Masks constructed of a variety of materials are in widespread use due to the COVID-19 pandemic, and people are exposed to chemicals inherent in the masks through inhalation. This work aims to survey commonly available mask materials to provide an overview of potential exposure. A total of 19 mask materials were analyzed using a nontargeted analysis twodimensional gas chromatography (GCxGC)−mass spectrometric (MS) workflow. Traditionally, there has been a lack of GCxGC− MS automated high-throughput screening methods, resulting in trade-offs with throughput and thoroughness. This work addresses the gap by introducing new machine learning software tools for high-throughput screening (Floodlight) and subsequent pattern analysis (Searchlight). A recursive workflow for chemical prioritization suitable for both manual curation and machine learning is introduced as a means of controlling the level of effort and equalizing sample loading while retaining key chemical signatures. Manual curation and machine learning were comparable with the mask materials clustering into three groups. The majority of the chemical signatures could be characterized by chemical class in seven categories: organophosphorus, long chain amides, polyethylene terephthalate oligomers, n-alkanes, olefins, branched alkanes and long-chain organic acids, alcohols, and aldehydes. The olefin, branched alkane, and organophosphorus components were primary contributors to clustering, with the other chemical classes having a significant degree of heterogeneity within the three clusters. Machine learning provided a means of rapidly extracting the key signatures of interest in agreement with the more traditional time-consuming and tedious manual curation process. Some identified signatures associated with plastics and flame retardants are potential toxins, warranting future study to understand the mask exposure route and potential health effects.

No abstract

The timely detection of small leaks from liquid pipelines poses a significant challenge for pipeline operations. One technology considered for continual monitoring is distributed temperature sensing (DTS), which utilizes a fiber-optic cable to provide distributed temperature measurements along a pipeline segment. This measurement technique allows for a high accuracy of temperature determination over long distances. Unexpected deviations in temperature at any given location can indicate various physical changes in the environment, including contact with a heated hydrocarbon due to a pipeline leak. The signals stemming from pipeline leaks may not be significantly greater than the noise in the DTS measurements, so care must be taken to configure the system in a manner that can detect small leaks while rejecting non-leak temperature anomalies. There are many factors that influence the frequency and intensity of the backscattered optical signal. This can result in noise in the fine-grained temperature sensing data. Thus, the DTS system must be tuned to the nominal temperature profile along the pipe segment. This customization allows for significant sensitivity and can utilize different leak detection thresholds at various locations based on normal temperature patterns. However, this segment-specific tuning can require a significant amount of resources and time. Additionally, this configuration exercise may have to be repeated as pipeline operating conditions change over time. Thus, there is a significant need and interest in advancing existing DTS processing techniques to enable the detection of leaks that today go undetected by DTS due to their signal response being too close to the noise floor and/or requiring significant resources to achieve positive results. This paper discusses the recent work focused on using machine learning (ML) techniques to detect leak signatures. Initial proof-of-concept results provide a more robust methodology for detecting leaks and allow for the detection of smaller leaks than are currently detectable by typical DTS systems, with low false alarm rates. A key use of ML approaches is that the system can “learn” about a given pipeline on its own without the need to utilize resources for pipeline segment-specific tuning. The potential to have a self-taught system is a powerful concept, and this paper discusses some key initial findings from applying ML-based techniques to optimize leak detection capabilities of an existing DTS system.

This report documents the result of research conducted by Southwest Research Institute (SwRI�) for the Pipeline Research Council International (PRCI) into the development of a Machine Learning (ML) model for improving the detection of leaks in liquid-carrying pipelines. Operators were surveyed as to their use of CPM systems for leak detection. Several operators provided data to support the research. The data was collected, curated, and analyzed by SwRI. Several ML models were investigated. A framework was developed to allow operators to use their own data to generate ML models for their pipelines to improve leak detection. A guideline was provided to facilitate use of the framework by operators. This document has been updated based on PRCI QC, committee, and PHSMA comments.