Integral nuclear data validation using experimental spent nuclear fuel compositions

Đăng ngày 7/3/2019 10:04:29 PM | Thể loại: | Lần tải: 0 | Lần xem: 11 | Page: 8 | FileSize: 2.26 M | File type: PDF
Integral nuclear data validation using experimental spent nuclear fuel compositions. Measurements of the isotopic contents of spent nuclear fuel provide experimental data that are a prerequisite for validating computer codes and nuclear data for many spent fuel applications. Under the auspices of the Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA) and guidance of the Expert Group on Assay Data of Spent Nuclear Fuel of the NEA Working Party on Nuclear Criticality Safety, a new database of expanded spent fuel isotopic compositions has been compiled.
Contents lists available at ScienceDirect
Nuclear Engineering and Technology
journal homepage: www.elsevier.com/locate/net
Original Article
Integral nuclear data validation using experimental spent nuclear fuel
compositions
Ian C. Gauld a, *, Mark L. Williams a, Franco Michel-Sendis b, Jesus S. Martinez a
a Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6170, USA
b OECD Nuclear Energy Agency, 46, Quai Alphonse Le Gallo, 92100 Boulogne-Billancourt, France
a r t i c l e
i n f o
a b s t r a c t
Article history:
Measurements of the isotopic contents of spent nuclear fuel provide experimental data that are a pre-
Received 1 June 2017
Accepted 5 July 2017
Available online 19 July 2017
requisite for validating computer codes and nuclear data for many spent fuel applications. Under the
auspices of the Organisation for Economic Co-operation and Development (OECD) Nuclear Energy
Agency (NEA) and guidance of the Expert Group on Assay Data of Spent Nuclear Fuel of the NEA Working
Keywords:
Adjoint Depletion
Nuclear Data
SFCOMPO Database
Spent Fuel
Uncertainty Analysis
Party on Nuclear Criticality Safety, a new database of expanded spent fuel isotopic compositions has been
compiled. The database, Spent Fuel Compositions (SFCOMPO) 2.0, includes measured data for more than
750 fuel samples acquired from 44 different reactors and representing eight different reactor technol-
ogies. Measurements for more than 90 isotopes are included. This new database provides data essential
for establishing the reliability of code systems for inventory predictions, but it also has broader potential
application to nuclear data evaluation. The database, together with adjoint based sensitivity and un-
certainty tools for transmutation systems developed to quantify the importance of nuclear data on
nuclide concentrations, are described.
© 2017 Korean Nuclear Society, Published by Elsevier Korea LLC. This is an open access article under the
1. Introduction
the criticality community and the availability of high-quality and
well-documented ICSBEP benchmarks.
Integral benchmark experiments provide global measures of
Under the auspices of the Organisation for Economic Co-
data performance for applications and are a valuable resource for
operation
and
Development
(OECD)
Nuclear
Energy
Agency
nuclear data testing and evaluation efforts. Integral testing of the
(NEA), physics benchmarks have been compiled and documented
US ENDF/B-VII.1 nuclear data library [1] and several international
in
the
International
Reactor
Physics
Benchmark
Experiment
nuclear data libraries rely heavily on testing using critical bench-
(IRPhEP) Handbook [3] that exercise nuclear data in broader range
marks as compiled in the International Criticality Safety Benchmark
of applications. As of 2014, the IRPhEP Handbook contained 136
Experiment Project (ICSBEP) Handbook [2]. Integral testing of
evaluations and measurements of critical buckling, spectral char-
ENDF/B-VII.1 included more than 2,000 critical benchmarks for
acteristics, reactivity coefcients, kinetics, reaction rates, and po-
compounds, metals, solutions, and other mixed systems in thermal,
wer distribution measurements. Both the ICSBEP and IRPhEP
intermediate, and fast neutron energy spectra. The use of criticality
benchmarks are applied to realistic applications, including uncer-
experiments as a centerpiece of data-testing efforts in the United
tainty analysis and cross section adjustment, uncertainty evalua-
States reects both the investment in nuclear data development by
tion for reactor core design methods, criticality safety problem
validation, and cross section data testing [4]. However, an intrinsic
limitation of both the ICSBEP and the IRPhEP benchmarks is that
Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-
their focus, for the most part, is on static systems of unirradiated
AC05-00OR22725 with the US Department of Energy (DOE). The US government
nuclear materials. Thus, many of these benchmarks exercise data
retains and the publisher, by accepting the article for publication, acknowledges
that the US government retains a nonexclusive, paid-up, irrevocable, worldwide
license to publish or reproduce the published form of this manuscript, or allow
others to do so, for US government purposes. DOE will provide public access to
these results of federally sponsored research in accordance with the DOE Public
for predominantly uranium and plutonium metals and compounds,
moderators/reectors materials, and absorber (criticality control)
materials. There are relatively few benchmarks for irradiated nu-
clear materials.
Interim dry fuel storage research and licensing activities are
* Corresponding author.
