Machinery Fault Diagnosis Using Cyclostationary Statistics and Quad Phase Minimum Average Correlation Energy Filters

摘要

This paper discussed about the implementation of automated machine condition monitoring using the cyclostationary statistics and Quad Phase Minimum Average Correlation Energy Filter for fault classification. The machine health identification is based on the comparison of acquired real-time vibration features with template features. A benchmark vibration data set which was collected from a U.S. Navy CH-46E helicopter aft gearbox by a British helicopter manufacturer known as Westland has been used to demonstrate the classification problem. The cylostationary statistics has been utilized in feature extraction of the vibration data. Encouraging and successful results have been obtained and presented in this paper.