397515 Process Learning Data Pipeline for Dealing with the Big Data Challenges

Wednesday, April 29, 2015: 1:30 PM
12B (Austin Convention Center)
Shu Xu1, Willy Wojsznis2, Mark Nixon3, Michael Baldea1 and Thomas F. Edgar1, (1)McKetta Department of Chemical Engineering, The University of Texas at Austin, Austin, TX, (2)Innovation Center, Emerson Process Management, Austin, TX, (3)Process Management, Emerson, Austin, TX

To uncover the valuable information from a large amount of process data stored in databases, a process learning data pipeline consisting of transformation, cleaning, re-sampling and exploratory data analysis (EDA) units is built using Python with a graphical user interface (GUI). The pipeline incorporates techniques and knowledge from various fields, such as statistics, computer science, and signal processing. A test case is provided based on industrial data set, and by analyzing the results coming from the pipeline, a better understanding of the process is obtained.

Extended Abstract: File Uploaded
See more of this Session: Smart Manufacturing Aspects of Big Data
See more of this Group/Topical: Big Data Analytics