Journal
EXPERIMENTAL TECHNIQUES
Volume 43, Issue 5, Pages 635-643Publisher
SPRINGER
DOI: 10.1007/s40799-019-00324-0
Keywords
Condition monitoring; Sensor fusion; KNN; Bearing defects; Vibration-based sensor
Funding
- Aeronautics Research and Development Board (ARDB), India
Ask authors/readers for more resources
This paper presents the novel technique for fault diagnosis of bearing by fusion of two different sensors: Vibration based and acoustic emission-based sensor. The diagnosis process involves the following steps: Data Acquisition and signal processing, Feature extraction, Classification of features, High-level data fusion and Decision making. Experiments are carried out upon test bearings with a fusion of sensors to obtain signals in time domain. Then, signal indicators for each signal have been calculated. Classifier called K-nearest neighbor (KNN) has been used for classification of fault conditions. Then, high-level sensor fusion was carried out to gain useful data for fault classification. The decision-making step allows understanding that vibration-based sensors are helpful in detecting inner race and outer race defect whereas the acoustic-based sensor is more useful for ball defects detection. These studies based on fusion helps to detect all the faults of rolling bearing at an early stage.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available