Using probabilistic methods for solving the problem of ensuring leak tightness of heat exchanger tubes of nuclear power plant steam generators - Probabilistic Safety Assessment for Optimum Nuclear Power Plant Life Management (PLiM) - page 299

Table 9.5 Parameters of defect distribution

Number of defects

Number
of steam
generators

Density of defects
per tube

Number of inspected tubes

MM CE Total

1 50 - 50 1608 0.0311
2 71 13 84 2838 0.0295
3 43 87 130 5518 0.0235
4 30 13 43 819 0.0525
5 30 12 42 5511 0.0076
6 5 2 7 883 0.0079
Comment . MM - 'lack of metal' defect'; CE - 'corrosion below the grid'-type defect

Statistical analysis of the defect size was performed for the variants of
set restriction of the maximum depth of the defect - 100%, 80%, 60%, 40%
in relation to wall thickness of the heat exchanger tubes.
Table 9.7 presents the results of calculations of the mathematical
expectations and variances (standard deviations) with the uncertainty of
the initial data on the statistical characteristics of defects, such as 'lack
of metal' (MM < 100%) for the elements of tubes of the steam generator
of power unit No. 1 of the Kola nuclear power plant, taking into account
the uncertainty of initial data.
Statistical analysis of defect sizes showed that the Weibull distribution
is the most suitable theoretical law for describing the depth distribution
of defects.
The law of distribution of the defect length was used to analyse the
propagation of defects in the metal of heat exchanger tubes based on the
experiments carried out at the Novovoronezh NPP. In this case, the length
of defects was described by a normal distribution with the mathematical
expectation of 13.5 mm and a standard deviation of 4.55 mm.
Performance analysis of heat exchanger tubes assumes that the rate
of nucleation of new defects in future will be the same as in 2002. This
premise requires mandatory testing, either through application of the model
of nucleation of defects described above, or by obtaining and processing
new data resulting from eddy current testing of tubes.
The following indices and parameters of the Paris equation were
used for the analysis of corrosion-fatigue defect growth: the exponent
for steel 08Cr18Ni10Ti m 0 = 3.53; experimental factor for a given
material in fatigue testing in air C 0 was assumed to be a random variable
obeying the log-normal distribution law with parameters q = -26.86 and
w = 0.6215; coefficient A 0 takes into account the influence of corrosive
environment on the fatigue crack growth rate. The value of this coefficient
for the contact zones of the heat exchanger tubes with spacing elements
is accepted as equal to 20 (higher concentration), and for the remaining
sections of the tube as 10.

￿ ￿ ￿ ￿ ￿