Asaad A. Dubaish; Alaa A. Jaber
Abstract
Various industrial applications, including rotating and reciprocating machinery, depend on gears. Therefore, a sudden breakdown of the gears could result in substantial financial losses. ...
Read More ...
Various industrial applications, including rotating and reciprocating machinery, depend on gears. Therefore, a sudden breakdown of the gears could result in substantial financial losses. Due to this, extensive studies have focused on defect diagnosis. Both machinery maintenance decisions and preventive maintenance techniques have been aided by vibration analysis. An increased vibration is a warning sign that a machine is about to malfunction or break down. Observing and evaluating the machine's vibration pulses can identify the nature and extent of the issue and, as a result, predict when the machine will fail. The vibration signal may identify gearbox defects early on and diagnose its problems. Hence, this research highlights the main crucial steps that can be followed for defect detection and identification, mainly based on vibration analysis methodologies. It provides an application methodology for various signal-processing techniques used successfully in rotating machinery. The study briefly explains the applied methods to diagnose problems that depend on hybrid artificial intelligence approaches, such as fuzzy sets, expert systems, and neural networks. The key aspect of the present paper is the parametric comparison of the performance of various artificial Intelligence systems used in rotary machines. As such, the paper reports a comprehensive study of the gearbox defect diagnosis and provides useful analysis, which would be helpful for the usage of such techniques in the engineering industry.
