Recent Papers

As the manufacturing sector grows, it finds itself resorting to more efforts in combatting the hardships caused due breakdown and unscheduled downtimes. Predictive maintenance, enabled through advanced technologies, has been critical in achieving those goals. This study intends to focus on fault diagnosis in mechanical systems with the aid of deep learning, thus providing an alternative approach to predictive maintenance in manufacturing processes. The paper commences by examining the nature of routine maintenance practice and its limitation in maintaining the machine such that mechanical failure does not escalate. Analytics-driven approach of predictive maintenance kicks in to provide solutions in a timely manner and at the most suitable cost. A subset of artificial intelligence, deep learning is particularly successful in detecting complex and nonlinear interactions and dependencies from mechanical system data patterns. The major component of this research is focused on the development and application of deep learning algorithms specifically designed to identify faults on production machinery. Such models are capable of prediction based on historical data that demonstrates various fault types and operational conditions. Deep neural networks enable the system to learn specific patterns which indicate weaknesses, thus achieving a level of accuracy which conventional methods are incapable of attaining.

Copyright © 2024 Rupesh Ambavane. This is an open access article distributed under the Creative Commons Attribution License.