Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/87527
- Development of online sensors for bubble stirred vessels
- Xu, Xiaodong; Brooks, Geoffrey; Yang, William
- Bubbling is commonly used in pyrometallurgieal vessels to enhance mass and heal transfer. These processes can be difficult to monitor because of the severe operating conditions associated with these processes. In this paper, we described the development of techniques for simultaneously analysing image, sound and vibration signals associated with bubbling in vessels using multivariate statistical analysis to form new 'latent' variables from the combined signals to control the process. This approach has been tested over a range of conditions for a bottom stirred cold model with two liquid layers (to simulate metal and slag). PCA (Principal Component Analysis) results showed that the image, sound and vibration signals are highly correlated and they can be reduced to one combined signal which describes most of the variation of the process. It was found that the combined signals from sound and vibration are also highly correlated and there was a clear linear relationship between this combined signal and the stirring power. Therefore, it should be possible to monitor the stirring process by this combined signal from sound and vibration measurement.
- Publication type
- Conference paper
- Research centre
- Swinburne University of Technology. Faculty of Engineering and Industrial Sciences
- Proceedings of the TMS 2010 Annual Meeting and Exhbition: Jim Evans Honorary Symposium, Seattle, Washington, United States, 14-18 February 2010 / B. Li, B. Thomas, L. Zhang, F. Doyle and A. Campbell (eds.), pp. 173-182
- Publication year
- Bubbling phenomena; Ladle metallurgy; Online sensors; PCA; Principal component analysis; Process control; Pyrometallurgy
- John Wiley & Sons
- 9780873397506, 0873397509
- Publisher URL
- Copyright © 2010.
- Peer reviewed