This work provides a detection method for failure in rotating machines based on a change of vibration pattern and offers the diagnosis about the operation conditions using Fuzzy Logic. A mechanic structure (as an experimental prototype where faults can be inserted) called Rotating System has been used. The vibration standard of the Rotating System, called "The Spectral Signature", has been obtained. The changes in the vibration standard have been analyzed and used as parameters for detecting incipient failures, as well as their condition evolution, allowing predictive monitoring and planning of maintenance. The faults analyzed in this work are caused due to insertion of asymmetric masses for unbalancing in the axle wheel. The system for diagnosing Fuzzy System was calibrated to detect and diagnose the conditions: normal, incipient failure, maintenance, and danger, using linguistic variables. The frequency of rotation and the amplitudes of vibration of the axle wheel are considered in each situation as parameters for analysis, diagnostic, for the decision by the Expert System based on Fuzzy rules. The results confirm that the proposed method is useful for detecting incipient failures, monitoring the evolution of severity and offering grants for planning and decision making about maintenance or prevention of rotating machines.
A New Hybrid "Park's Vector - Time Synchronous Averaging" Approach to the Induction Motor-fault Monitoring and Diagnosis
