MSc Thesis Presentation: Classifying Formulations and Tracking Accelerated Aging of Crosslinked Polyethylene Pipes by Applying Machine Learning Concepts to Infrared Spectra

Candidate

Melanie Hiles

Abstract

Crosslinked polyethylene (PEX-a) pipes are emerging as promising replacements for traditional metal or concrete pipes used for water, gas, and sewage transport. However, PEX-a is susceptible to oxidative degradation during its manufacturing and end-use applications that can ultimately result in premature pipe failure. Therefore, understanding the relationship between pipe formulation and performance is critical to their proper design and implementation. We have developed a methodology using the machine learning techniques of principal component analysis (PCA), k-means clustering and support vector machines (SVM) to compare and classify different PEX-a pipe formulations based on characteristic infrared (IR) spectroscopy absorbance peaks. PEX-a pipes of one such formulation was subjected to accelerated aging under different conditions to examine the effect of external stresses on changes to the chemical composition of the pipe samples. PCA and Decision Tree (DT) based techniques were used to identify characteristic IR signatures related to aging of PEX-a pipes.

Examination Committee

• Dr. Ralf Gellert, Chair
• Dr. John Dutcher, Advisor
• Dr. Robert Wickham