Sample-efficient Plasma Spray Process Configuration with Constrained Bayesian Optimization

Recent work has shown constrained Bayesian optimization to be a powerful technique for the optimization of industrial processes. We adapt this framework to the set-up and optimization of atmospheric plasma spraying processes. We propose and validate a Gaussian process modeling structure to predict coatings properties. We introduce a parallel acquisition procedure tailored on the process characteristics and propose an algorithm that adapts to real-time process measurements to improve reproducibility. We validate our optimization method numerically and experimentally, and demonstrate that it can efficiently find input parameters that produce the desired coating and minimize the process cost.

[1]  Martin Friis,et al.  Control of thermal spray processes by means of process maps and process windows , 2003 .

[2]  Konrad Wegener,et al.  Self-optimizing grinding machines using Gaussian process models and constrained Bayesian optimization , 2020, The International Journal of Advanced Manufacturing Technology.

[3]  Zhenhua Wu.  Empirical modeling for processing parameters’ effects on coating properties in plasma spraying process , 2015 .

[4]  S. Lee,et al.  Influence of feedstock and spraying parameters on the depositing efficiency and microhardness of plasma-sprayed zirconia coatings , 2004 .

[5]  John Lygeros,et al.  Safe and Efficient Model-free Adaptive Control via Bayesian Optimization , 2021, ICRA 2021.

[6]  Peter I. Frazier,et al.  Parallel Bayesian Global Optimization of Expensive Functions , 2016, Oper. Res..

[7]  F. Gao,et al.  Optimization of Plasma Spray Process Using Statistical Methods , 2011, Journal of Thermal Spray Technology.

[8]  Konrad Wegener,et al.  Bayesian optimization for autonomous process set-up in turning , 2019, CIRP Journal of Manufacturing Science and Technology.

[9]  Ghislain Montavon,et al.  Artificial neural networks implementation in plasma spray process: Prediction of power parameters and in-flight particle characteristics vs. desired coating structural attributes , 2009 .

[10]  Ali Jalali,et al.  Hybrid Batch Bayesian Optimization , 2012, ICML.

[11]  A. Vaidya,et al.  Sensing, Control, and In Situ Measurement of Coating Properties: An Integrated Approach Toward Establishing Process-Property Correlations , 2009 .

[12]  D. Thirumalaikumarasamy,et al.  Effect of experimental parameters on the micro hardness of plasma sprayed alumina coatings on AZ31B magnesium alloy , 2015 .

[13]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[14]  Alkis Gotovos,et al.  Safe Exploration for Optimization with Gaussian Processes , 2015, ICML.

[15]  Ghislain Montavon,et al.  Artificial Neural Networks vs. Fuzzy Logic: Simple Tools to Predict and Control Complex Processes—Application to Plasma Spray Processes , 2007 .

[16]  Christian Moreau,et al.  The long-term stability of plasma spraying , 2002 .

[17]  Dilip Kumar Pratihar,et al.  Modeling of plasma spray coating process using statistical regression analysis , 2013 .

[18]  Matt J. Kusner,et al.  Bayesian Optimization with Inequality Constraints , 2014, ICML.

[19]  Andreas Krause,et al.  Safe controller optimization for quadrotors with Gaussian processes , 2015, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[20]  D. Ginsbourger,et al.  A Multi-points Criterion for Deterministic Parallel Global Optimization based on Gaussian Processes , 2008 .

[21]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.