DEEP LEARNING APPLICATION FOR MULTIPLE FAULT DETECTION IN THREE-PHASE INDUCTION MOTORS

Creative Commons License

Noğay H. S.

7th INTERNATIONAL CONGRESS ON ENGINEERING SCIENCES AND MULTIDISCIPLINARY APPROACHES, İstanbul, Türkiye, 25 - 26 Mayıs 2024, ss.236-243, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.236-243
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Bursa Uludağ Üniversitesi Adresli: Evet

Özet

Abstract: This study presents a deep learning application designed to detect and classify multiple fault conditions in three-phase induction motors using thermal imaging. Leveraging the capabilities of convolutional neural networks (CNNs), the model aims to predict ten specific types of fault conditions and one healthy operating condition. The CNN model was trained on a diverse dataset of thermal images capturing these fault conditions, achieving an impressive accuracy rate of 98.7%. The high accuracy underscores the model's effectiveness in identifying and classifying various fault states, which is critical for timely maintenance and prevention of motor failures. Keywords: Deep Learning, CNN, Fault Detection, Three-Phase Induction Motors, Thermal Imaging