Real-Time Chemical Process Monitoring with UMAP
Document Type
Article
Publication Date
1-1-2021
Abstract
In the modern chemical world, engineers have access to millions of data points at their fingertips. Using this data properly can help us to recognize vast improvements in the way plants are run, especially in process monitoring. In this paper three aspects of process monitoring are considered: visualization, fault identification, and fault diagnosis. For each aspect, the use of modern machine learning techniques for addressing these issues is discussed, and improvements over outdated methods are illustrated. This proposed approach is tested using the Tennessee Eastman Process (TEP) for several types of faults.
Publication Source (Journal or Book title)
Computer Aided Chemical Engineering
First Page
2077
Last Page
2082
Recommended Citation
webb, Z., & Romagnoli, J. (2021). Real-Time Chemical Process Monitoring with UMAP. Computer Aided Chemical Engineering, 50, 2077-2082. https://doi.org/10.1016/B978-0-323-88506-5.50321-1