Classifying formulations of crosslinked polyethylene pipe by applying machine‐learning concepts to infrared spectra

Abstract

Crosslinked polyethylene (PEX-a) pipes are emerging as promising replacements for traditional metal or concrete pipes used for water, gas, and sewage transport. Understanding the relationship between pipe formulation and performance is critical to their proper design and implementation. We have developed a methodology using principal component analysis (PCA) and the machine learning techniques of 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. The application of PCA revealed that a large percentage (89%) of the total variance could be explained by the first three principal components (PC1-PC3), with distinct clustering of the data for each formulation. By examining the contribution of the individual IR

Deep Generative Modeling of Infrared Images Provides Signature of Cracking in Cross-Linked Polyethylene Pipe

“…From the best fit to the sigmoidal function, the center of the hydrolysis transition zone is identified by the inflection point d 0 , and the width of the transition is characterized by the width parameter Δ = 43 μm. This gradient width likely reflects fundamental aspects of the complex hot water-driven secondorder autocatalytic hydrolysis process 3 For the L2 row heat maps in Figure 4f−j, the latent values range from about −0.1 to 0.3 for virgin and extensively oxidized samples, respectively. The regions near the inner wall surface of all samples exposed to in-service conditions exhibited elevated L2 values associated with oxidation, chain scission, and amorphous PE loss.…”

“…These peaks correspond to major products of PE degradation: ketone carbonyl and terminal vinyl groups, respectively. 3,5 Major decreases occur in the 1400−1275 cm −1 region that is associated with the CH 3 bending and CH 2 wagging modes of amorphous PE and weaker decreases at 1080 cm −1 that is associated with the PE backbone C−C stretching modes. 43 Taken together, these changes are consistent with the oxidative degradation and chain scission of amorphous PE components of the PEX-a pipe matrix.…”

Deep Learning and Infrared Spectroscopy: Representation Learning with a β-Variational Autoencoder

“…However, this approach can be difficult to implement for complex heterogeneous systems with overlapping absorption signals from different chemical species, which in turn can limit the interpretation of IR spectra and the applications of IR spectroscopy. To overcome these limitations, multivariate methods such as principal component analysis (PCA) − are often used to extract as much interpretable information as possible from IR spectra of complex heterogeneous systems.…”

“…In this Letter, we have used a deep learning approach to study the spectroscopic and corresponding chemical changes that occur during the aging of cross-linked polyethylene (PEX-a) pipe that is increasingly being used for domestic and industrial water transport and heating. PEX-a pipes, especially those that have experienced in-service conditions or undergone accelerated aging, are complex heterogeneous systems, in which the IR spectra exhibit contributions from the polyethylene matrix, multiple stabilizing additives, and numerous degradation products. ,,− Variance in the IR spectra can be induced by different aging processes, degradation processes, and chemical reactions that occur simultaneously in the pipe during in-service use and accelerated aging. To better understand these processes and identify distinct spectroscopic and chemical changes, we have trained an artificial neural network β-variational autoencoder (β-VAE) on a diverse data set of ∼25 000 IR spectra of virgin, in-service, cracked, and aged pipes (see Supporting Information for the β-VAE model architecture, training details, and reconstruction quality statistics).…”

Evaluation of Performance of Chlorinated Polyethylene Using Wireless Network and Artificial Intelligence Technology

“…Nonetheless, the field of polymer informatics is still in its infancy. Researchers are attempting to improve ML algorithms and to integrate data accumulation and ML algorithms more deeply in specific applications [7]. In this paper, it is proposed using artificial intelligence (AI) in conjunction with wireless network technology to precisely assess the performance of chlorinated polyethylene in polymer research.…”