Gulf Coast Conference 2024: Previewing Two Oral Sessions on Machine Learning and Instrument Calibration & Hydraulic Fracturing Fluids
On October 16th from 9:30 am – 9:55 am, Brian Rohrback of Infometrix, Inc. will give a talk about the impact of machine learning on spectrometer calibration and Jesus Acapulco of Analytik Jena will give a talk on using inductively coupled plasma–optical emission spectroscopy (ICP-OES) to determine the composition of water used in fracking. We preview both sessions here.
Improvements in spectroscopic instrumentation have allowed analysts to expand the capabilities of the technology at their disposal. Whether it means conducting their analysis more efficiently or ensuring they are getting access to more data and more information, spectroscopic instrumentation has contributed positively to realizing both these goals.
The advent of artificial intelligence (AI) and machine learning (ML) has propelled spectroscopic instrumentation to new heights. AI systems are programmed to compare measured and computer-generated spectra, which allows analysts to obtain more data from spectral analysis experiments (1). ML, meanwhile, as a subfield of AI technologies, can extract and summarize large, complex data sets for both chromatography and spectroscopy applications (1). Because analysts are realizing that manual data analysis is increasingly becoming ineffective in diagnostic and biomedical work, AI and ML has helped bridge the gap in instrumentation to help analysts perform the work needed in these fields (1).
Recent studies have demonstrated how AI and ML have influenced Raman spectroscopy-based classification applications (2,3). For example, ML algorithms can rapidly process and decode complex Raman spectra, enabling faster, more accurate identification of chemical compounds (3). This is particularly beneficial in fields like pharmaceuticals, materials science, and biological research, where large datasets need to be analyzed efficiently (3). AI also improves peak detection, background noise reduction, and calibration processes, leading to more precise measurements (2,3). By automating these tasks, AI-powered Raman spectroscopy instruments offer improved sensitivity, real-time monitoring, and faster insights, significantly advancing both research and industrial applications.
At the Gulf Coast Conference in Galveston, Texas, Dr. Brian Rohrback will be exploring how AI and ML has been applied to laboratory instrumentation and process analyzers. Rohrback is the President of Infometrix, a company that is known for using these new technologies and applying them to their products (4). The new technologies that Infometrix has developed for the past four decades were primarily designed for the hydrocarbon processing industry (4).
Galveston Texas from the air | Image Credit: © eric - stock.adobe.com.
In his talk, Rohrback will talk about the inevitability of AI becoming routine in chromatography and spectroscopy. He will address in his talk how advancements in AI and ML will create more benefits that will outweigh the costs and drawbacks in applying these techniques in normal workflows (5). The talk will also discuss how ML enhances and improves spectrometer output interpretation and information flow, offering practical guidance to navigate these fields (5). Rohrback’s talk will highlight how the main objective with AI and ML is to achieve full automation for enhanced operational efficiency, freeing up analysts for other tasks.
Meanwhile, another talk led by Jesus Acapulco of Analytik Jena will take place, titled “The Analysis of Foresh and Produced Waters in Hydraulic Fracturing Fluids with the PlasmaQuant 9100 Elite ICP-OES.” Acapulco’s talk will focus on showing how the Analytik Jena PlasmaQuant 9100 Elite (PQ9100E) ICP-OES, equipped with an ASX-560 autosampler and salt kit, demonstrated high precision (<3% RSD) and accuracy (within ±20%) for measuring trace to major elements in fluids from the Permian Basin, Texas (6). This talk will demonstrate how this system can be used in various applications for both fracking fluid analysis and other industries requiring complex chemical analysis (6).
References
- Workman, Jr., J. Artificial Intelligence in Analytical Spectroscopy, Part II: Examples in Spectroscopy. Spectroscopy 2023, 38 (6), 10–15. DOI: 10.56530/spectroscopy.js8781e3
- Houhou, R.; Bocklitz, T. Trends in Artificial Intelligence, Machine Learning, and Chemometrics Applied to Chemical Data. Anal. Sci. Adv. 2021, 2 (3–4), 128–141. DOI: 10.1002/ansa.202000162
- Qi, Y.; Hu, D.; Jiang, Y.; et al. Recent Progresses in Machine Learning Assisted Raman Spectroscopy. Adv. Opt. Mater. 2023, 11 (14), 2203104. DOI: 10.1002/adom.202203104
- Gulf Coast Conference, 2024 – Brian Rohrback. Gulf Coast Conference. Available at: https://www.gulfcoastconference.com/speakerBio.php?pid=7849&y=2024&r=a (accessed 2024-09-30).
- Gulf Coast Conference, Presentation Profile. Gulf Coast Conference. Available at: https://www.gulfcoastconference.com/presentationProfile.php?pid=7849&y=2024&r=bio&rid=7849 (accessed 2024-09-30).
- Gulf Coast Conference, Presentation Profile. Gulf Coast Conference. Available at: https://www.gulfcoastconference.com/presentationProfile.php?pid=7862&y=2024&r=bio&rid=7862 (accessed 2024-09-30).
