Hybrid Machine Learning Techniques and Computational Mechanics: Estimating the Dynamic Behavior of Oxide Precipitation Hardened Steel

A new generation of Oxide Dispersion Strengthened (ODS) alloys called Precipitation Hardened (OPH) alloys, has recently been developed by the authors. The excellent mechanical properties can be improved optimizing chemical composition in combination with heat treatment. However, behavior such materials requires consideration a large number variables, nonlinearities, and uncertainties analyses, modeling analytical methods is not accurate enough. Therefore, artificial intelligence (AI) methods, as machine learning (ML), beneficial to alleviate problems associated complexity these alloys. In this work, three different hybrid ML techniques have employed estimate ultimate tensile strength (UTS) elongation special proposed include feedforward neural network (FF-ANN) trained using particle swarm optimization (PSO) two adaptive neuro-fuzzy inference system (ANFIS) both fuzzy C-means (FCM) clustering subtractive (SC). Since OPH are mainly produced via alloying (MA) mixture powder components followed consolidation hot rolling, series standard tests were performed on variants alloy. way, some critical parameters UTS could extracted from experimental results. main contribution present study important based material including Aluminum (Al), Molybdenum (Mo), Iron (Fe), Chromium (Cr), Tantalum (Ta), Yttrium (Y) Oxygen (O), MA treatment conditions. results show that strategies only able accurately determine complex alloy an accuracy about 95%, but they also help designer benefit powerful tools design versions without calculations.