Abstract
The development of condition monitoring in continuous-time mass production is a traditional but critical topic owing to its role in increasing cost-effectiveness. A new method for intelligent manufacturing processes monitor was presented using the newly developed learning approach, Support Vector Regression, and related noise density model for efficient and effective modeling. A new similarity measurement method was introduced resulting in a real-time system developed. The system was applied for certain manufacturing process condition monitoring. The experimental results showed that the proposed method is substantially more effective than others. In addition, the new monitor requires only normal condition samples for implementation, an attractive feature for shop floor applications. The novel system ensures the automatic and intelligent operation, and it has potential applications on the other manufacturing processes.
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Editor information
Rights and permissions
About this paper
Cite this paper
Ge, M., Xu, Y. A Novel Intelligent Monitor for Manufacturing Processes. In: Tarn, TJ., Zhou, C., Chen, SB. (eds) Robotic Welding, Intelligence and Automation. Lecture Notes in Control and Information Science, vol 299. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-44415-2_5
Download citation
DOI: https://doi.org/10.1007/978-3-540-44415-2_5
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20804-4
Online ISBN: 978-3-540-44415-2
eBook Packages: Springer Book Archive