SURVEY OF GEAR FAULT DIAGNOSIS USING VARIOUS STATISTICAL SIGNALS PARAMETERS

Abstract

Gears are critical components of industrial equipment, where gear failure results machinery failure and that consider as a significant reduction in productivity. It is always critical to keep track of the machine's health in time. Consequently, researchers have been working on developing methods for identifying and diagnosing gear problems. The purpose of this paper is focused to provide a review of a variety of diagnosis techniques that have been shown to be successful when applied to rotating machinery such as gears, as well as to highlight fault detection and identification techniques that are primarily based on vibration analysis. fluctuations from these standards generate distinctive vibration signals whose help in monitoring the gearbox malfunctions. The main sources of these fluctuations are crack tooth, chipped tooth, missing tooth, the surface wear during heat treatment or gearbox assembly, and the geometrical errors, resulting from the gear cutting process and wear. In conclusions, a brief explanation of a novel method of diagnosis based on hybrid artificial intelligence approaches that incorporate neural networks, fuzzy sets, expert systems, and fault detection is provided.