skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Integrating Physical Modeling, Neural Computing, and Statistical Analysis for On-Line Process Monitoring

Abstract

Two basic approaches can be mentioned to model physical systems. One approach derives a model structure from the known physical laws. However, obtaining a model with the required fidelity may be difficult if the system is not well understood. A second approach is to employ a black-box structure to learn the implicit input-output relationships from measurements in which no particular attention is paid to modeling the underlying processes. A method that draws on the respective strengths of each of these two approaches is described. The technique integrates known first-principles knowledge derived from physical modeling with measured input-output mappings derived from neural processing to produce a computer model of a dynamical process. The technique is used to detect operational changes of mechanical equipment by statistically comparing, using a likelihood test, the predicted model output for the given measured input with the actual process output. Experimental results with a peristaltic pump are presented.

Authors:
;  [1]
  1. Argonne National Laboratory (United States)
Publication Date:
OSTI Identifier:
20826820
Resource Type:
Journal Article
Journal Name:
Nuclear Technology
Additional Journal Information:
Journal Volume: 141; Journal Issue: 1; Other Information: Copyright (c) 2006 American Nuclear Society (ANS), United States, All rights reserved. http://epubs.ans.org/; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0029-5450
Country of Publication:
United States
Language:
English
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; COMPUTERS; EQUIPMENT; MAPPING; MONITORING; NEURAL NETWORKS; ON-LINE SYSTEMS; SIMULATION

Citation Formats

Garcia, Humberto E, and Vilim, Richard B. Integrating Physical Modeling, Neural Computing, and Statistical Analysis for On-Line Process Monitoring. United States: N. p., 2003. Web.
Garcia, Humberto E, & Vilim, Richard B. Integrating Physical Modeling, Neural Computing, and Statistical Analysis for On-Line Process Monitoring. United States.
Garcia, Humberto E, and Vilim, Richard B. Wed . "Integrating Physical Modeling, Neural Computing, and Statistical Analysis for On-Line Process Monitoring". United States.
@article{osti_20826820,
title = {Integrating Physical Modeling, Neural Computing, and Statistical Analysis for On-Line Process Monitoring},
author = {Garcia, Humberto E and Vilim, Richard B},
abstractNote = {Two basic approaches can be mentioned to model physical systems. One approach derives a model structure from the known physical laws. However, obtaining a model with the required fidelity may be difficult if the system is not well understood. A second approach is to employ a black-box structure to learn the implicit input-output relationships from measurements in which no particular attention is paid to modeling the underlying processes. A method that draws on the respective strengths of each of these two approaches is described. The technique integrates known first-principles knowledge derived from physical modeling with measured input-output mappings derived from neural processing to produce a computer model of a dynamical process. The technique is used to detect operational changes of mechanical equipment by statistically comparing, using a likelihood test, the predicted model output for the given measured input with the actual process output. Experimental results with a peristaltic pump are presented.},
doi = {},
url = {https://www.osti.gov/biblio/20826820}, journal = {Nuclear Technology},
issn = {0029-5450},
number = 1,
volume = 141,
place = {United States},
year = {2003},
month = {1}
}