gas centrifuge enrichment plant safeguards system modeling/67531/metadc876944/...gas centrifuge...
TRANSCRIPT
UCRL-CONF-221981
Gas Centrifuge Enrichment PlantSafeguards System Modeling
Hatem A. Elayat, W. J. O'Connell, Brian D. Boyer
June 12, 2006
47th Annual Meeting for the Institute of Nuclear MaterialsManagementNashville, TN, United StatesJuly 16, 2006 through July 20, 2006
Disclaimer
This document was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the University of California nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the University of California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or the University of California, and shall not be used for advertising or product endorsement purposes.
Gas Centrifuge Enrichment Plant Safeguards System Modeling H. A. Elayat, W. J. O’Connell
Lawrence Livermore National Laboratory 7000 East Ave., Livermore, CA 94550-9234, USA (925) 422-8983
B. D. Boyer Brookhaven National Laboratory
ABSTRACT The U.S. Department of Energy (DOE) is interested in developing tools and methods for potential U.S. use in designing and evaluating safeguards systems used in enrichment facilities. This research focuses on analyzing the effectiveness of the safeguards in protecting against the range of safeguards concerns for enrichment plants, including diversion of attractive material and unauthorized modes of use. We developed an Extend simulation model for a generic medium-sized centrifuge enrichment plant. We modeled the material flow in normal operation, plant operational upset modes, and selected diversion scenarios, for selected safeguards systems. Simulation modeling is used to analyze both authorized and unauthorized use of a plant and the flow of safeguards information. Simulation tracks the movement of materials and isotopes, identifies the signatures of unauthorized use, tracks the flow and compilation of safeguards data, and evaluates the effectiveness of the safeguards system in detecting misuse signatures. The simulation model developed could be of use to the International Atomic Energy Agency IAEA, enabling the IAEA to observe and draw conclusions that uranium enrichment facilities are being used only within authorized limits for peaceful uses of nuclear energy. It will evaluate improved approaches to nonproliferation concerns, facilitating deployment of enhanced and cost-effective safeguards systems for an important part of the nuclear power fuel cycle.
INTRODUCTION The U.S. Department of Energy (DOE) is interested in developing tools and methods for potential U.S. use in designing and evaluating safeguards systems used in enrichment facilities [3]. The International Atomic Energy Agency is also continuing to review needs, capabilities, and efficiency in safeguarding enrichment plants. The IAEA hosted a technical meeting in Vienna on April 18-22, 2005 with the aim of further strengthening its inspection and verification approaches applied to uranium enrichment activities. The present applied research focuses on analyzing the effectiveness of the safeguards in protecting against the range of safeguards concerns for gas centrifuge enrichment plants (GCEPs), including diversion of attractive material and unauthorized modes of use. It is part of a multi-laboratory DOE project, following on from an earlier examination of possible safeguards for natural uranium conversion plants [2, 4, 6]. Based on the earlier work, we have put together a tool suite for safeguards analysis, the Lawrence Livermore National Laboratory LLNL Integrated Safeguards System Analysis Tool (LISSAT) as outlined in Figure 1. It is a framework for performing systems analysis for evaluating the effectiveness of a safeguard system for a nuclear fuel cycle facility. LISSAT can be used to evaluate safeguards effectiveness for several nuclear fuel cycle facilities, as shown in Figure 2, like reactors [7], fuel fabrication plants, natural uranium conversion plants [2, 4, 6], nuclear fuel reprocessing facilities, enrichment facilities, spent fuel storage facilities, and Material tracking. In this paper we describe the application of LISSAT to evaluate the effectiveness of the safeguards for a generic enrichment facility. LISSAT is a suite of systems analysis tools including directed graphs and fault trees to structure the vulnerabilities, potential misuse actions, plant and safeguards configurations, and responses to observations. A second set of tools includes simulation modeling to analyze both authorized and unauthorized use of a plant and the flow of
safeguards information. We will analyze and simulate a generic medium-sized centrifuge enrichment plant’s material flow in normal operation, plant operational upset modes, and selected diversion scenarios; for selected safeguards systems. We will evaluate the effectiveness of selected new instruments in the information flow of safeguards information for timely detection of undeclared operations or material.
