Convoluted filtering for process cycle modeling

Principles of materials science and engineering, physics, mathematics, information are used to extract knowledge insights from the process-structure–property-performance relationships hidden in data. The process-structure modeling can be accelerated without loss interpretability, with artificial intelligence tools that mimic salient features process relations. In this work, a novel convoluted model-filtering technique was exploited build successfully train Convoluted Filter (CoFi) artifacts for Fe-based alloy heat treatment cycles. were pre-trained filter out deep models change surrogate microstructure state after at ambient conditions. Direct representation thermal cycle within Graph facilitated development meaningful data evolution, which reduce overfitting limited datasets.