Development and Implementation of a Vibration Signal Collection and Analysis System

The main objective of this study is to demonstrate the relationship between the behavior of rotating machines and vibration analysis, with an emphasis on fault identification based on the signal spectrum in the frequency domain. Vibration analysis, a technique widely used in predictive maintenance, enables continuous monitoring of equipment operating conditions and early detection of issues that may compromise performance. The methods applied in this study include the collection of vibration data from electric motors and generators, followed by processing these data using spectral analysis techniques. The development of a Python code was essential for real-time signal input and processing, facilitating result visualization and enabling a more accurate diagnosis of machine conditions. The ISO 2372 standard was considered in determining characteristic frequencies associated with different types of faults. The final considerations highlight the importance of integrating advanced signal processing methods with traditional vibration analysis techniques. This combination allows for more detailed and effective monitoring of machine operating conditions, contributing to the reduction of unplanned downtime and the optimization of maintenance processes. The study also reinforces the relevance of predictive maintenance in the current industrial context, where equipment reliability is crucial for operational competitiveness and sustainability.