A Data-Driven Multivariate Process Monitoring Platform For Knowledge Discovery And Model Building In Industrial Applications
Semester of Graduation
Master of Science in Chemical Engineering (MSChE)
The Gordon A. and Mary Cain Department of Chemical Engineering
In industrial chemical manufacturing processes, the amount of raw data generated can add complexity in the analysis and understanding of the process dynamics. Being able to properly interpret this data can help improve plant operation, especially regarding safety and profitability. This research has culminated in FastMan-JMP, a platform proposed for monitoring of industrial processes and optimization of the offline data-driven model-building process as part of the process monitoring workflow. FastMan-JMP is a tool developed in Python to apply various data mining and machine learning techniques quickly and easily to better understand valuable patterns and hidden trends in process data. One of these important trends of particular interest is that of global and local structures in data, how they relate to the process operation, and how they are preserved and analyzed using the available data mining methods.
It is shown that local and global structures in the data set can be visualized and related to actual process operations though the identification of the process variables responsible for the division between any particular clusters. Additionally, comparisons between different data mining methods can be difficult, due to different algorithms’ unique structures and parameters. One of the main aims of this research has been to decipher these algorithms, how they perform in the context of industrial data, and to develop this tool, FastMan-JMP, to help others do the same. Results are discussed for several industrial case studies to illustrate the applicability of this work to real-world scenarios.
Seghers, Estelle E., "A Data-Driven Multivariate Process Monitoring Platform For Knowledge Discovery And Model Building In Industrial Applications" (2023). LSU Master's Theses. 5707.
Available for download on Saturday, March 16, 2024