E-mail address: gauldi@ornl.gov (I.C. Gauld).
increasing as spent fuel pools at nuclear plants in many countries
1738-5733/© 2017 Korean Nuclear Society, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233
1227
are reaching capacity. In addition, geological repositories in Finland
design concept evaluations. The measured assay data have been
and Sweden are expected to be operational in the next decade.
compiled and reviewed by international radiochemistry experts
Recognizing the need for improved experimental benchmarks to
and modeling practitioners as coordinated through the activities of
support international activities in spent fuel management, the
the OECD/NEA EGADSNF [13].
Expert Group on Assay Data for Spent Nuclear Fuel (EGADSNF),
Since 2007, the database has been expanded beyond pres-
under the guidance of the Working Party on Nuclear Criticality
surized water reactor (PWR) and boiling water reactor (BWR) fuel
Safety of the OECD/NEA, has developed and expanded a database of
types that were included in earlier versions. The database now
spent fuel isotope compositions obtained by destructive radio-
includes eight different reactor designs: light water PWR, BWR,
chemical analysis of irradiated fuel samples from commercial nu-
waterewater
energetic
reactor
(VVER)-440,
and
VVER-1000;
clear reactors. Under this initiative, the Spent Fuel Compositions
heavy water CANDU (CANada Deuterium Uranium), graphite-
database, SFCOMPO 2.0, originally developed by the Japan Atomic
moderated Magnox, AGR (advanced gas-cooled reactor), and
Energy Research Institution [5], has been expanded from 246
RBMK (Reaktor Bolshoy Moshchnosti Kanalnyy) reactor designs.
samples to more than 750 samples from 44 reactors representing
Measurements have been acquired from 67 different fuel as-
eight different international reactor technologies.
semblies and 291 fuel rods. The updated database, SFCOMPO 2.0,
Because of the design complexities and uncertainties in the
contains more than 20,000 measurement entries, with associated
actual conditions of an operating reactor, applying these data to
uncertainties as reported by the measurement laboratories.
obtain realistic estimates of the accuracy of calculated results can
Each reactor design class includes a range of fuel assembly de-
be extremely difcult. In addition, there can be large uncertainties
signs. The diversity of designs, coolants, moderators, enrichments,
in the measurement data attributed to the complex radiochemical
and fuel burnup values provides a wide range of neutron spectral
measurement processes not present in criticality and reactor
conditionsduringirradiationthatcanbeusedtotesttheperformance
physics experiments. Under the OECD/NEA, these data have been
of nuclear data over different neutron energy regions. For example,
more completely documented and peer reviewed for accuracy by
the database includes spent fuel measurements for samples with
international experts. Although measurement accuracies and the
enrichments from 0.71 wt.% (natural U) to 4.9 wt.%, and burnup
quality and extent of documentation often fall short of the high
values from 0.85 GWd/tU to more than 70 GWd/tU. Coolant and
standards set by the ICSBEP and IRPhEP benchmarks, these mea-
moderator materials include light water, heavy water, gas, and
surements nevertheless represent valuable validation tests and
graphite.Thereactors,countryoforigin,andthenumberofmeasured
have been widely used to evaluate the performance characteristics
fuel samples available for each reactor are summarized in Table 1.
of computer codes, models, and nuclear data used for calculating
The database is designed with a top-to-bottom hierarchical
spent fuel compositions [6,7].
structure organized by reactors, fuel assemblies, fuel rods, and
Because of the signicant complexity of spent fuel measure-
samples, with the measurement data usually being associated at
ments and past inconsistent documentation of reactor design,
the sample level. However, some experiments involving dissolution
operating data, and the design and measurement uncertainties,
of entire assemblies from a reprocessing plant are included and are
these measurements have not been widely adopted for integral
therefore associated with the assembly level.
testing of nuclear data. France is an exception, where measure-
The database includes measured nuclide concentrations, activ-
ments of irradiated fuel compositions from the French Gravelines
ities, isotopic atom and mass ratios, measurement method, uncer-
and Bugey reactors, and the Swiss Beznau reactor, were used in the
tainty, laboratory, and measurement date. Measurements for 91
development and validation of the European JEF-2.2 libraries [8].
isotopes are included (Fig. 1). Measurement data are always
More recently, additional assay measurements, including data
included as the values and units reported by the laboratory or
made available in SFCOMPO, have been used in the validation of
experimental program. Where possible, the measurement data are
JEFF-3.1 [9,10].
automatically converted by the program to a common mass basis of
Unlike the ICSBEP and IRPhEP benchmarks involving primarily
mg/g fuel or mg/g initial U.
steady-state
systems,
spent
fuel
experiments
involve
time-
With the addition of newer experimental program data that
dependent isotope depletion and decay as dened by the trans-
include high-precision measurements and improved design and
mutation equations. Therefore, identifying the relationships be-
operating data documentation, the quality of the database has
tween
computational
bias
and
nuclear
data
is
particularly
improved signicantly compared to earlier versions. Moreover, as
challenging. Tools to help understand and quantify the complex
part of the data review process, multiple independent cross checks
relationships between nuclear data and measured compositions
of the information have been performed. Also, preliminary bench-
are therefore needed in order to better utilize databases such as
mark models have been developed for many of the experiments.
