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

Title: The use of artificial intelligence for safeguarding fuel reprocessing plants

Abstract

Recorded process data from the ''Minirun'' campaigns conducted at the Barnwell Nuclear Fuel Plant (BNFP) in Barnwell, South Carolina during 1980 to 1981 have been utilized to study the suitability of computer-based Artificial Intelligence (AI) methods for process monitoring for safeguards purposes. The techniques of knowledge engineering were used to formulate the decision-making software which operates on the process data customarily used for process operations. The OPS5 AI language was used to construct an Expert System for this purpose. Such systems are able to form reasoned conclusions from incomplete, inaccurate or otherwise ''fuzzy'' data, and to explain the reasoning that led to them. The programs were tested using minirun data taken during simulated diversions ranging in size from 1 to 20 L of solution that had been monitored previously using conventional procedural techniques. 13 refs., 3 figs.

Authors:
;
Publication Date:
Research Org.:
Oak Ridge National Lab., TN (USA); Carnegie-Mellon Univ., Pittsburgh, PA (USA)
OSTI Identifier:
5458760
Report Number(s):
CONF-871110-6-FP
ON: DE88005053
DOE Contract Number:
AC05-84OR21400
Resource Type:
Conference
Resource Relation:
Conference: 3. international conference on facility operations safeguards interface, San Diego, CA, USA, 29 Nov 1987; Other Information: Paper copy only, copy does not permit microfiche production
Country of Publication:
United States
Language:
English
Subject:
98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL PROTECTION; 99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; FUEL REPROCESSING PLANTS; NUCLEAR MATERIALS MANAGEMENT; SAFEGUARDS; NUCLEAR MATERIALS DIVERSION; DETECTION; ARTIFICIAL INTELLIGENCE; EXPERT SYSTEMS; MONITORS; URANIUM; ACTINIDES; ELEMENTS; MANAGEMENT; MEASURING INSTRUMENTS; METALS; NUCLEAR FACILITIES; 055001* - Nuclear Fuels- Safeguards, Inspection, & Accountability- Technical Aspects; 990210 - Supercomputers- (1987-1989)

