Presented by : Jim Gulliford

1
(No Transcript)
2
Summary of UK PIE dataRichard Moore
Presented by Jim Gulliford
3
Overview

• Sellafield Dataset
• Other Data
• CERES, UK BUC validation measurements
• Consistency, Completeness, Uncertainties
• Need for more data
• Classification suitability for benchmarking
• Lessons Learned
• UK participation in OECD-NEA PIE Experts Group

4
Sellafield Dataset

• PIE data used to validate spent fuel inventory
  calculations to support UK nuclear operations
• Database of PIE measurements from around the
  world
• Database includes the results of calculations
  performed
• WIMS TRAIL FISPIN
• Classification of validation data

5
Sellafield database

• Includes
• Experimental result
• Calculated result
• C/E
• Cooling
• Enrichment
• Assembly and sample irradiation
• Laboratory
• Classification
• Calculation code and details

6
LWR PIE Summary
7
LWR MOX Data
pulled-pin samples not used for benchmarking
8
Other PIE Data

• Burnup Credit Programmes at Winfrith
• Used to valid WIMS MONK
• CERES
• Reactivity and PIE measurements on PWR BWR
  samples from France USA
• PIE included analysis of 15 major BUC fission
  products
• Pre-CERES
• Reactivity and PIE on HEU research reactor fuel,
  AGR PWR (Zorita Besnau) samples
• UK ready to make data available (need to get
  agreement from US and French partners for CERES
  data)

9
CERES Reactivity Measurements fuel samples
10
CERES Reactivity Measurements FP samples
11
CERES PIE Analysis

• Result for Sm149 appears to be due to problem
  with measurement. Later PIE work gave much better
  agreement

12
Inter lab agreement
Sometimes excellent, sometimes not
13
Inter-lab agreement

• For measurements made in two laboratories
• Some studies show up to 77 of results agree to
  within 2-sigma errors
• Statistically it should be 95
• 77 is good when compared with other studies
• 36 for fission products
• 22 for actinides
• Demonstrates a problem with measurements or
  uncertainty estimation need some other means to
  assess the reliability of validation data.

14
Data classifications

• Class A Most consistent and reliable data
  laboratory cross checks performed and consistent
• Class B Multiple laboratory measurement on
  dissolved sample and results consistent
• Class C Single laboratory measurement on
  multiple similar samples and results consistent
• Class D Reliable data as assessed by experts,
  without laboratory cross check
• Class F Results unsuitable for validation

15
Classification overview
Actinides
Fission products
16
Lessons Learned

• Chemical separation process is very delicate
  (particularly for Fission Products) good idea to
  get independent verification
• Pulled-pin irradiations difficult to analyse
  try to avoid if possible
• Rh chemistry difficult we have experienced
  problems with sample manufacture and PIE
• Need to do thorough check on completeness of
  description of irradiation history environment
• Inconsistent results - measurement uncertainty
  analysis appears incomplete in some cases (i.e.
  uncertainties in chemical separation)

17
UK Participation in PIE Experts Group

• Donation of data as/when available
• Review of others contributions
• Seek to identify remaining UK expertise in
  chemical separation to add to lesson learned
• Build consensus on reliable experimental
  techniques
• Benchmark new data and present summary of new and
  old benchmark results
• Identify gaps in database
• Highlight problem areas in calculations

18
Finally

• Currently investigating whether we can give all
  this information for use in SFCOMPO
• Hopefully we will be able to provide the data
  soon
• Suggest inclusion of similar procedure to ICSBEP
  where evaluation includes results of indicative
  calculation results
• Provides test of completeness of data and gives
  early indication of gross errors