Failure probabilistic analysis of steam generator heat-transfer tubing with pitting corrosion

Pitting corrosion damages seriously steam generator (SG) heat-transfer tubing of nuclear power plants. To estimate accumulative failure probability of the SG heat-transfer tubing, a failure probabilistic analysis method with pitting corrosion of stochastic processes is proposed in this paper. Pitting corrosion can be considered as a combination process of pit initiation and growth. The pit generation process is built as a non-homogenous Poisson process. The pit growth process is modeled by a stationary gamma process. A cumulative failure probability model is derived from the stationary gamma process. The proposed model is evaluated with the maximum likelihood estimation method according to the experiment data. Finally, Monte Carlo method is applied to simulate the cumulative failure probability model. The results illustrated that the proposed method is feasible and practical to estimate the accumulative failure probability of the SG heat-transfer tubing.

► The cumulative failure probability can be formulated under conditions of great uncertainty using only four parameters.
► This is a drastic reduction compared to other statistical approaches in the literature which require a tremendous amount of data for the statistical models.
► The proposed modeling method is valid and practical and has the capability to estimate and simulate the pitting failure of the SG heat-transfer tubing.
► This paper introduced the integration of NHPP and gamma process to model stochastic deterioration, which has a great potential in optimizing preventive maintenance plans and repair plans.
► This is because gamma process is well suited for modeling the temporal variability of deterioration, and has proved to be very useful in determining optimal inspection and maintenance decisions.