SFCOMPO 2.0 for nuclear data testing and evaluation.
An important caveat on the database is that these reviews have
This paper describes the current status of the new SFCOMPO 2.0
been performed to ensure that the database accurately reects the
database and benchmark development efforts, and it summarizes
experimental data as reported and are traceable to primary refer-
and demonstrates new nuclear data sensitivity and uncertainty
ence reports. The reviews do not constitute a formal evaluation of
analysis tools for depletion problems being developed and targeted
the data to assess quality and accuracy of the measurements,
for future release in the SCALE nuclear systems analysis code sys-
consistency of the data, completeness of the documentation,
tem [11].
development of benchmarks, or an analysis of uncertainties.
However, procedures for such evaluations have been developed
2. SFCOMPO database
[14], and preliminary detailed evaluation reports have been initi-
ated for several datasets.
2.1. Database content
2.2. Database application interface
SFCOMPO 2.0 is a database of well-documented, peer-reviewed
measurements of spent fuel assay data [12] that can be used by the
SFCOMPO 2.0 is a structured query language Java H2 relational
nuclear community for validating computational systems and nu-
database
accessible
through
a
graphical
application
interface
clear data in applications of reactor physics, nuclear fuel cycle, used
developed by the NEA. The database has a similar interface and
fuel storage and disposal, as well as advanced fuels and reactor
functionality as the IRPhEP Database and Analysis Tool [15] and
1228
I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233
Table 1
2.3. Search capabilities
Summary of SFCOMPO 2.0 database content.
Design
Reactor
Country
Samples
Many database elds can be queried using the Basic and
AGR
Hinkleyd3
Hinkleyd4
United Kingdom
United Kingdom
21
36
Advanced search panels. The Basic search query includes reactor
type, measured fuel type (i.e., U-metal, UO2 or MOX, UO2-Gd2O3),
Hunterston Bd1
United Kingdom
6
sample burnup range, and measured nuclide (see the nuclide
BWR
Cooper
Dodewaard
Forsmarkd3
United States
The Netherlands
Sweden
17
5
2
search panel in Fig. 1). Advanced query parameters include the
following:
Fukushima Daiichid3
Fukushima Dainid1
Japan
Japan
36
13
 Reactor name
Fukushima Dainid2
Japan
44
 Coolant type
Gariglianod1
Gundremmingend1
JPDR
Monticello
Italy
Germany
Japan
Italy
26
18
30
30
 Moderator type
 Assembly identiers
 Lattice type
Quad Citiesd1
United States
18
 Lattice dimension
CANDU
Magnox
Tsurugad1
Bruced1
NPD
Pickering Ad1
Bradwelld1
Japan
Canada
Canada
Canada
United Kingdom
10
3
27
1
1
 Fuel rod pitch (range)
 Fuel rod diameter (range)
 Enrichment (range)
 Axial position or height of the sample (range)
Hunterston Ad1
United Kingdom
3
 Measurement type
PWR
Beznaud1
Calvert Cliffsd1
Switzerland
United States
6
33
 Measurement uncertainty (range)
Genkaid1
Gosgend1
Japan
Switzerland
2
4
These search features allow the user to rapidly identify mea-
H. R. Robinsond2
United States
7
surement data for isotopes, the reactor, and fuel design parameter
Mihamad3
Neckarwestheimd2
Obrigheimd1
Ohid1
Ohid2
Japan
Germany
Germany
Japan
Japan
9
1
33
1
5
characteristics of interest. The search panel returns the nuclide
concentration values and uncertainties of experimental measure-
ments as queried by the user for inspection and comparison pur-
poses. SFCOMPO 2.0 allows the user to visualize measurement data
Takahamad3
Japan
16
of selected samples and provides the numeric measurements data,
Three Mile Islandd1
Trino Vercellesed1
Turkey Pointd3
Vandellosd2
United States
Italy
United States
Spain
24
49
18
9
along with other information queried. The search results list the
sample level information for all samples matching the search
criteria and the measurement data in separate panels. A graphical
Yankee
United States
78
plotter displays nuclide concentrations provided in normalized
RBMK
VVER-1000
Leningradd1
Balakovod2
Balakovod3
Kalinind1
Novovoronezhd5
Russia
Russia
Russia
Russia
Russia
41
3
2
5
10
units of initial (unirradiated) fuel mass as a function of sample
burnup. A more general plotting capability can be used to display
most of the measurement data against other elds used as the x
axis. Fig. 4 shows the results of a database query for 239Pu mea-
VVER-440
Kolad3
Russia
12
surements for all uranium-based fuels and a plot of the results and
Novovoronezhd3
Novovoronezhd4
Russia
Russia
7
28
uncertainties as a function of sample burnup.
