Helical Gear Defect Detection System Using Symmetrized Dot Pattern with Convolutional Neural Network, Support Vector Machine and Random Forest

View at publisher Summarise with VIRAT
Authors: Arunkumar T, Keerthika S, V. Manimaran, Suriyaa N S, Vikas P, Vishal K K
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Helical gears operate under continuous load variations, making them prone to wear, pitting, cracks and tooth damage, all of which can reduce system efficiency and reliability. Early fault identification is essential for maintaining smooth power transmission and preventing unexpected failures in industrial machinery. In this work, vibration signals from healthy and defective helical gears are transformed into Symmetrized Dot Pattern images to reveal changes in frequency behaviour caused by gear degradation. These visual patterns are then analysed using a combination of deep learning feature extraction and machine learning classifiers to distinguish fault conditions accurately. The study focuses on developing a robust, image-based diagnostic framework capable of identifying problem areas quickly and supporting real time monitoring. The findings demonstrate that integrating visual representation of vibration signals with advanced classification techniques enhances detection capability and offers a reliable solution for improving maintenance strategies in helical gear systems.

Fields of Study & Subjects

Machine Fault Diagnosis Techniques Gear and Bearing Dynamics Analysis Industrial Vision Systems and Defect Detection

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1618–1623
Publication date
21 January 2026
Language
EN