Improved model for predicting J-R curves from charpy data: Final report

It is necessary to estimate J-R data for aging and residual life studies, using data from surveillance capsule specimens that reflect the current state of the pressure vessel steel. An improved model is needed that will predict J-R data from Charpy data, tensile properties, material composition, temperature, and fluence. This report describes Phase I of a project to develop such a model, using innovative pattern recognition and surface-fitting codes to identify the key parameters and optimal modeling functions. In Phase I, a readily-available sample of data on reactor pressure vessel and piping steels was collected and analyzed by the pattern recognition code. A preliminary model was fitted to the data, producing a more convenient model that gives a better fit with fewer fitting constants, compared to earlier models. The results are sufficiently encouraging to justify a Phase II project which would produce a calibrated, validated model for J-R data, with significantly better predictive capability and less data scatter than any model now available. The data available for a Phase II effort are evaluated, and found to be adequate for weld metal modeling. A modest testing effort would make the data base adequate for calibrating separate weld and base metal models.more » 26 refs., 14 figs., 2 tabs.« less