Análise de vibrações para a investigação de falhas mecânicas em grupos hidrogeradores

Abstract

In the current scenario where the Brazilian electrical matrix is still heavily dependent on hydroelectric generation, accounting for approximately 61.9% of energy generated in 2022, actions aimed at increasing the availability of generating units (UG) and improving efficiency become crucial. Investment in vibration control and monitoring is fundamental. This work aims to contribute to advancing knowledge in the field of vibration analysis applied to the investigation of failures in UG. The central focus is on using vibration analysis techniques as a tool to investigate mechanically induced failures in hydroelectric plants. Vibrations may result from mechanical, electrical defects, as well as hydraulic excitations leading to different modes of excitation directly impacting machine availability and compromising power generation. The paper covers signal processing techniques, the main instruments used in vibration monitoring, and the most common failures in generating units. In this work, detailed attention is given to the main mechanical failures that result in vibrations, presenting the characteristic behavior of the machine and the interpretation of defects in relation to the vibration spectrum. The equipment used for tests, configurations, and inputs of the vibration analysis system (SDAV) for data collection are also specified in detail, along with the entire procedure for machine balancing. The results of the vibration data acquisition from the analyzed generating unit, the case study of this work, initially indicated severe unbalance with an amplitude of 5.07 mm/s, well above the limit allowed by the NRB 10,816 standard. Following the balancing procedure, a significant improvement in vibration was observed, reaching levels of 1.48 mm/s, allowing the machine to operate within the A zone limit, which defines a tolerance of up to 1.6 mm/s for this range. In conclusion, the importance of the work lies in supporting the investigation of defects through vibration analysis in machines, adopting a predictive approach to avoid major downtimes for corrective maintenance and losses due to equipment breakdowns. In this study, we emphasize that the use of an affordable portable device proves effective in collecting and analyzing vibration data. We conclude that it is essential to conduct preventive vibration monitoring in Generating Units (UG) to enhance operational safety.