The digraph-fault tree analysis is at the heart of the process. It structures possible diversion activities in a diversion scenario together with the safeguards measures and activities relevant to the diversion scenario. Then it incorporates possible failure modes of the safeguards measures and develops a fault tree for the safeguard system in this situation. Among the inputs to the fault tree are the analysis of the inspector’s verification of the facility material declarations. Specifically these inputs are the probabilities of detection of various diversion activities meant to influence the facility’s declared nuclear material balance. Outputs of the digraph-fault tree analysis are the probability of success, quantity, and value of the material removed in the
diversion scenarios. The most attractive diversion scenarios are selected for time-domain simulation. The simulations track the uranium flow through the facility. The simulations include normal operation, intermediate storage, normal variations of input flow, and diversion scenarios. Simulation outputs are the time series of material outputs, which illustrate the data signatures of normal operation and diversion schemes. The simulation model can be used as an inspection tool. The model can be stored on the laptops of inspectors while inspection of the plant is occurring. The simulation model can be run to conduct “what if“ scenarios and can be helpful in identifying data signatures that are indicative of diversion. Several important results are generated when LISSAT is used. First, the directed graph/ fault tree analysis provides a structured systematic approach to incorporate all root causes for each diversion scenario including operator misdeclarations. Second, it helps quantify the change in the probability of diversion due to the introduction or use of material accounting, surveillance cameras, detectors, new safeguard measures/tools, new technology, and changes in plant design.
Digraph - Fault Tree Methodology
Simulation Model
Analysis of Verification of Facil ity Material Declarations
Diversion Scenarios (Probability, Sensitivity & Importance Analysis)
Selection of Most Attractive Diversion Scenarios
Plant Signatures – 1. Normal operation 2. Most Attractive Diversion Scenarios
F acility & Safeguards System s Characterization
Figure 1. LLNL Integrated Safeguards System Analysis Tool LISSAT for evaluating the the effectiveness of a safeguard system for a nuclear fuel cycle facility
Third, the simulation modeling helps identify plant signatures (normal versus abnormal) that might assist IAEA inspectors as indicators of diversion. Fourth, it helps identify the ideal location
Fig. 2. Current & Potential use of LISSAT in Nuclear Fuel Cycle Safeguards of detectors, surveillance cameras, and inspection points. In this paper we present the simulation model of the plant and example results. The statistical and digraph-fault tree analysis are currently ongoing.
Current use of LISSAT *
* *
*
Potential use of LISSAT
* * *
*
Material Tracking Potential use of LISSAT
* * *
FACILITY & SAFEGUARDS SYSTEM CHARACTERIZATION The facility analyzed in this paper is a medium sized generic enrichment facility based on Ref. [5]. It has a separative capacity of 500,000 SWU/yr. It has 50 cascades, each with 250 centrifuges, and each enriching to the same value of 3.5% enrichment. The feed to the plant is 900,000 kgU/yr in the form of UF6. The product is 115,000 kgU/yr in the form of UF6 enriched to 3.5%. The block diagram for this generic facility is shown in Figure 3. It consists of nine modules. There are two material balance areas (MBAs), the storage area and the operations area. The more detailed block diagram for the Cascade Hall is shown in Figure 4.
Fig. 3. Block diagram for generic enrichment facility
1. Feed storage 8. Tails storage 7. Product storage
2. Feed area 6 autoclaves
4. Product withdrawal
5. Tails withdrawal
6. Weighing and sampling area
9. Blending and rebatching area
3. Cascade halls 50 cascades 250 centrifuge per cascade
U-containing waste
U-containing waste
Receipts
Shipments
Fig. 4. Block diagram for Cascade Hall, autoclaves & product & tail withdrawal DIVERSION SCENARIO There are several safeguards concerns regarding GCEPs, including diversion of low-enriched uranium (LEU), excess production of LEU, and reconfiguration of part of the plant to produce HEU. As an example diversion scenario, we examine the steady skimming of 2% of the product (the product of one cascade) over a year, to divert 2000 kg of LEU, containing 70 kg of 235U. This is near the IAEA significant quantity of LEU, which is LEU containing 75 kg of 235U. SIMULATION MODEL An Extend simulation model was developed for this generic enrichment facility to model the flow of material throughout the facility. The model represents a continuous process. The simulation model consists of nine modules, each representing part of the enrichment process, from receipt of material at the feed storage to the enriched uranium at the product storage. The Product Withdrawal module is shown in Figure 5. It consists of iconic-blocks representing what is being modeled, the Desublimers, valves, and charts to monitor the flow of material at every stage. Input parameters to the simulation model were sized to match plant throughput. The simulation was run for one year with a time step of one hour. Figure 6 shows the plant signature when 2000Kg of LEU was diverted by skimming from one cascade throughout the year, it also shows the total amount of LEU produced throughout the year as well as the number of cylinders used.
gas
Autoclave
solid
Pressure reduction
Cascade
Autoclave
solid
Tails
Product
Feed
solid
solid
16 product Desublimers, 16 tails Desublimers; compress (pump) or cool to solidify
gas
gas solid gas
gas Product
Tails ….
Gas, 0.05 atmosphere or less
Transfer as needed
(6 autoclaves total)
Holdup drum (w. gas)
Gaseous impurities at startup; sample tube
Note:
represents a valve.
