作者: Adam Glowacz , Zygfryd Glowacz
DOI: 10.1016/J.APACOUST.2016.10.012
关键词:
摘要: Abstract An early diagnosis of faults prevents financial loss and downtimes in the industry. In this paper authors presented fault diagnostic technique stator single-phase induction motor. The proposed was based on recognition acoustic signals. measured analysed 3 states motor: a healthy motor, motor with shorted coils auxiliary winding, winding main winding. an original method feature extraction called MSAF-RATIO30-MULTIEXPANDED (Method Selection Amplitudes Frequency - Ratio 30% maximum amplitude Multiexpanded) described. This used to form vectors. A classification obtained vectors performed by KNN (K-Nearest Neighbour classifier), K-Means clustering Linear Perceptron. can find application for protection motors. It be also other rotating electrical machines.