Fault classification of polymer spur gear using advanced signal processing and machine learning techniques

Abstract

Polymer gears possess numerous advantages over metal gears, including their low weight, low vibration and noise, affordability high corrosion re-sistance, and ease of manufacturing. These characteristics make polymer gears highly suitable for power transmission applications in various do-mains such as aviation, electric vehicles, textile machines, windshield wip-ers, printers, packaging machines, and mixers. Over the past few decades, extensive research has been conducted on polymer gear materials, loading conditions, the influence of gear pairs, and tooth modifications. Among these studies, researchers have identified that polymer gears often fail due to wear, pitting, and thermal damage. Researchers have employed a variety of techniques to modify the teeth of polymer gears, such as inserting steel pins into the internal holes of the gear teeth. This modification aims to en-hance gear durability by mitigating thermal effects on the teeth. However, surprisingly, there is a significant gap in research exploring how these tooth modifications affect gear noise and vibration levels. Apart from that, only a few research studies have been conducted on detecting faults in polymer gears. Therefore, the objective of this thesis is to investigate the influence of tooth modification on the noise and vibration characteristics of polymer gears. Furthermore, the study aims to develop approaches for detecting and classifying faults in polymer gears.