Feed storage
pcontent3
pcontent1
pcontent1
pcontent2pcontent2
pcontent3
pcontent3
pcontent4
pcontent4
pcontent1pcontent2pcontent3pcontent4
pcontent1pcontent2pcontent3pcontent4
Eqn
Eqn
getC
RS wan
t
Product UF6 from Cascade Pipe
getC
RS wan
t
getC
RS wan
t
getC
RS wan
t
T
Eqn
EqnN
B
A
Ya>=
product cylinder1
product cylinder1Eqn
N
B
A
Ya>=
N
B
A
Ya>=
N
B
A
Ya>=
Eqn
residual 1
residual 1
Eqn residual 2
Eqn
residual 3
Eqn residual 4
residual 2residual 3residual 4
product cylinder2
product cylinder2
Eqn
product cylinder3
product cylinder3
product cylinder4
product cylinder4
4. PRODUCT WITHDRAWAL
Fig. 5. Product Withdrawal Module in Extend Simulation Model
Fig. 6. Simulated Total Production of LEU (100,000 kg) & Diversion of 2000 kg of LEU by Skimming from One Cascade Throughout the Year
0 543.75 1087.5 1631.25 2175 2718.75 3262.5 3806.25 4350 4893.75 5437.5 5981.25 6525 7068.75 7612.5 8156.25 8700
0
3125
6250
9375
12500
15625
18750
21875
25000
28125
31250
34375
37500
40625
43750
46875
50000
53125
56250
59375
62500
65625
68750
71875
75000
78125
81250
84375
87500
90625
93750
96875
100000
Time
ValuePlotter I/O
LEU PRODUCT PRO… LEU CYLINDERS P… LEU PRODUCT DIV… Black
LEU Produced
Diversion of LEU Product by Skimming from one Cascade
100,000 Kg
2000 Kg
TIME (Hrs)
Fig. 7. Simulated Total Declared Production of LEU (100,000 kg) Under Normal Plant Operation Versus under Diversion of 2000 kg of LEU by Means of Skimming
REFERENCES
[1] B. Boyer, private communication, Brookhaven National Laboratory, May 12, 2005.
[2] H. A. Elayat, H. E. Lambert, and W. J. O’Connell, Systems Analysis of Safeguards Effectiveness in a Uranium Conversion Facility,” Paper 346, Proceedings of the 45th Annual Conference of the Institute of Nuclear Materials Management INMM, July 18-22, 2004, Orlando, FL. [3] B. McGinnis, B. Rollen, M. Whitaker, D. Lockwood, J. Murphy, and B. Moran, “Gas Centrifuge Uranium Enrichment Facilities in The United States – IAEA Safeguards Implementation,” Paper 259, Proceedings of the 46th Annual Conference of the Institute of Nuclear Materials Management INMM, July 10-14, 2005, Phoenix, AZ. [4] B. Boyer, et al, U.S. DOE Efforts to Enhance Conversion and Enrichment Plant Safeguards, presented at the 7th International Conference on Facility Operations-Safeguards Interface, February 29-March 4, 2004, Charleston, SC. [5] D. M. Gordon, B. D. Boyer, J. M. Whitaker ,and J. M. Younkin, Evaluation of IAEA Safeguards at Medium-Sized Gas Centrifuge Uranium Enrichment Plants, BNL/NCT-03-002, Brookhaven National Laboratory, and ORNL/ESP02-102, Oak Ridge National Laboratory, January 20, 2004 (Report is Official Use Only). [6] H. A. Elayat, H. E. Lambert, and W. J. O’Connell, Systems Analysis of Safeguards Effectiveness in a Uranium Conversion, Lawrence Livermore National Laboratory, UCRL-TR-212441, August 2005 (Report is Official Use Only).
0 725 1450 2175 2900 3625 4350 5075 5800 6525 7250 7975 87000
4141.667
8283.333
12425
16566.67
20708.33
24850
28991.67
33133.33
37275
41416.67
45558.33
49700
53841.67
57983.33
62125
66266.67
70408.33
74550
78691.67
82833.33
86975
91116.67
95258.33
99400
Time
ValuePlotter, Mult iSim
LEU PRODUCT PRO… LEU PRODUCT PRO… LEU PRODUCT PRO… LEU PRODUCT PRO…
LEU Produced without Diversion
LEU Produced with Diversion
[7] A. Bernstein, et. al., Use of Antineutrino Detectors for Nuclear Reactor Safeguards Effectiveness Assessment, Paper 284, Proceedings of the 47th Annual Conference of the Institute of Nuclear Materials Management INMM, July 16-20, 2006, Nashville, TN.