Analysis of the DWPF glass pouring system using neural networks

PDF Version Also Available for Download.

Description

Neural networks were used to determine the sensitivity of 39 selected Melter/Melter Off Gas and Melter Feed System process parameters as related to the Defense Waste Processing Facility (DWPF) Melter Pour Spout Pressure during the overall analysis and resolution of the DWPF glass production and pouring issues. Two different commercial neural network software packages were used for this analysis. Models were developed and used to determine the critical parameters which accurately describe the DWPF Pour Spout Pressure. The model created using a low-end software package has a root mean square error of {+-} 0.35 inwc (< 2% of the instrument`s ... continued below

Physical Description

23 p.

Creation Information

Calloway, T.B. Jr.; Jantzen, C.M.; Medich, L. & Spennato, N. August 5, 1997.

Context

This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided by UNT Libraries Government Documents Department to Digital Library, a digital repository hosted by the UNT Libraries. More information about this article can be viewed below.

Who

People and organizations associated with either the creation of this article or its content.

Authors

Sponsor

Provided By

UNT Libraries Government Documents Department

Serving as both a federal and a state depository library, the UNT Libraries Government Documents Department maintains millions of items in a variety of formats. The department is a member of the FDLP Content Partnerships Program and an Affiliated Archive of the National Archives.

Contact Us

What

Descriptive information to help identify this article. Follow the links below to find similar items on the Digital Library.

Description

Neural networks were used to determine the sensitivity of 39 selected Melter/Melter Off Gas and Melter Feed System process parameters as related to the Defense Waste Processing Facility (DWPF) Melter Pour Spout Pressure during the overall analysis and resolution of the DWPF glass production and pouring issues. Two different commercial neural network software packages were used for this analysis. Models were developed and used to determine the critical parameters which accurately describe the DWPF Pour Spout Pressure. The model created using a low-end software package has a root mean square error of {+-} 0.35 inwc (< 2% of the instrument`s measured range, R{sup 2} = 0.77) with respect to the plant data used to validate and test the model. The model created using a high-end software package has a R{sub 2} = 0.97 with respect to the plant data used to validate and test the model. The models developed for this application identified the key process parameters which contribute to the control of the DWPF Melter Pour Spout pressure during glass pouring operations. The relative contribution and ranking of the selected parameters was determined using the modeling software. Neural network computing software was determined to be a cost-effective software tool for process engineers performing troubleshooting and system performance monitoring activities. In remote high-level waste processing environments, neural network software is especially useful as a replacement for sensors which have failed and are costly to replace. The software can be used to accurately model critical remotely installed plant instrumentation. When the instrumentation fails, the software can be used to provide a soft sensor to replace the actual sensor, thereby decreasing the overall operating cost. Additionally, neural network software tools require very little training and are especially useful in mining or selecting critical variables from the vast amounts of data collected from process computers.

Physical Description

23 p.

Notes

INIS; OSTI as DE98051768

Source

  • Waste management `98, Tucson, AZ (United States), 1-5 Mar 1998

Language

Item Type

Identifier

Unique identifying numbers for this article in the Digital Library or other systems.

  • Other: DE98051768
  • Report No.: WSRC-MS--97-00616
  • Report No.: CONF-980307--
  • Grant Number: AC09-96SR18500
  • Office of Scientific & Technical Information Report Number: 629338
  • Archival Resource Key: ark:/67531/metadc692110

Collections

This article is part of the following collection of related materials.

Office of Scientific & Technical Information Technical Reports

What responsibilities do I have when using this article?

When

Dates and time periods associated with this article.

Creation Date

  • August 5, 1997

Added to The UNT Digital Library

  • Aug. 14, 2015, 8:43 a.m.

Description Last Updated

  • Nov. 30, 2015, 3:39 p.m.

Usage Statistics

When was this article last used?

Yesterday: 0
Past 30 days: 0
Total Uses: 6

Interact With This Article

Here are some suggestions for what to do next.

Start Reading

PDF Version Also Available for Download.

Citations, Rights, Re-Use

Calloway, T.B. Jr.; Jantzen, C.M.; Medich, L. & Spennato, N. Analysis of the DWPF glass pouring system using neural networks, article, August 5, 1997; United States. (digital.library.unt.edu/ark:/67531/metadc692110/: accessed September 25, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.