirdelen
Karamanoğlu Mehmetbey Üniversitesi, Türkiye
Mehmet Tümay
Çukurova Üniversitesi, Türkiye
Devamını Göster
Özet
For industrial applications, Bidirectional DC-DC converters (BDCs) are used in recent years. And also their efficiency results are improved to apply different control methods. ANN algorithms is one of the new control topic in literature. This paper attempts to improve the dynamic performance of bidirectional dc-dc converter. And it deals with a novel control scheme related with an adaptive input voltage control by using ANN algorithms based on SOM. Firstly, adaptive input voltages are classified by ANN algorithms (SOM, SVM, FF) and the best suitable and proposed ANN algorithm is SOM which is selected for this controller. The best efficiency case is selected in this algorithm. Then, the voltage values are predicted. This voltage values are used in simulation models. Thus, the proposed system works both effective and high efficiency. Theoretical analysis and simulation results obtained from an actual industrial network model in PSCAD verify the viability and effectiveness of the proposed Bidirectional DC-DC Converter (BDC).
Anahtar Kelimeler
Algorithms | Artificial Neural Network | Bidirectional DC-DC Converter | Feed-Forward | Self Organizing Map | Support Vector Machine | Voltage Control
Bildiri Türü Tebliğ/Bildiri
Bildiri Alt Türü Tam Metin Olarak Yayımlanan Tebliğ (Uluslararası Kongre/Sempozyum)
Bildiri Niteliği Web of Science Kapsamındaki Kongre/Sempozyum
Bildiri Dili İngilizce
Kongre Adı Second International Conference on Systems Informatics, Modelling and Simulation
Kongre Tarihi 01-06-2016 / 03-06-2016
Basıldığı Ülke Letonya
Basıldığı Şehir Rıga