Further studies in A.I. Augmented process control using the BOXES methodology

D. W. Russell

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The application of the BOXES paradigm to continuous, non failing systems seems anomalous. The paper discusses such an application with special reference to a simulation of a trash to steam plant. The paper indicates cost savings attributable to intelligent control and opens for discussion real time BOXES applications for poorly or non-definable systems. The BOXES method must be modified to accommodate endless operation and focuses on the notion of augmenting a safe, non optimal system with AI generated control values. The paper compares the resulting plant performance to that of a simple rule based algorithm and a random idiot exercising the same control authority as BOXES.

Original languageEnglish (US)
Pages (from-to)75-79
Number of pages5
JournalAnnual Review in Automatic Programming
Volume16
Issue numberPART 1
DOIs
StatePublished - 1991

Fingerprint

Process control
Optimal systems
Intelligent control
Steam
Costs

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

@article{80c95d38f41449deada3c109c86b5ba5,
title = "Further studies in A.I. Augmented process control using the BOXES methodology",
abstract = "The application of the BOXES paradigm to continuous, non failing systems seems anomalous. The paper discusses such an application with special reference to a simulation of a trash to steam plant. The paper indicates cost savings attributable to intelligent control and opens for discussion real time BOXES applications for poorly or non-definable systems. The BOXES method must be modified to accommodate endless operation and focuses on the notion of augmenting a safe, non optimal system with AI generated control values. The paper compares the resulting plant performance to that of a simple rule based algorithm and a random idiot exercising the same control authority as BOXES.",
author = "Russell, {D. W.}",
year = "1991",
doi = "10.1016/0066-4138(91)90012-Z",
language = "English (US)",
volume = "16",
pages = "75--79",
journal = "Annual Reviews in Control",
issn = "1367-5788",
publisher = "Elsevier Limited",
number = "PART 1",

}

Further studies in A.I. Augmented process control using the BOXES methodology. / Russell, D. W.

In: Annual Review in Automatic Programming, Vol. 16, No. PART 1, 1991, p. 75-79.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Further studies in A.I. Augmented process control using the BOXES methodology

AU - Russell, D. W.

PY - 1991

Y1 - 1991

N2 - The application of the BOXES paradigm to continuous, non failing systems seems anomalous. The paper discusses such an application with special reference to a simulation of a trash to steam plant. The paper indicates cost savings attributable to intelligent control and opens for discussion real time BOXES applications for poorly or non-definable systems. The BOXES method must be modified to accommodate endless operation and focuses on the notion of augmenting a safe, non optimal system with AI generated control values. The paper compares the resulting plant performance to that of a simple rule based algorithm and a random idiot exercising the same control authority as BOXES.

AB - The application of the BOXES paradigm to continuous, non failing systems seems anomalous. The paper discusses such an application with special reference to a simulation of a trash to steam plant. The paper indicates cost savings attributable to intelligent control and opens for discussion real time BOXES applications for poorly or non-definable systems. The BOXES method must be modified to accommodate endless operation and focuses on the notion of augmenting a safe, non optimal system with AI generated control values. The paper compares the resulting plant performance to that of a simple rule based algorithm and a random idiot exercising the same control authority as BOXES.

UR - http://www.scopus.com/inward/record.url?scp=44949283615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44949283615&partnerID=8YFLogxK

U2 - 10.1016/0066-4138(91)90012-Z

DO - 10.1016/0066-4138(91)90012-Z

M3 - Article

AN - SCOPUS:44949283615

VL - 16

SP - 75

EP - 79

JO - Annual Reviews in Control

JF - Annual Reviews in Control

SN - 1367-5788

IS - PART 1

ER -