Autonomous Calibration and the Impact on Process Analysis and Control – The use of multivariate analysis in processing spectroscopic data is far from new; the processing of NIR data in petroleum refineries dates to the early 1980s and in the food industry well before that. Although the computers have improved in performance leading to speed ups in the calibration process, the procedures being followed have not changed significantly since the 1980s. Intriguingly, we have made decisions on the corporate level that work against each other. We are installing more spectrometers and at the same time we are reducing staffing for spectrometer calibration and maintenance. A change in approach is mandated. In the spirit of automation, there are tools from the general statistics realm that can be applied to simplify the work involved in optimizing a calibration through selection of samples. Robust techniques require some set-up of parameters, but once the parameters are established for an application, they are often usable in every other instance of that application. The result is a one-pass means of selecting optimal samples for a calibration problem and in turn simplifies the assignment of model rank. This approach is completely independent of hardware configuration and can be used with any software – plus it solves two problems: It is a selection process that can be completely automated; and It is objective and does not rely on the relative skill of a specific analyst. The ultimate goal is to integrate spectroscopic measurements in a process setting with the same simplicity-of-effort with which we install temperature sensors. Presented by Brian Rohrback, Infometrix, Inc.
Re-engineering Calibration for Process Spectrometers – This is a report of a multi-company, multi-industry, hydrocarbon processing consortium, established six years ago, to re-evaluate how the calibration process for analyzers could be managed more efficiently. The first focus was optical spectroscopy, an increasingly important source of process chemistry knowledge due to its advantage of speed, sensitivity, and simple safety requirements. As one of very few analyzer technologies that can measure chemistry, spectroscopy has become a workhorse in the chemical, petrochemical, and petroleum industries. But, even as the number of optical systems is continuing to increase, companies have been decreasing the number of employees who are tasked with their management. As a result, a paradigm shift is required for industry to adapt to a higher workload combined with changes in the fitness level and longevity of the technicians responsible for installation, calibration, and maintenance. What if we could put a spectroscopy system in place and have it handle the application and communicate results as soon as it is turned on? Then, if performance does not match legacy standards, the system dials itself in or calls for help. The systematic approach discussed is not constrained by the brand of hardware or by the software vendor and, as such, the approach can be used to manage any new or any in-place system. Presented by Brian Rohrback, Infometrix, Inc.
Multivariate Analysis for Inferentials and Analyzers This course examines a series of algorithmic approaches with the goal of streamlining multivariate model construction to make the analyzers significantly more robust when put into routine practice. This class will benefit those who want to better understand the tools available to build qualitative and quantitative inferentials for their process. It will also benefit technicians who perform both routine and irregular maintenance of chromatographic and spectroscopic instruments in both process and laboratory settings. Managing pressure, temperature, flow and level as an ensemble rather than as independent measurements to track will be highlighted. Attention will also be aimed at optical spectrometers, in particular near infrared and Raman. Chromatographic applications will be discussed with the purpose of simplifying maintenance. These topics are designed to cover the best practices of multivariate technology and will prepare participants for tackling a vast array of problems. This course is useful for any chemist or engineer who is concerned about optimizing their analytical methods to get the most out of their process operations. This course will be presented by Brian Rohrback, president of Infometrix, Inc.
Infometrix now offers Ai-Metrix; its new simplified calibration service to save you time and resources. With efficiency in automation, intelligent data management and speedy calibration, ease burden of maintenance, simplify calibration, and improve model quality. See details on our Ai-Metrix page.
Infometrix will be attending the 2018 Gulf Coast Conference in October. See abstracts below for the 2018 presentations at the GCC conference by Infometrix. Join us or contact us for more information.
Abstract # 169 – Paper
10/16/2018 – 1:20 PM – 1:40 PM – Tulip Room
Optimizing Gas Chromatography
The heartbeat of the process environment is in the data we collect, but we are not always efficient in translating our data streams into actionable information. The richest source of hydrocarbon process information comes from spectrometers and chromatographs and, for many applications, gas chromatography is the cheapest, most adaptable, and most reliable technology available. We can use tech borrowed from other fields to provide more consistent and objective GC results, automate translation of the raw traces into real-time information streams, and create databases that can be used across plant sites or even across industries.
Abstract # 145 – Paper
10/16/2018 – 10:45 AM – 11:05 AM – Tulip Room
Automating Spectroscopy Calibrations
– Infometrix, Inc.
A consortium of companies has undertaken a project to reduce the effort devoted to producing, maintaining, and stabilizing optical spectroscopy performance in routine quality assessment. Over a five-year period, the group has examined an unprecedented historical collection of spectra from multiple spectrometers spanning 1-5 years from sixteen oil refineries, with the goal of developing and maintaining stable models for long-term deployment. Even though the data are tied to petroleum work, the lessons learned are true for all applications. The technologies utilized follow a pattern of best practices, including the use of Robust outlier diagnostics, local and hierarchical modeling, and model augmentation. The effort has resulted in significant progress towards automation of model creation, stability, and maintenance in an industrial process. Additionally, we will share a comparison of prediction capabilities of different spectroscopy technologies and form-factors that can lead to a significant reduction in deployment and maintenance cost.