AGR, advanced gas-cooled reactor; BWR, boiling water reactor; CANDU, CANada
Deuterium Uranium; PWR, pressurized water reactor; RBMK, Reaktor Bolshoy
3. Applications of the database
Moshchnosti Kanalnyy; SFCOMPO, Spent Fuel Compositions database; VVER,
waterewater energetic reactor.
Destructive radiochemical assay data are one of the primary
means used to validate depletion codes and nuclear data libraries
for spent fuel applications. Measured isotopic data in SFCOMPO 2.0
Database for the ICSBEP [16] programs, also developed by the NEA.
include many nuclides that dominate applications in nuclear crit-
Information in the database can be accessed by browsing the data in
icality safety (burnup credit), gamma radiation sources, passive
a hierarchical organization structure by reactor technology type
neutron sources, decay heat, and long-lived radiological isotopes
(e.g., BWR, Magnox), reactor name, assembly ID, fuel rod, and
important
to
geological
repository
disposal.
A
widely
used
measured fuel sample level. Fig. 2 illustrates the database structural
approach to burnup code and data validation is to apply the code to
organization
as
viewed
in
the
application
interface
browser,
calculate the same nuclide inventories for which measurements are
expanded to showthedata forone of theFukushima-Daini-1reactor
available. This technique has underpinned uncertainty analysis for
data sets. Fig. 3 provides a screen shot of the database browse panel
many spent fuel applications, including burnup credit, in nuclear
showing the organizational structure (left side) and the display of
criticality safety. In addition to measured nuclide concentration
design and operating information at the assembly level.
data, these applications also require the detailed design and reactor
A major advancement is the ability to store and display more
operating conditions needed to model the fuel assembly irradiation
complete and detailed fuel and assembly design information and
and predict the nuclide concentrations in the fuel.
time-dependent operating history data for the reactor, assembly,
Another application of the data includes the use of measured
fuel rod, and measured sample. SFCOMPO 2.0 also provides refer-
data alone without the need to model individual experiments.
ences for all the data it displays. These references are linked to
SFCOMPO 2.0 has been used recently to develop uncertainty bands
primary reports such as radiochemistry laboratory reports, open
for waste categorization in instances where detailed information
journal papers, and assay data evaluation reports, providing full
on the waste is not available [17]. For this application, there is no
traceability of the data. Primaryexperimental and reference reports
requirement for detailed operating history information.
are accessible as electronic portable document format (PDF) ver-
Comparing
calculated-to-measured
nuclide
concentrations
sions of the reports that can be accessed through a built-in viewer.
provides information necessary to validate codes and estimate the
HƯỚNG DẪN DOWNLOAD TÀI LIỆU

Bước 1:Tại trang tài liệu slideshare.vn bạn muốn tải, click vào nút Download màu xanh lá cây ở phía trên.
Bước 2: Tại liên kết tải về, bạn chọn liên kết để tải File về máy tính. Tại đây sẽ có lựa chọn tải File được lưu trên slideshare.vn
Bước 3: Một thông báo xuất hiện ở phía cuối trình duyệt, hỏi bạn muốn lưu . - Nếu click vào Save, file sẽ được lưu về máy (Quá trình tải file nhanh hay chậm phụ thuộc vào đường truyền internet, dung lượng file bạn muốn tải)
Có nhiều phần mềm hỗ trợ việc download file về máy tính với tốc độ tải file nhanh như: Internet Download Manager (IDM), Free Download Manager, ... Tùy vào sở thích của từng người mà người dùng chọn lựa phần mềm hỗ trợ download cho máy tính của mình  
11 lần xem

Integral nuclear data validation using experimental spent nuclear fuel compositions. Measurements of the isotopic contents of spent nuclear fuel provide experimental data that are a prerequisite for validating computer codes and nuclear data for many spent fuel applications. Under the auspices of the Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA) and guidance of the Expert Group on Assay Data of Spent Nuclear Fuel of the NEA Working Party on Nuclear Criticality Safety, a new database of expanded spent fuel isotopic compositions has been compiled..

