On the Prediction of Flow Stress Behavior of Additively Manufactured AlSi10Mg for High Temperature Applications

The high-temperature deformation behavior of laser powder bed fabricated (LPBF) AlSi10Mg alloy was investigated using an isothermal hot compression test over a wide range of deformation conditions (150-300 degrees C and 0.01-1 s-1). Different phenomenological models, namely the Johnson-Cook model, modified Johnson-Cook, strain-compensated Arrhenius equation, modified Zerilli-Armstrong model, modified Fields-Backofen model and artificial neural network (ANN) with feed-forward back propagation learning algorithm, were used for predicting the flow stress dependency on strain, strain rate, and temperature. The accuracy of the predictive capability of these models was determined using different statistical parameters such as correlation coefficient (R), average absolute relative error, and root mean square error. The modified Fields-Backofen model and strain-compensated Arrhenius model were identified as the best-suited models for predicting the flow stress behavior of additively manufactured AlSi10Mg, with an average error of 3.3% and 3.9% and correlation coefficient of 0.96 and 0.97, respectively. The ANN model exhibited the highest accuracy in predicting the hot deformation behavior of LPBF-fabricated AlSi10Mg, with an average error of 0.5% and a correlation coefficient of 0.99.