CNN Based Ensemble Approach for Malfunction Detection from Machine Sounds
Abstract
Together with the meaning and essence of data for the company nowadays; The variety of data also differs. One of these differentiating data types is sound. Borusan Makina ve Güç Sistemleri A.Ş. the data obtained from the Caterpillar construction machines of. The machine sound gives clues about many malfunctions. Artificial intelligence systems of the heard sound will be integrated into business processes. Every tone can be converted. With this, the properties and estimates of the sound grids are used. In this direction; While the incident is getting in the way of his business, an unfortunate project occurs with a similar visitor. The traditional will use a meaningful method by listening to the producer's sound and technology and innovation to develop easy blueprints of decisions that cannot be diverted to sound data. Thanks to the real-time model with short-term audio recording, it is instantly predicted whether there is a problem in the machine. Free from personal and technical comments; By examining the patterns of sound waves, it is aimed to be made without cancellation.
References
- 1.Jena DP, Panigrahi SN. Motorbike piston-bore fault identification from engine noise signature analysis. J Appl Acoust 2014; 76: 35-47.
- 2.Fitzgerald, Derry “Harmonic/percussive separation using median filtering,” 13th International Conference on Digital Audio Effects (DAFX10), Graz, Austria, 2010.
- 3.Tan, Mingxing, and Quoc Le. “Efficientnet: Rethinking model scaling for convolutional neural networks,” International Conference on Machine Learning, PMLR, 2019.
- 4.Yang, Ruo-Yu, and Rahul Rai. “Machine auscultation: enabling machine diagnostics using convolutional neural networks and large-scale machine audio data,” Advances in Manufacturing 7.2, 174-187, 2019.
- 5.Chen, Tianqi, and Carlos Guestrin. "Xgboost: A scalable tree boosting system," Proceedings of the 22nd ACM signed international conference on knowledge discovery and data mining. 2016.
Akca, E., Özçay, T., Dinç, Y., Yalçı, N., Taşabat, S. E., Varol, M. A., Kayı, B., Öğütcen, M. Y., Öztürk, B. (2022). CNN Based Ensemble Approach for Malfunction Detection from Machine Sounds. *The European Journal of Research and Development*, 2(2), 411 - 420. https://doi.org/10.56038/ejrnd.v2i2.37
Bibliographic Info