Nội dung

Nuclear Engineering and Technology 49 (2017) 1226e1233 Contents lists available at ScienceDirect Nuclear Engineering and Technology journal homepage: www.elsevier.com/locate/net Original Article Integral nuclear data validation using experimental spent nuclear fuel compositions Ian C. Gauld a, *, Mark L. Williams a, Franco Michel-Sendis b, Jesus S. Martinez a a Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6170, USA b OECD Nuclear Energy Agency, 46, Quai Alphonse Le Gallo, 92100 Boulogne-Billancourt, France a r t i c l e i n f o a b s t r a c t Article history: Received 1 June 2017 Accepted 5 July 2017 Available online 19 July 2017 Keywords: Adjoint Depletion Nuclear Data SFCOMPO Database Spent Fuel Uncertainty Analysis 1. Introduction Measurements of the isotopic contents of spent nuclear fuel provide experimental data that are a pre-requisite for validating computer codes and nuclear data for many spent fuel applications. Under the auspices of the Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA) and guidance of the Expert Group on Assay Data of Spent Nuclear Fuel of the NEA Working Party on Nuclear Criticality Safety, a new database of expanded spent fuel isotopic compositions has been compiled. The database, Spent Fuel Compositions (SFCOMPO) 2.0, includes measured data for more than 750 fuel samples acquired from 44 different reactors and representing eight different reactor technol-ogies. Measurements for more than 90 isotopes are included. This new database provides data essential for establishing the reliability of code systems for inventory predictions, but it also has broader potential application to nuclear data evaluation. The database, together with adjoint based sensitivity and un-certainty tools for transmutation systems developed to quantify the importance of nuclear data on nuclide concentrations, are described. © 2017 Korean Nuclear Society, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). the criticality community and the availability of high-quality and well-documented ICSBEP benchmarks. Integral benchmark experiments provide global measures of data performance for applications and are a valuable resource for nuclear data testing and evaluation efforts. Integral testing of the US ENDF/B-VII.1 nuclear data library [1] and several international nuclear data libraries rely heavily on testing using critical bench-marks as compiled in the International Criticality Safety Benchmark Experiment Project (ICSBEP) Handbook [2]. Integral testing of ENDF/B-VII.1 included more than 2,000 critical benchmarks for compounds, metals, solutions, and other mixed systems in thermal, intermediate, and fast neutron energy spectra. The use of criticality experiments as a centerpiece of data-testing efforts in the United States reflects both the investment in nuclear data development by Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). * Corresponding author. E-mail address: gauldi@ornl.gov (I.C. Gauld). Under the auspices of the Organisation for Economic Co-operation and Development (OECD) Nuclear Energy Agency (NEA), physics benchmarks have been compiled and documented in the International Reactor Physics Benchmark Experiment (IRPhEP) Handbook [3] that exercise nuclear data in broader range of applications. As of 2014, the IRPhEP Handbook contained 136 evaluations and measurements of critical buckling, spectral char-acteristics, reactivity coefficients, kinetics, reaction rates, and po-wer distribution measurements. Both the ICSBEP and IRPhEP benchmarks are applied to realistic applications, including uncer-tainty analysis and cross section adjustment, uncertainty evalua-tion for reactor core design methods, criticality safety problem validation, and cross section data testing [4]. However, an intrinsic limitation of both the ICSBEP and the IRPhEP benchmarks is that their focus, for the most part, is on static systems of unirradiated nuclear materials. Thus, many of these benchmarks exercise data for predominantly uranium and plutonium metals and compounds, moderators/reflectors materials, and absorber (criticality control) materials. There are relatively few benchmarks for irradiated nu-clear materials. Interim dry fuel storage research and licensing activities are increasing as spent fuel pools at nuclear plants in many countries http://dx.doi.org/10.1016/j.net.2017.07.002 1738-5733/© 2017 Korean Nuclear Society, Published by Elsevier Korea LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233 1227 are reaching capacity. In addition, geological repositories in Finland and Sweden are expected to be operational in the next decade. Recognizing the need for improved experimental benchmarks to support international activities in spent fuel management, the Expert Group on Assay Data for Spent Nuclear Fuel (EGADSNF), under the guidance of the Working Party on Nuclear Criticality Safety of the OECD/NEA, has developed and expanded a database of spent fuel isotope compositions obtained by destructive radio-chemical analysis of irradiated fuel samples from commercial nu-clear reactors. Under this initiative, the Spent Fuel Compositions database, SFCOMPO 2.0, originally developed by the Japan Atomic Energy Research Institution [5], has been expanded from 246 samples to more than 750 samples from 44 reactors representing eight different international reactor technologies. Because of the design complexities and uncertainties in the actual conditions of an operating reactor, applying these data to obtain realistic estimates of the accuracy of calculated results can be extremely difficult. In addition, there can be large uncertainties in the measurement data attributed to the complex radiochemical measurement processes not present in criticality and reactor physics experiments. Under the OECD/NEA, these data have been more completely documented and peer reviewed