Sistema de monitoreo de vibraciones por medio de modelos de mantenimiento predictivo 4.0 PHM y sistemas inerciales con el propósito de diagnosticar el estado y salud de activos

Abstract

Industrial rotating machines such as induction motors require constant maintenance in order to extend their productive and operational life as long as possible. Nowadays, there are several maintenance concepts, one of which is predictive maintenance, which is planned and systematic with the objective of preventing failures, minimizing the risk of corrective maintenance. With the rise of new technologies, the concept of Maintenance 4.0 has been created, which is the evolution of preventive maintenance. This is based on the use of Industry 4.0 technologies, such as the Internet of Things (IoT), Big Data, Artificial Intelligence, Augmented Reality, Digital Twins, among others. The main concept is to optimize maintenance processes by using real-time digital tools and data analysis to provide an accurate diagnosis of the state of the machine, motor, or asset being analyzed. The state monitoring is carried out through sensors such as accelerometers for the case of vibration analysis in motors. Many rotating machines vibrate due to the rotational movement of their internal components. Components such as stator, rotor, bearings, and contactors may have faults and be responsible for malfunction or damage to a motor, generating atypical vibrations. Motors are susceptible to noise and abnormal vibrations generating imbalance that can cause problems in the motors such as misalignment, shaft breakage, imbalance, cavitation or turbulence, clearances, bearing problems, and even asset destruction. Therefore, this document presents an alternative solution based on Maintenance 4.0 capable of analyzing, detecting, and predicting damage to rotating machinery through vibrational data analysis using machine learning. In addition, the design of an electronic system for data acquisition is carried out.