Creep Lifetime Prediction of Alloy 617 Using Black Box Machine Learning Approach

Abstract

This study explored the application of black box machine learning (ML) to build high throughput models that predict the creep response of Ni-based Alloy 617. Black box ML refers to highly complex machine learning algorithms that generate outputs from inputs without an interpretable internal process. The rapid implementation of suitable heat of alloys into targeted service is impeded by the extended qualification process involving chemistry-processing-structure-performance assessment. The ASME B&PV Code III outlines the requirement of 10,000 h of creep testing before each heat can be qualified for service and 30,000 h for heats that exhibit metastable phases. There is a critical need to shorten the development-to-deployment timeline for heats of an alloy at specific applications. Recent advancement in ML offers the ability to identify correlations in data which is difficult to elucidate by other approaches. To that end, black box ML is employed to expedite the HEAT qualification of Alloy 617 and predict performance from HEAT chemistry out to an unprecedented timescale. In this study, creep data for Ni-based Alloy 617—a solid solution strengthened material is gathered from a wide range of data sources. The alloy chemistry, phases, precipitates, and microstructural features are analyzed to obtain the key strengthening mechanism. Service conditions, mechanical properties, chemistry, and chemical ratios are provided as potential input features. The Pearson correlation coefficient with a 95% confidence bound is employed for input feature screening where poorly correlated inputs are eliminated to speed up the ML process and prevent under- and/or over-fitting of predictions. In the ML algorithm, the selected input features are regarded as predictors, and rupture is regarded as the response. An algorithm evaluation is performed where 20 ML algorithms are trained with the training set. The three-layered neural network (NN) was observed to be the best algorithm for the given data based on statistical rationale. The NN accurately predicts rupture across a range of isotherms and data sources. The interpolative and extrapolative predictions are in compliance with ECCC V5 guidelines. A post-audit validation exhibits neither under- nor over-fitting and confirms the applicability of NN algorithms to unseen data. The black box ML provides a pathway to predict the performance directly from chemistry and opens avenues to rapid heat qualification.