for accuracy by international experts. Although measurement accuracies and the quality and extent of documentation often fall short of the high standards set by the ICSBEP and IRPhEP benchmarks, these mea-surements nevertheless represent valuable validation tests and have been widely used to evaluate the performance characteristics of computer codes, models, and nuclear data used for calculating spent fuel compositions [6,7]. Because of the significant complexity of spent fuel measure-ments and past inconsistent documentation of reactor design, operating data, and the design and measurement uncertainties, these measurements have not been widely adopted for integral testing of nuclear data. France is an exception, where measure-ments of irradiated fuel compositions from the French Gravelines and Bugey reactors, and the Swiss Beznau reactor, were used in the development and validation of the European JEF-2.2 libraries [8]. More recently, additional assay measurements, including data made available in SFCOMPO, have been used in the validation of JEFF-3.1 [9,10]. Unlike the ICSBEP and IRPhEP benchmarks involving primarily design concept evaluations. The measured assay data have been compiled and reviewed by international radiochemistry experts and modeling practitioners as coordinated through the activities of the OECD/NEA EGADSNF [13]. Since 2007, the database has been expanded beyond pres-surized water reactor (PWR) and boiling water reactor (BWR) fuel types that were included in earlier versions. The database now includes eight different reactor designs: light water PWR, BWR, waterewater energetic reactor (VVER)-440, and VVER-1000; heavy water CANDU (CANada Deuterium Uranium), graphite-moderated Magnox, AGR (advanced gas-cooled reactor), and RBMK (Reaktor Bolshoy Moshchnosti Kanalnyy) reactor designs. Measurements have been acquired from 67 different fuel as-semblies and 291 fuel rods. The updated database, SFCOMPO 2.0, contains more than 20,000 measurement entries, with associated uncertainties as reported by the measurement laboratories. Each reactor design class includes a range of fuel assembly de-signs. The diversity of designs, coolants, moderators, enrichments, and fuel burnup values provides a wide range of neutron spectral conditionsduringirradiationthatcanbeusedtotesttheperformance of nuclear data over different neutron energy regions. For example, the database includes spent fuel measurements for samples with enrichments from 0.71 wt.% (natural U) to 4.9 wt.%, and burnup values from 0.85 GWd/tU to more than 70 GWd/tU. Coolant and moderator materials include light water, heavy water, gas, and graphite.Thereactors,countryoforigin,andthenumberofmeasured fuel samples available for each reactor are summarized in Table 1. The database is designed with a top-to-bottom hierarchical structure organized by reactors, fuel assemblies, fuel rods, and samples, with the measurement data usually being associated at the sample level. However, some experiments involving dissolution of entire assemblies from a reprocessing plant are included and are therefore associated with the assembly level. The database includes measured nuclide concentrations, activ-ities, isotopic atom and mass ratios, measurement method, uncer-tainty, laboratory, and measurement date. Measurements for 91 isotopes are included (Fig. 1). Measurement data are always included as the values and units reported by the laboratory or experimental program. Where possible, the measurement data are automatically converted by the program to a common mass basis of mg/g fuel or mg/g initial U. steady-state systems, spent fuel experiments involve time- With the addition of newer experimental program data that dependent isotope depletion and decay as defined by the trans- mutation equations. Therefore, identifying the relationships be- include high-precision measurements and improved design and operating data documentation, the quality of the database has tween computational bias and nuclear data is particularly improved significantly compared to earlier versions. Moreover, as challenging. Tools to help understand and quantify the complex relationships between nuclear data and measured compositions are therefore needed in order to better utilize databases such as SFCOMPO 2.0 for nuclear data testing and evaluation. This paper describes the current status of the new SFCOMPO 2.0 database and benchmark development efforts, and it summarizes and demonstrates new nuclear data sensitivity and uncertainty analysis tools for depletion problems being developed and targeted for future release in the SCALE nuclear systems analysis code sys-tem [11]. 2. SFCOMPO database 2.1. Database content SFCOMPO 2.0 is a database of well-documented, peer-reviewed measurements of spent fuel assay data [12] that can be used by the nuclear community for validating computational systems and nu-clear data in applications of reactor physics, nuclear fuel cycle, used fuel storage and disposal, as well as advanced fuels and reactor part of the data review process, multiple independent cross checks of the information have been performed. Also, preliminary bench-mark models have been developed for many of the experiments. An important caveat on the database is that these reviews have been performed to ensure that the database accurately reflects the experimental data as reported and are traceable to primary refer-ence reports. The reviews do not constitute a formal evaluation of the data to assess quality and accuracy of the measurements, consistency of the data, completeness of the documentation, development of benchmarks, or an analysis of uncertainties. However, procedures for such evaluations have been developed [14], and preliminary detailed evaluation reports have been initi-ated for several datasets. 