Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Development of inferential measurements using neural networks.

S Bhartiya1, J R Whiteley

  • 1School of Chemical Engineering, Oklahoma State University, Stillwater 74078-5021, USA.

ISA Transactions
|October 2, 2001
PubMed
Summary

This study presents a systematic neural network approach for developing nonlinear correlations to infer key industrial process variables when direct measurements are infrequent. The method uses statistical analysis to select optimal inputs, enabling better process control and optimization.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Results of mapping patients expectation using SERVQUAL.

Journal of healthcare quality research·2020
Same author

Co oxide nanostructures for electrocatalytic water-oxidation: effects of dimensionality and related properties.

Nanoscale·2018
Same author

Trypan blue staining of filtering bleb in eyes with operate trabeculectomy.

Nepalese journal of ophthalmology : a biannual peer-reviewed academic journal of the Nepal Ophthalmic Society : NEPJOPH·2012
Same author

Growth-related model of the GAL system in Saccharomyces cerevisiae predicts behaviour of several mutant strains.

IET systems biology·2012
Same author

Multiple delivery cesium oven system for negative ion sources.

The Review of scientific instruments·2012
Same author

System-level analysis of tryptophan regulation in Escherichia coli--performance under starved and well-fed conditions.

IET systems biology·2007

Area of Science:

  • Chemical Engineering
  • Process Control
  • Data Science

Background:

  • Industrial processes often rely on infrequent off-line laboratory measurements for critical variable control.
  • Advanced process control and optimization necessitate real-time inferred measurements derived from correlations.
  • Complex industrial processes lack readily available correlations and optimal input variable identification.

Purpose of the Study:

  • To introduce a systematic methodology for developing nonlinear correlations for inferential measurements.
  • To leverage neural networks for creating robust inferential models in complex industrial settings.
  • To demonstrate the practical application of this approach in inferring a specific petroleum product quality parameter.

Main Methods:

  • A three-step procedure involving data collection, preprocessing, and statistical analysis for input variable selection.

Related Experiment Videos

  • Utilizing neural networks to generate nonlinear correlations for inferential measurements.
  • Applying the developed methodology to infer the ASTM 95% endpoint of a petroleum product.

    • Main Results:

    • Successfully developed and demonstrated a systematic approach for inferential measurement using neural networks.
    • Identified a subset of process variables through statistical analysis for effective correlation development.
    • Validated the methodology by accurately inferring the ASTM 95% endpoint in a refinery setting.

    Conclusions:

    • The proposed systematic approach provides an effective means to develop nonlinear inferential measurement correlations.
    • Neural networks are a powerful tool for modeling complex industrial processes where traditional methods fall short.
    • This methodology enhances process understanding and control by enabling the inference of critical quality parameters.