Analisador LAMAR

Maintaining high reliability standards in rotating machines, widely used in industry, can be a crucial competitive differentiator. In this scenario, vibration analysis stands out as a technique capable of offering this reliability, promoting business growth. The application of predictive maintenance allows you to identify potential problems and plan interventions to correct failures. Vibration measurement is usually done with accelerometer sensors, and most smartphones already have this sensor integrated into their hardware. Thus, considering the increasing use of smartphones, their application in detecting faults in rotating machines can popularize vibration analysis.
However, the literature still barely explores the use of smartphones in vibration analysis, especially to detect faults in rotating machines. This research aims to fill this gap, exploring the potential of smartphones in this field. The objective is also to help industry professionals improve the reliability of their machines with a low-cost device.
To this end, the "LAMAR Analyzer" application was developed for the Android system, making it possible to test the viability of smartphones in detecting faults. With this application, vibration signals were collected at a sampling rate of 500Hz from an electric motor under six different conditions, varying load and criticality. Using 70% of the samples for training and 30% for testing, the results showed a convergence rate of 40.6% and an accuracy of 96.6% in predicting the engine condition.
The proposed solution presents promising results, with potential for improvements, and paves the way for the development of future solutions that enable the use of smartphones to identify faults in rotating machines.