Gaussian Process Surrogates for Modeling Uncertainties in a Use Case of Forging Superalloys

Abstract

The avoidance of scrap and the adherence to tolerances is an important goal in manufacturing. This requires a good engineering understanding of the underlying process. To achieve this, real physical experiments can be conducted. However, they are expensive in time and resources, and can slow down production. A promising way to overcome these drawbacks is process exploration through simulation, where the finite element method (FEM) is a well-established and robust simulation method. While FEM simulation can provide high-resolution results, it requires extensive computing resources to do so. In addition, the simulation design often depends on unknown process properties. To circumvent these drawbacks, we present a Gaussian Process surrogate model approach that accounts for real physical manufacturing process uncertainties and acts as a substitute for expensive FEM simulation, resulting in a fast and robust method that adequately depicts reality. We demonstrate that active learning can be easily applied with our surrogate model to improve computational resources. On top of that, we present a novel optimization method that treats aleatoric and epistemic uncertainties separately, allowing for greater flexibility in solving inverse problems. We evaluate our model using a typical manufacturing use case, the preforming of an Inconel 625 superalloy billet on a forging press.

Active Learning as a Tool for Optimizing “Plug‐n‐Play” Electrochemical Atom Transfer Radical Polymerization

“…[ 41 ] First, Gaussian process regression (GPR) is used as a surrogate function to fit data from an unknown objective and the corresponding expected utility surface, which is maximized to select the next experiment. [ 42 ] Second, the probability improvement (PI) function is used to iteratively guide selection of the next unexplored data as the acquisition function. [ 43 ] Data generated is used to augment the training data which iteratively adapts the model until the optimum output ( Ɖ m in this case) is satisfied.…”

Assessment of ground motion amplitude scaling using interpretable Gaussian process regression: Application to steel moment frames

“…42 Furthermore, it is observed that many Bayesian regression models based on artificial neural networks converge to the Gaussian process with an infinite number of hidden units. 43 Due to the efficacy of GPRs, it has been widely used as surrogate models for various modeling purposes and applications (e.g., [44][45][46][47][48][49] ).…”

Combining Gaussian Process Regression with Poisson Blending for Seamless Cloud Removal from Optical Remote Sensing Imagery for Cropland Monitoring

“…Predictive uncertainty in GPR is epistemic uncertainty that indicates how confident the model is with respect to its predictions [51]. Thus, predictive uncertainty may not always be related directly to prediction errors.…”