Abstract ▼ AI-Helper

Since the failure of steam generator (S/G) tubes affects the safety of a nuclear power plant directly and severely, it is very important to sustain the structural integrity of S/G tubes above the required level. Deterministic structural integrity assessment methods have been used for the evaluation ...

Since the failure of steam generator (S/G) tubes affects the safety of a nuclear power plant directly and severely, it is very important to sustain the structural integrity of S/G tubes above the required level. Deterministic structural integrity assessment methods have been used for the evaluation of the structural integrity of S/G tubes. However the deterministic methods usually show conservative results compared with probabilistic methods. Therefore in this paper, the integrity of S/G tubes is evaluated using probabilistic methods to obtain more realistic results. In order to assess the integrity of S/G tubes, it is necessary to evaluate the number and the sizes of physical flaws that actually exit in S/G tubes. The evaluation of flaw initiation and flaw growth, and the calculation of the probability of failure of S/G tubes are also necessary. And the analysis of uncertainties of non-destructive inspection (NDI) is required for the evaluation of the number of physical flaws and their sizes. In this paper these required processes are considered for the probabilistic evaluation of the structural integrity of S/G tubes. The uncertainty of NDI is quantitatively evaluated using the eddy current test results for outside diameter (OD) axial cracks. And the uncertainties due to the imperfection of analyst and inspection technique are distinguished quantitatively. The probabilistic crack growth rate is calculated using the in-service inspection (ISI) data. The crack growth rate is expressed as a function of crack length and a random variable is used in order to represent the variation of the value. The number of flaws detected at each ISI is influenced by the probability of detection (POD) of NDI system and related with ISI schedule. Therefore the relation between the number of detected flaws and POD is analyzed using the periodic ISI data. From this analysis the number of physical flaws can be estimated. In the estimation of the number of physical flaws, the effective POD is proposed. The effective POD is the total detection probability, which is calculated considering the number of ISI, ISI interval and the probabilistic flaw growth rate. The number of initiated flaws is predicted using Weibull distribution. The fluctuation of increasing pattern of the number of initiated flaws is quantitatively assessed using joint confidence region of Weibull parameters. And a method of master curve is also proposed to predict the number of initiated flaws even in the early stage of operation when there is little flaw data. The probability of failure for a cracked tube is calculated using the first-order reliability method (FORM). The rate of crack propagation and the resistance of tubes to cracks are considered as random variables in the calculation of probability of failure. In the evaluation of structural integrity of S/G using probability of failure, the method of relative importance is proposed. The influence of each cracked tube to the structural integrity of whole S/G can be quantified by the relative importance. When allowable probability of failure is designated, the repair and operation plans can be established based on the above results.