Análise de falhas em caixa multiplicadora de aerogeradores usando transformada wavelet contínua

Abstract

Wind power is one of the oldest-exploited energy sources by humans and today is the most seasoned and efficient energy of all renewable energies. To make wind power more competitive with other sources, characteristics as availability, reliability and the life of turbines will all need to be improved. Therefore, reducing inspection and maintenance costs is vital to maintaining the competitiveness of this industry. The maintenance applied to wind turbines implies high costs and long periods without energy production. Vibration and oil signal analysis are frequently used in predictive maintenance for diagnosing mechanical component failures as they contain information about the condition of their components. Vibration signals from a gearbox are often noisy, making it difficult to extract information from signal components, especially in practical situations. One of the solutions to this problem is the application of signal processing techniques to make getting information easier. The present thesis aimed to apply the Continuous Wavelet Transform – CWT, in the extraction of characteristics and subsequent diagnosis of wear on gears. In this way, signals obtained in a developed experimental bench, which consists of a motor coupled in pairs of gear. Oil analysis and optical microscopy were used to monitor oil temperature and its composition, as well as the type of wear that occurred on the tooth flank. The CWT proved to be effective in the analysis of the vibration signal indicating the presence of wear due to contact fatigue (scuffing and pitting), which was corroborated by the analysis of oil and microscopy.