Importance Of AI In Acoustic Emission Testing and Enhancement Provided Through Compound AI
7th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, ICHORA 2025, Ankara, Türkiye, 23 - 24 Mayıs 2025, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/ichora65333.2025.11017232
- Basıldığı Şehir: Ankara
- Basıldığı Ülke: Türkiye
- Anahtar Kelimeler: acoustic emission testing, artificial intelligence, railway engineering, transportation
- Ankara Yıldırım Beyazıt Üniversitesi Adresli: Evet
Özet
Acoustic emission has evolved as one of the essential non-destructive testing methods for railway track monitoring. The growing complexity of railway infrastructure, artificial intelligence (AI) is transforming AE testing by the automation of defect detection and classification. This paper presents a extensive review of literature on the use of AI techniques for AE testing, assessing their advantages, limitations, and the promise of compound AI models. Standalone methods like Artificial Neural ANN (Artificial Neural Networks), SVM (Support Vector Machines), CNN (Convolutional Neural Networks) offer an efficient feature extraction and classification but suffers away from data dependency, computation costs, and temporal analysis. Compound models - like CNN with Combine LSTM or Random Forest with CNN - improve accuracy by taking advantage of complementary strengths. The AI for AE testing futures research will focus on integrating for better transparency and decision making - such as Explainable AI (XAI). AI-driven AE Real time monitoring, reduced manual intervention, and better defect prediction are provided by testing. enhancing safety and efficiency within rail networks.