Developing a neural network to model the mechanical properties of 13-8 PH stainless steel alloy
(1) Avantus Aerospace, Compton, California; Woodbridge High School, Irvine, California, (2) Global Youth Mission STEM Club, Brea, California
We systematically evaluated the effects of raw material composition, heat treatment, and mechanical properties on 13-8PH stainless steel alloy. 13-8PH is widely used in the aerospace industry for various structural components. Little is known about the predictive modeling of mechanical properties for 13-8 PH per heat treatment. To achieve the desired properties, multiple back-and-forth heat treatments and mechanical tests are often required which utilize significant amounts of resources. Thus, we hypothesized that no specific heat treatment temperature can be used to achieve the desired property ranges across all 13-8PH compositions. We heat treated 13-8PH alloy through solution treatment, cryogenic treatment, and final elevated temperature aging. We conducted double shear testing to measure this mechanical property. Consistent with expectations, experimental data confirmed that no single specific aging heat treatment temperature could be used to achieve desired property range for all compositions. A unique aging heat treatment temperature is required and there is a need to develop a model to predict heat treatment response for each composition. The results of the neural network models were in agreement with our experimental results and aided in the evaluation of the effects of aging temperature on double shear strength. The data suggests that this model can be used to determine the appropriate 13-8PH alloy aging temperature needed to achieve the desired mechanical properties, eliminating the need for many costly trials and errors through re-heat treatments.
This article has been tagged with: