A new computational method for stator faults recognition in induction machines based on hyper-volumes

摘要: To remain competitive, the manufacturing industry is always innovating and developing new cost-efficient ways to produce goods. That why today, extensive automation applied in nearly every type of assembly processes. Automation improves productivity, quality robustness products. It also increases predictability production lines mainly constituted asynchronous machines. These machines, however, need regular maintenance. Time-based maintenance labor-intensive, ineffective identifying problems that develop between scheduled inspections, not cost-effective. For these reasons, researchers companies are now investigating methods what called preventive involves use sensors (vibrations, load cells, electrical, etc.) placed on machine monitor its actual state order detect engine failures. some years, works presenting interesting results [1-35] have been published, but few investigated effective capable clearly characterize importance In this paper, we propose a computational approach for detection characterization stator faults machines based electrical signal analysis. Our method able detect, locate, quantify severity failure. do so, frequency characteristics [6, 7] simple detection, currents [6] performance speed induction localization quantification Moreover, exploit hyper-volumes model defective We present an experiment conducted which shows promising results.