Citation Formats

Wachter, J.W., and Forgy, C.L. The use of artificial intelligence for safeguarding fuel reprocessing plants. United States: N. p., 1987. Web.
Wachter, J.W., & Forgy, C.L. The use of artificial intelligence for safeguarding fuel reprocessing plants. United States.
Wachter, J.W., and Forgy, C.L. Thu . "The use of artificial intelligence for safeguarding fuel reprocessing plants". United States. doi:. https://www.osti.gov/servlets/purl/5458760.
@article{osti_5458760,
title = {The use of artificial intelligence for safeguarding fuel reprocessing plants},
author = {Wachter, J.W. and Forgy, C.L.},
abstractNote = {Recorded process data from the ''Minirun'' campaigns conducted at the Barnwell Nuclear Fuel Plant (BNFP) in Barnwell, South Carolina during 1980 to 1981 have been utilized to study the suitability of computer-based Artificial Intelligence (AI) methods for process monitoring for safeguards purposes. The techniques of knowledge engineering were used to formulate the decision-making software which operates on the process data customarily used for process operations. The OPS5 AI language was used to construct an Expert System for this purpose. Such systems are able to form reasoned conclusions from incomplete, inaccurate or otherwise ''fuzzy'' data, and to explain the reasoning that led to them. The programs were tested using minirun data taken during simulated diversions ranging in size from 1 to 20 L of solution that had been monitored previously using conventional procedural techniques. 13 refs., 3 figs.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Jan 01 00:00:00 EST 1987},
month = {Thu Jan 01 00:00:00 EST 1987}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Recorded process data from minirun campaigns conducted at the Barnwell Nuclear Fuels Plant have been utilized to study the suitability of computer-based artificial intelligence (AI) methods for process monitoring for safeguards purposes. The techniques of knowledge engineering were used to formulate the decision-making software. The computer software accepted as input process data customarily used for process operations that had previously been recorded on magnetic tape during the 1980 miniruns. The OPS5 AI language was used to construct an expert system for simulated monitoring of the process. Such expert systems facilitate the employment of the heuristic reasoning used by human observersmore » to form reasoned conclusions from incomplete, inaccurate, or otherwise fuzzy data.« less
  • The International Atomic Energy Agency (IAEA) has established international safeguards standards for fissionable material at spent nuclear fuel reprocessing plants to ensure that significant quantities of weapons-grade nuclear material are not diverted from these facilities. Currently, methods to verify material control and accountancy (MC&A) at these facilities require time-consuming and resource-intensive destructive assay (DA). Leveraging new on-line non-destructive assay (NDA) techniques in conjunction with the traditional and highly precise DA methods may provide a more timely, cost-effective and resource-efficient means for MC&A verification at such facilities. Pacific Northwest National Laboratory (PNNL) is developing on-line NDA process monitoring technologies, including amore » spectroscopy-based monitoring system, to potentially reduce the time and resource burden associated with current techniques. The spectroscopic monitor continuously measures chemical compositions of the process streams including actinide metal ions (U, Pu, Np), selected fission products, and major non-radioactive flowsheet chemicals using UV-vis-near infrared and Raman spectroscopy. This paper provides an overview of the methods and reports our on-going efforts to develop and demonstrate the technologies. Our ability to identify material intentionally diverted from a liquid-liquid extraction contactor system was successfully tested using on-line process monitoring as a means to detect the amount of material diverted. A chemical diversion and detection of that diversion, from a solvent extraction scheme was demonstrated using a centrifugal contactor system operating a tributyl phosphate based extraction. A portion of the feed from a counter-current extraction system was diverted while a continuous extraction experiment was underway; the spectroscopic on-line process monitoring system was simultaneously measuring the feed, raffinate and organic products streams. The amount observed to be diverted by on-line spectroscopic process monitoring was in excellent agreement with values based from the known mass of sample directly diverted from system feed solution.« less
  • In the operation of a nuclear power plant, the sheer magnitude of the number of process parameters and systems interactions poses difficulties for the operators, particularly during abnormal or emergency situations. Recovery from an upset situation depends upon the facility with which the available raw data can be converted into and assimilated as meaningful knowledge. Plant personnel are sometimes affected by stress and emotion, which may have varying degrees of influence on their performance. Expert systems can take some of the uncertainty and guesswork out of their decisions by providing expert advice and rapid access to a large information base.more » Application of artificial intelligence technologies, particularly expert systems, to control room activities in a nuclear power plant has the potential to reduce operator error and improve power plant safety and reliability. 12 refs.« less
  • Safeguarding the plutonium passing through a large-scale reprocessing plant (such as one with 800 t of uranium per year) involves nondestructive assay measurements for plutonium at key points. The gamma-ray and neutron signals from the plutonium are generally hidden by the much larger backgrounds from fission products and actinides, so indirect measurements are routinely used. The intense neutron emission rate from spent fuel is from curium. In a spent fuel assembly at the head-end of a plant, the curium neutrons are used to deduce the amount of plutonium present. Coincidence and multiplicity counting are alternative ways to measure neutrons frommore » spent fuel; they have advantages over total neutron counting in certain conditions and offer new opportunities for examining assemblies. New uses for measurements of curium`s neutrons are proposed to safeguard waste streams. From a year`s work at a large-scale plant, 4 to 7 kg of plutonium can remain in leached hulls and 4 to 22 kg of plutonium can remain in the vitrified high-level liquid waste. While the plutonium in these wastes has the safeguards advantage of being dilute, it is important to verify (a) that the many kilograms involved are in fact present and (b) that the declared masses are not higher than the actual amounts so that more concentrated plutonium cannot pass through the plant by masquerading as waste. Curium measurements on spent fuel assemblies, the accountability tank, and leached hulls would form a safeguards system around all the inputs and outputs of a plant`s head-end where the plutonium is always intimately mixed with the curium. A neutron measurement of the vitrified waste would help identify the presence of a diversion path upstream because essentially all of the curium measured in the spent fuel assemblies should also be found in the vitrified waste (on a batch basis). 7 refs., 4 figs.« less