1 / 20

Multi-factorial analysis of occupational outage doses dispersion in the French NPPs 1998-2002

Multi-factorial analysis of occupational outage doses dispersion in the French NPPs 1998-2002. C. Lefaure, L . D’Ascenzo, P. Crouail, CEPN G. Cordier, J. Lebeau, A. Rocher, G. Machicoane, EDF. OBJECTIVES OF THE STUDY.

masato
Download Presentation

Multi-factorial analysis of occupational outage doses dispersion in the French NPPs 1998-2002

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Multi-factorial analysis ofoccupational outage doses dispersion in the French NPPs 1998-2002 C. Lefaure, L . D’Ascenzo, P. Crouail, CEPNG. Cordier, J. Lebeau, A. Rocher, G. Machicoane, EDF

  2. OBJECTIVES OF THE STUDY • Since 1992, at EDF, many efforts in reducing occupational radiological exposures. • Without knowing what are most important factors • In depth statistical analysis of influence of factors explaining outage doses dispersion in France, • To determine which are still potential levers for continuing dose optimisation.

  3. METHOD • First step: data from 137 outages (1998 – 2000); • Second step: extension to 2001 and 2002, has allowed data from 226 outages. • 11 qualitative variables and • 39 quantitative variables

  4. METHOD • reactors design features (10 variables), • characteristics of the operation of the plants (15 variables), and • characteristics of the shutdown (15 variables), • and of theoutages themselves (13 variables) • Many variables similar to those requested in the ISOE level 2 questionnaire • (materials and components, primary water chemistry and contamination levels, hot spots…)

  5. METHOD • Analysis of the qualitative variables influence on outage doses, • Analysis of correlations between doses and each quantitative variable. A well as quantitative analysis of the evolutions, • Use of STATGRAPHICS Plus software. • Sample as a whole; each type of French PWR ; each major type of outages (simple refuelling outage (ASR), short maintenance outage (VP) and long ten years outage (VD)

  6. QUALITATIVE VARIABLES ANALYSIS

  7. QUALITATIVE VARIABLES ANALYSIS • Very important impact of the design • Type of reactor (30%) • statistically significant impact of SG tubes material and fabrication, and SG channel head electropolishing • no relationship with MOX fuel, which let consider that there is no significant impact of the MOX fuel on the source term

  8. QUALITATIVE VARIABLES ANALYSIS • Impact of “reactor boron and water make up system” type • Two types of covers (on the 900 MWe) with air or nitrogen • No noticeable impact • Impact of hot spots presence: ± 30 % extra dose • Normalisation of doses by 1.3

  9. QUANTITATIVE VARIABLES ANALYSIS • Time spent in controlled area is by far the most important variable (cc >.7) at all levels of analysis • Should be important to follow other variables influencing the time spent: mishaps, reworks, fortuitous works… • Comfort the expectations from ISOE when issuing a book on “work management” • Time spent better approximates the exposed workload than length of outage in terms of days

  10. QUANTITATIVE VARIABLES ANALYSIS • Most other variables impacts hidden by the time spent • Use the dose index to go further • Ratio « outage dose divided by time spent » • Impact of other variables become significant to explain the dispersion of the dose index.

  11. QUANTITATIVE VARIABLES ANALYSIS • The radiological state of the primary circuit (average of dose rates measured following EPRI SMRP recommendations) • Not as powerful as “time spent”, for explaining dose dispersion (role of auxiliary circuits) • But for ten years outages

  12. QUANTITATIVE VARIABLES ANALYSIS • cobalt 60 very important during the first cycles • it remains a significant pollutant impacting dose rates and doses with ageing of the units as seen in the study for the 900 MWe units. • importance of stellite reduction programmes for that type of reactors • importance of reducing CO59 content in the steels at both design and modification stages.

  13. QUANTITATIVE VARIABLES ANALYSIS • other radio-elements potentially contributing to dose through gamma emissions, become also significant contributors for the 1300 MWe units as cobalt is less important. • It re-emphasises the importance of a circuit purification strategy not only focussed on cobalt, but adapting the chemical specifications according to the type of pollutant (Ag, Sb…). • EDF has developed such a strategy that will be now implemented.

  14. QUANTITATIVE VARIABLES ANALYSIS • The time spent in the controlled area by the health physicists: an important variable for the 900 MWe units. • The highest the time, the lowest are the doses with quite good correlations whatever the type of outage (ASR, VP or VD). • Why not for the 1300MWe ? • Average time +66% with smaller dispersion • How to increase time spent in the 900 Mwe units?

  15. QUANTITATIVE VARIABLES ANALYSIS • Time spent by the managers was also introduced as a variable • Unfortunately, the correlations are quite good but positive • It does not mean that the presence of the mangers is not useful, • It just demonstrate that the longer are the outages, the highest the time spent in the controlled area by the managers and the highest the outage doses.

  16. QUANTITATIVE VARIABLES ANALYSIS • In a first time (three first years), the time spent in the controlled area by the individuals in charge of decontamination has appeared important for the 1300 MWe units. • Not any more the case with two more years. • It corresponds to an important reduction of the dispersion of the time spent during the different outages (by more than a factor 3).

  17. QUANTITATIVE VARIABLES ANALYSIS • number of rapid and important “load modifications” during the cycle, were introduced to check if the corresponding water movements may have introduced a significant oxidation and therefore an increase of the dose rates and doses. • Correlations are no more significant with these variables than wih the qualitative variable “reactor boron and water make up system” type .

  18. QUANTITATIVE VARIABLES ANALYSIS • Another variable for checking the impact of the corrosion: the number of days during the cycle with pH lower than 6.9. • Expected that the higher that number, the higher the acidity, the corrosion in the primary circuit, the higher the dose rates nearby the primary circuit • in most situations no significant correlation with either the dose index or the radiological state of primary circuit have been observed • but for the 1300 MWe during short outages only during the first period of the analysis (1998-2000) with both the radiological state of primary circuit and the dose index

  19. QUANTITATIVE VARIABLES ANALYSIS • that sub-sample shows an important increase of the average number of days between 1999 and 2000 (from 12 days to 23) with an increase of the dispersion, while it remains stable from 2000 to 2002. • even if some phenomenon’s are obvious (impact of acidity on corrosion…) the discrepancies between the units on the selected indicators are often not enough important to show significantly that they play a role for explaining the dispersion in terms of outage doses or dose index

  20. CONCLUSION • just examples of what can be done with such a study, both in France and at the international level with ISOE2 data • It has pointed out interest of mixing qualitative and quantitative variables; descriptive statistical analysis with the analysis of correlation dispersions as well as there evolution

More Related