Abstract:
The widespread use of asynchronous motors is primarily attributed to their superiority over other motor types. They are notably cost-effective when it comes to maintenance. As such, it is essential to implement robust protective measures against major faults and to establish an effective functional monitoring system incorporating efficient preventive maintenance strategies. Predictive maintenance, in particular, can significantly enhance the cost-effectiveness and reliability of installations. The machine’s behavior evolves under varying fault conditions, making accurate fault identification critical and requiring a detailed and thorough analysis of the motor. In this context, the objective of our work is to diagnose faults that affect the motor throughout its operational lifespan. This is achieved using a model based on the detailed calculation of all the machine’s inductances. Among the most common faults is bearing failure, which is a key focus for devising an optimal maintenance strategy. Therefore, it is crucial to identify specific fault signatures that enable the prediction of fault progression. By analyzing various signals, we aim to extract reliable indicators and signatures that can be used not only for accurate fault prediction but also in industrial applications. To validate our analytical developments, experimental validation is indispensable. This step serves to confirm the accuracy of the simulated results for the various faults studied. To achieve this, test benches were employed to corroborate the findings derived from analytical calculations. Ultimately, the results obtained are highly indicative, particularly for faults such as power supply imbalances and bearing failures, and show strong alignment with findings reported in the literature.
