Nuevos modelos para la Caracterización, Detección y Diagnóstico de Fallas en Máquinas Eléctricas Rotativas
Identifying faults in electric motors is crucial for optimizing their efficiency and preventing failures. Monitoring and inspecting critical systems improves availability and operational reliability, ensuring personnel safety, environmental compliance, and legal compliance, reducing costs in manufa...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Fundación de Estudios Superiores Comfanorte
2023-07-01
|
| Series: | Mundo Fesc |
| Subjects: | |
| Online Access: | https://www.fesc.edu.co/Revistas/OJS/index.php/mundofesc/article/view/1656 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Identifying faults in electric motors is crucial for optimizing their efficiency and preventing failures. Monitoring and inspecting critical systems improves availability and operational reliability, ensuring personnel safety, environmental compliance, and legal compliance, reducing costs in manufacturing and business operations.
Physical methods and vibration analysis are currently used to detect problems, although these approaches present limitations in both accuracy and efficiency due to the subjectivity of each case and the loss of information in each case. The main objective is to explore and propose new methodologies for characterizing faults in electric motors through the use of data analysis and machine learning techniques, reviewing traditional and current methods for fault diagnosis, including vibration analysis techniques and supervised and unsupervised learning algorithms.
The potential of artificial intelligence algorithms to improve diagnosis and reduce subjectivity in data processing is also examined. Therefore, modern techniques based on machine learning offer significant improvements in fault detection and prediction, allowing the identification of complex patterns and more precise diagnoses, expanding the capabilities of conventional approaches to facilitate predictive maintenance.
Therefore, new methodologies based on data analysis and machine learning represent a breakthrough in fault diagnosis in electric motors. Their implementation in the industry can reduce maintenance costs, optimize motor operation, and prevent unexpected failures. |
|---|---|
| ISSN: | 2216-0353 2216-0388 |