2.2. Database application interface SFCOMPO 2.0 is a structured query language Java H2 relational database accessible through a graphical application interface developed by the NEA. The database has a similar interface and functionality as the IRPhEP Database and Analysis Tool [15] and 1228 Table 1 Summary of SFCOMPO 2.0 database content. I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233 2.3. Search capabilities Design AGR BWR CANDU Magnox PWR RBMK VVER-1000 VVER-440 Reactor Hinkleyd3 Hinkleyd4 Hunterston Bd1 Cooper Dodewaard Forsmarkd3 Fukushima Daiichid3 Fukushima Dainid1 Fukushima Dainid2 Gariglianod1 Gundremmingend1 JPDR Monticello Quad Citiesd1 Tsurugad1 Bruced1 NPD Pickering Ad1 Bradwelld1 Hunterston Ad1 Beznaud1 Calvert Cliffsd1 Genkaid1 Gosgend1 H. R. Robinsond2 Mihamad3 Neckarwestheimd2 Obrigheimd1 Ohid1 Ohid2 Takahamad3 Three Mile Islandd1 Trino Vercellesed1 Turkey Pointd3 Vandellosd2 Yankee Leningradd1 Balakovod2 Balakovod3 Kalinind1 Novovoronezhd5 Kolad3 Novovoronezhd3 Novovoronezhd4 Country United Kingdom United Kingdom United Kingdom United States The Netherlands Sweden Japan Japan Japan Italy Germany Japan Italy United States Japan Canada Canada Canada United Kingdom United Kingdom Switzerland United States Japan Switzerland United States Japan Germany Germany Japan Japan Japan United States Italy United States Spain United States Russia Russia Russia Russia Russia Russia Russia Russia Samples 21 36 6 17 5 2 36 13 44 26 18 30 30 18 10 3 27 1 1 3 6 33 2 4 7 9 1 33 1 5 16 24 49 18 9 78 41 3 2 5 10 12 7 28 Many database fields can be queried using the Basic and Advanced search panels. The Basic search query includes reactor type, measured fuel type (i.e., U-metal, UO2 or MOX, UO2-Gd2O3), sample burnup range, and measured nuclide (see the nuclide search panel in Fig. 1). Advanced query parameters include the following: Reactor name Coolant type Moderator type Assembly identifiers Lattice type Lattice dimension Fuel rod pitch (range) Fuel rod diameter (range) Enrichment (range) Axial position or height of the sample (range) Measurement type Measurement uncertainty (range) These search features allow the user to rapidly identify mea-surement data for isotopes, the reactor, and fuel design parameter characteristics of interest. The search panel returns the nuclide concentration values and uncertainties of experimental measure-ments as queried by the user for inspection and comparison pur-poses. SFCOMPO 2.0 allows the user to visualize measurement data of selected samples and provides the numeric measurements data, along with other information queried. The search results list the sample level information for all samples matching the search criteria and the measurement data in separate panels. A graphical plotter displays nuclide concentrations provided in normalized units of initial (unirradiated) fuel mass as a function of sample burnup. A more general plotting capability can be used to display most of the measurement data against other fields used as the x axis. Fig. 4 shows the results of a database query for 239Pu mea- surements for all uranium-based fuels and a plot of the results and uncertainties as a function of sample burnup. AGR, advanced gas-cooled reactor; BWR, boiling water reactor; CANDU, CANada Deuterium Uranium; PWR, pressurized water reactor; RBMK, Reaktor Bolshoy Moshchnosti Kanalnyy; SFCOMPO, Spent Fuel Compositions database; VVER, waterewater energetic reactor. Database for the ICSBEP [16] programs, also developed by the NEA. Information in the database can be accessed by browsing the data in a hierarchical organization structure by reactor technology type (e.g., BWR, Magnox), reactor name, assembly ID, fuel rod, and measured fuel sample level. Fig. 2 illustrates the database structural organization as viewed in the application interface browser, expanded to showthedata forone of theFukushima-Daini-1reactor data sets. Fig. 3 provides a screen shot of the database browse panel showing the organizational structure (left side) and the display of design and operating information at the assembly level. A major advancement is the ability to store and display more complete and detailed fuel and assembly design information and time-dependent operating history data for the reactor, assembly, fuel rod, and measured sample. SFCOMPO 2.0 also provides refer-ences for all the data it displays. These references are linked to primary reports such as radiochemistry laboratory reports, open journal papers, and assay data evaluation reports, providing full traceability of the data. Primaryexperimental and reference reports 3. Applications of the database Destructive radiochemical assay data are one of the primary means used to validate depletion codes and nuclear data libraries for spent fuel applications. Measured isotopic data in SFCOMPO 2.0 include many nuclides that dominate applications in nuclear crit-icality safety (burnup credit), gamma radiation sources, passive neutron sources, decay heat, and long-lived radiological isotopes important to geological repository disposal. A widely used approach to burnup code and data validation is to apply the code to calculate the same nuclide inventories for which measurements are available. This technique has underpinned uncertainty analysis for many spent fuel applications, including burnup credit, in nuclear criticality safety. In addition to measured nuclide concentration data, these applications also require the detailed design and reactor operating conditions needed to model the fuel assembly irradiation and predict the nuclide concentrations in the fuel. Another application of the data includes the use of measured data alone without the need to model individual experiments. SFCOMPO 2.0 has been used recently to develop uncertainty bands for waste categorization in instances where detailed information on the waste is not available [17]. For this application, there is no requirement