The Long Beach Convention Center
Long Beach, CA
1:30pm, April 27th
Brian Rohrback
Infometrix, Inc.
Will Warkentin
Chevron Richmond Refinery
KEYWORDS
Best Practices, Calibration, Cloud Computing, Database, Gasoline Blending, Optical Spectroscopy, PLS, Process Control
ABSTRACT
Optical spectroscopy is a favored technology to measure chemistry and is ubiquitous in the hydrocarbon processing industry. In a previous paper, we focused on a generic, machine-learning approach that addressed the primary bottlenecks of mustering data, automating analyzer calibration, and tracking data and model performance over time. The gain in efficiency has been considerable, and the fact that the approach does not disturb any of the legacy (i.e., no changes or alterations to any analyzer or software in place) made deployment simple.
We also standardized a procedure for doing calibrations that, adheres to best practices, archives all data and models, provides ease of access, and delivers the models in any format. What remains is to assess the speed of processing and the quality of the models. To that end a series of calibration experts were tasked with model optimization, restricting the work to selecting the proper samples to include in the computation and setting the number of factors in PLS. The amount of time and the quality of the models were then compared. The automated system performed the work in minutes rather than hours and the quality of the predictions at least matched the best experts and performed significantly better than the average expert. The conclusion is that there is a large amount of recoverable giveaway that can be avoided through automation of this process and the consistency it brings to the PLS model construction.
INTRODUCTION
There is a lot of mundane work tied to the assembly of spectra and laboratory reference values to enable quality calibration work. There is also insufficient guidance when it comes to the model construction task. How much time should be spent on this task? How to best assess whether a spectrum-reference pair is an outlier or not? How many cycles of regression-sample elimination make sense? Where do we switch over from improving the model by adding PLS factors to overfitting and incorporating destabilizing noise?
For more information or the full paper, contact us.
Venue:
March 3-6, 2024
Bethesda North Marriott Hotel and Conference Center
Presented by:
Brian Rohrback, Ph.D., MBA, President, Infometrix, Inc.
Abstract:
Addressing challenges is why we are here. As scientists in our organizations, we are awash in data, but the mustering of these data into their actionable information content is a significant challenge. Clearly, we have tools from the univariate and multivariate statistical trove that help. Relatively recently, artificial intelligence or novel machine learning concepts are being applied with some success. Choosing the tools we apply and using these tools appropriately presents us with another challenge. Noting that the desired property metric will likely be only lightly correlated to the bits of data being assembled is challenging; blending the information extraction process for disparate sources of data adds to the challenge. From manufacturing quality monitoring to personalized medicine, diagnosing the health status needs to be done in real time for us to maintain reasonable control. The first step is to ensure that the input data is as noise-free as possible. Next is to turn data into an information feed that can be combined with other sources. Challenges reappear as we look to mine these data-information streams for a best-fit solution.
Register at www.IFPACglobal.org/attendee-registration.
infometrix.com 