This article proposes a new method for bearing acoustic detection based on a parabolic acoustic mirror. The parabolic acoustic mirror is used instead of the traditional bearing end cover, and a directional microphone is placed at the focal point of the parabolic acoustic mirror. The fault sound emitted by the bearing is focused on the directional microphone after being reflected by the parabolic acoustic mirror, achieving signal gain acquisition. Due to the inherent properties of paraboloids, non-parallel incident sound waves cannot converge at the focal point, thus achieving directional acquisition of bearing fault signals. Further, the single channel underdetermined blind source separation algorithm is used to extract the sound signal emitted by the bearing fault point, achieving high-quality acquisition of fault signals. Through experimental comparison and analysis with vibration signals and microphone direct sampling signals, it has been proven that the bearing fault signals collected by the parabolic acoustic mirror have certain advantages compared to the other two methods, indicating the feasibility of collecting bearing fault signals by the parabolic acoustic mirror.
In this paper, a new method for bearing acoustic detection is proposed based on the parabolic acoustic mirror, which uses the parabolic acoustic mirror instead of the traditional bearing end cap, places a directional microphone at the focal point of the parabolic acoustic mirror, and focuses the fault information signal emitted by the bearing at the directional microphone after the reflection of the parabolic acoustic mirror to achieve gain acquisition. At the same time, due to the inherent properties of the parabola, the non-parallel incident sound waves cannot converge at the focal point, so as to realize the directional acquisition of the bearing fault signal. Combined with the support vector machine, the acoustic features are extracted, and the high-dimensional features are reduced to three-dimensional features by using the KJADE algorithm, and they are input into the support vector machine as feature vectors for identification and classification of various bearings. Experiments show that the proposed method can accurately classify bearing faults.
A fault diagnosis model based on Motif Difference Field (MDF) and Swin Transformer is proposed to address the issue of scarce fault samples in actual working conditions, which leads to poor diagnostic and generalization capabilities of Deep Learning based fault diagnosis models. Using MDF instead of Gramian Angle Field (GAF), the one-dimensional signal is transformed into a two-dimensional image, retaining features while performing data augmentation; Using the Swin Transformer network model instead of the CNN network for bearing fault measurement. The results indicate that the model has higher accuracy and generalization compared to other deep learning methods.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.