for detailed operating history information. are accessible as electronic portable document format (PDF) ver- Comparing calculated-to-measured nuclide concentrations sions of the reports that can be accessed through a built-in viewer. provides information necessary to validate codes and estimate the I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233 1229 Fig. 1. Measured isotopes available in SFCOMPO 2.0. SFCOMPO, Spent Fuel Compositions database. bias and uncertainty in isotope predictions and other integral quantities. However, these comparisons do not provide direct in-formation on the accuracy of the specific nuclear data used by the codes. The time-dependent relationship between nuclide concen-trations, Ni, and nucleardata is given by the transmutation equation vti ¼ liNi þ Xlj/iNj þXNj0sj0j/i〈f〉 Ni XhsiiðrÞ〈f〉 |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} Decay Reactions (1) where decay constants l, energy-averaged neutron cross sections s and energy-integrated neutron flux f define the transition rates. Provided that the system description is given such that accurate es-timates of fðEÞ can be obtained by transport calculations of the fuel assembly,andthenuclidecontentismeasuredwithlowuncertainty, spent fuel measurements can be used for nuclear data testing becausetheresultsofcomputationsimulationsoftheseexperiments rely directly on the accuracy of the data used by the codes. Application to nuclear data evaluation is complicated by the large amount of data used by the codes for burnup simulations and time dependence of data importance in transmutation analysis. The production of any nuclides usually depends on cross sections and half-lives for many different nuclides in the production chain, and fission products also depend on their fission yields. Software tools to aid in the evaluation of nuclear data uncertainty in trans-mutation systems are being developed at Oak Ridge National Lab-oratory (ORNL). The application of these tools to nuclear data evaluation and testing using spent fuel databases such as SFCOMPO is briefly summarized below. 4. Tools for nuclear data assessment Fig. 2. Screen capture of the SFCOMPO 2.0 organizational structure showing infor-mation for the Fukushima-Daini-1 samples of assembly 2F1ZN3 rod A9. SFCOMPO, Spent Fuel Compositions database. Computational analysis of irradiated fuel compositions can provide a wealth of information concerning bias and uncertainty in the nuclear data used in modeling and simulation codes, including 1230 I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233 Fig. 3. Browser panel showing CANDU reactor design data (37-element design) and fuel bundle map of measured fuel rod locations. CANDU, CANada Deuterium Uranium. neutron cross section data, fissionproduct yields, and nucleardecay data.Transmutationcalculationsusealargeamountofnucleardata, and it is essential in nucleardata testing to identify which data have thelargestsensitivityontheresponsesandwhichdataaretherefore equations in a coupled neutronenuclide field, such as in a reactor core, has also been developed [22]. In the uncoupled formalism, the absolute sensitivity of a calculated response R to nucleardata element acan be shown to be, most responsible for biases in the calculated results. The trans-mutation matrix used by ORIGEN contains more than 50,000 tran- sition element coefficients that represent energy-averaged neutron SR ¼ N*ðtÞvAðaÞNTðtÞ; (2) reactions, decay branch transitions, and fission product yields. The cross sections alone include data for 774 nuclides and 14,000 neutron reactions. Therefore, direct perturbation of nuclear data for thesecomplex systemstodetermine theimportanceof nucleardata on different observables is often not efficient or practical. An efficient approach to calculating nuclear data sensitivities and uncertainties from nuclear data is to use adjoint-based perturbation theory. Adjoint techniques for criticality safety ana-lyses have been developed and are now widely adopted for nuclear data sensitivity analysis, which is required for numerous applica-tions including experiment similarity analysis [18] and nuclear data adjustment for eigenvalue problems [19]. However, similar tech-niques are not widely available to analyze time-dependent trans-mutation or depletion systems. The development of the adjoint equations for depletion was first published by Gandini [20] and implemented in the ORIGEN code by Williams and Weisbin [21]. This implementation of perturbation theory assumes that the where NT is the transpose of the forward solution (column) of the nuclide composition vector and N* is the adjoint solution of the nuclide vector solved for a user defined response R, and AðaÞ is the transmutation matrix containing transition elements a that define the energy-average neutron cross sections, the fission product yields, and decay transition data [18]. This expression is integrated from t ¼ 0 to t ¼ Tf . Note that the partial derivative of AðaÞ with respect to a results in zeros in the matrix locations where the data element a does not contribute. The sensitivity coefficients in Eq. (2) give the absolute change in a defined response R to a change in data parameter a. The response can be defined for an individual isotope or aggregate isotopes that contribute to an integral response such as decay heat, gamma emission, or other spent fuel properties. The forward transmutation equation is expressed in matrix notation as nuclide field is uncoupled from the neutron flux field, i.e., a perturbation of the nuclide field does not significantly affect the flux field, and each can be varied independently. A more general dNðtÞ ¼ A NðtÞ; t2 0;Tf ; (3) perturbation formulation for the spaceeenergy-dependent burnup with the solution I.C. Gauld et al. / Nuclear Engineering and Technology 49 (2017) 1226e1233 1231

Tài liệu liên quan