Computational Framework for Microstructure Identification and Hardness Prediction in Ultrahigh-Strength Steel Components Fabricated by Wire-Arc Directed Energy Deposition

This work demonstrates the application of an FEA-based modeling framework for identifying the microstructure and hardness of an ultrahigh-strength steel component fabricated by wire-arc directed energy deposition. The framework adapts an FEA process model using a new, calibrated dual double ellipsoid Gaussian heat source with microstructure identification and hardness prediction approaches used in previous temper bead welding research. The microstructure identification relied on predicted peak temperature sequences of supercritical, intercritical, and subcritical or tempering reheats. Microstructure-specific tempering response relationships derived from the modified Grange-Baughman approach predicted hardness. The framework was validated against metallurgical characterization and hardness measurements along the build height of a 64-layer, single bead per layer WA-DED wall. The calibrated DDEG heat source demonstrated reliable predictions of the multi-reheat thermal cycling, with average peak temperature differences of 3°C to 21°C when comparing 9 sequential predicted and in-situ measured reheat cycles. The microstructure identification procedure correctly identified three distinct regions along the build height: the as-solidified fresh martensite, super- and intercritically reheated fresh martensite heat affected zones, and tempered martensite heat affected zones. Combining the identified microstructures with previously developed tempering response relationships predicted the cyclic hardness variations in the tempered regions with 94.5% overlap between the 99% confidence interval of predicted hardness and the measured hardness range. The validated framework provides a basis for component property optimization by controlling local microstructures and tempering through WA-DED process parameters.