An image recognition method for gear fault diagnosis in the manufacturing line of short filament fibres

Abstract

The manufacturing line is a fundamental element in short filament fibre production, in which the gearbox is the key
\nmechanical part. Any faults in the gearbox will greatly affect the quality o f the short filament fibres. However, due to
\nthe harsh working environment, the gearbox is vulnerable to failure. Due to the complexity o f the manufacturing line,
\neffective and efficient feature extraction o f gear faults is still a challenge. To this end, a new fault diagnosis method based
\non image recognition is proposed in this paper for gear fault detection in fibre manufacturing lines. In this method,
\nwavelet packet bispectrum analysis (WPBA) is proposed to process the gear vibration signals. The bispectrum texture is
\nobtained and then analysed by an image fusion algorithm for texture feature extraction. The grey-level co-occurrence
\nmatrix is used in the image fusion and the extracted texture features are four parameters o f the grey-level co-occurrence
\nmatrix. Finally, a support vector machine (SVM) is adapted to recognise the gear fault type and location. Experimental
\ndata acquired from a real-world manufacturing line o f short filament fibres are used to evaluate the performance o f the
\nproposed image-based gear fault detection method. The analysis results demonstrate that the newly proposed method is
\ncapable o f accurate gear fault detection in fibre manufacturing lines.