Hardness Prediction for Temper Bead Welding of Non-Consistent Layer Technique

Abstract

Temper bead welding (TBW) is one effective repair welding method for the large-scale nuclear power plants. Consistent Layer (CSL) technique is the theoretically most authoritative method among the five temper bead welding techniques. However in the actual operation, CSL technique is difficult to perform, and non-CSL techniques (Controlled Deposition technique, Half Bead technique, et al) are mainly used in the actual repair process. The thermal cycles in heat affect zone (HAZ) of non-CSL technique are more complicated than that of CSL techniques. Through simplifying the complicated thermal cycles to 4 types of thermal cycles, the neural network-based hardness prediction system for non-CSL techniques has been constructed. The hardness distribution in HAZ of non-CSL techniques was calculated based on the thermal cycles numerically obtained by finite element method (FEM). The predicted hardness was in good accordance with the experimental results. It follows that the thermal cycle simplification methods are effective for estimating the tempering effect during temper bead welding of non-CSL techniques.