Use of On-Line Fatigue Monitoring of Nuclear Reactor Components as a Tool for Plant Life Extension

Abstract

The application of an on-line fatigue monitoring system for tracking fatigue usage in operating power plants is described. The system, like several others which have been developed, uses the influence function approach, operates on a microcomputer, and determines component stresses using temperature, pressure, and flow rate data that are typically available from plant computers. Using plant-unique influence functions developed specifically for each component location, the system calculates stresses as a function of time and computes the fatigue usage. Stress values are calculated at time intervals defined by the user and the fatigue values are saved on files for use at a later time. The application of the GE Fatigue Monitoring System (GEFMS) to calculate fatigue usage in the feedwater nozzle of a GE Boiling Water Reactor is described in this paper. Stress predictions using GEFMS for a sample temperature transient show excellent agreement with results from finite element thermal and stress analysis performed on a mainframe computer. Fatigue usage calculations for a simulated 24-hr temperature record confirm that the system provides accurate results at a cost that is significantly lower than similar analysis done on mainframe computer systems. The system, which has been installed in a boiling water reactor plant, provides the technical basis to evaluate actual reactor conditions and justify plant life extension.

A neural network approach for determining spatial and geometry dependent Green's functions for thermal stress approximation in power plant header components

“…It is defined as the point at which the stress contribution predicted by G (P, t) decays to a value which is considered to be negligible. Here, t CH is taken as the time when G (P, t) is 5% [14] is the maximum value (σ). An exponential sum of terms (see equation (10)) is used in the present work to estimate unit temperature step stress responses.…”

“…The critical parameters are defined as the maximum stress value (σ, with units of MPa in the present work), the time at which the maximum stress is achieved (tσ, units of s here), and a characteristic time (t CH , taken to be the time at which the instantaneous stress value is 5% ofσ and having units of s here). It has been noted that defining an asymptotic value for the Green's function approximation can reduce the computational effort required to implement the method [14]. It is these critical parameters that will be used as outputs for the developed neural network.…”

“…The Green's function method (see section 2) has been shown to be a useful tool in predicting transient thermal stresses by several authors. In particular, the technique has been applied to fatigue analysis problems in the nuclear power industry [12,13,14].…”

On-Line and Off-Line Steam Turbine Component Strain States Monitoring for the Diagnostic System

“…Working out such variations by the e.g. Jain flow" method [2] allows an even more exact evaluation of the growth of life consumption in the analysed period.…”

A Comparison of Simple Methods to Incorporate Material Temperature Dependency in the Green’s Function Method for Estimating Transient Thermal Stresses in Thick-Walled Power Plant Components

“…The Green’s function method (see Section 2.2 ) has been show to be a useful tool in predicting transient thermal stresses by several authors. In particular, the technique has been applied to fatigue analysis problems in the nuclear power industry [ 14 , 15 , 16 ]. It has often been assumed (for simplicity of integration) that Green’s functions are temperature independent.…”