Plasma Spray Process Parameters Configuration using Sample-efficient Batch Bayesian Optimization

Guidetti, Xavier; Rupenyan, Alisa; Fassl, Lutz; Nabavi, Majid; Lygeros, John

doi:10.1109/case49439.2021.9551499

Cited by 7 publications

(13 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Details about the process data-driven modeling are available in [30]. Our goal is to select values for six controllable process inputs to regulate the coating microhardness and porosity and maximizing the equipment lifetime.…”

Section: Atmospheric Plasma Spraying Configurationmentioning

confidence: 99%

See 1 more Smart Citation

Advanced Manufacturing Configuration by Sample-Efficient Batch Bayesian Optimization

Guidetti¹,

Rupenyan²,

Fassl

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

We propose a framework for the configuration and operation of expensive-to-evaluate advanced manufacturing methods, based on Bayesian optimization. The framework unifies a tailored acquisition function, a parallel acquisition procedure, and the integration of process information providing context to the optimization procedure. The novel acquisition function is demonstrated, analyzed and compared on state-of-the-art benchmarking problems. We apply the optimization approach to atmospheric plasma spraying and fused deposition modeling. Our results demonstrate that the proposed framework can efficiently find input parameters that produce the desired outcome and minimize the process cost.

show abstract

Section: Atmospheric Plasma Spraying Configurationmentioning

confidence: 99%

“…To conduct simulated studies, we use the neural network model structure and data set from [30]. The neural network simulates the behavior of the APS machine and acts as an oracle during the optimization process, returning the microhardness and porosity of virtual coated samples.…”

Section: A Simulated Process Optimizationmentioning

confidence: 99%

Advanced Manufacturing Configuration by Sample-Efficient Batch Bayesian Optimization

Guidetti¹,

Rupenyan²,

Fassl

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Including all measurable process parameters in the modeling and optimization makes it possible to prevent these deviations during the configuration process. Here, we outline a generalized version of the method that was proposed and detailed in [27].…”

Section: B Status-aware Optimizationmentioning

confidence: 99%

“…application rate, thickness, porosity, microhardness) depend on multiple process input parameters [28]. Details about the process data-driven modeling are available in [27]. Our goal is to select values for six controllable process inputs to regulate the coating microhardness and porosity and maximizing the equipment lifetime.…”

Section: Atmospheric Plasma Spraying Configurationmentioning

confidence: 99%

See 1 more Smart Citation

Advanced Manufacturing Configuration by Sample-efficient Batch Bayesian Optimization

Guidetti¹,

Rupenyan²,

Fassl³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We propose a framework for the configuration and operation of expensive-to-evaluate advanced manufacturing methods, based on Bayesian optimization. The framework unifies a tailored acquisition function, a parallel acquisition procedure, and the integration of process information providing context to the optimization procedure. The novel acquisition function is demonstrated and analyzed on benchmark illustrative problems. We apply the optimization approach to atmospheric plasma spraying in simulation and experiments. Our results demonstrate that the proposed framework can efficiently find input parameters that produce the desired outcome and minimize the process cost.

show abstract

Spraying Parameters Selection Based on Predicted Equipment Status: A Study on Measured Voltage

et al. 2022

Self Cite

View full text Add to dashboard Cite

Previous work conducted on atmospheric plasma spraying has shown the importance of including the measured gun voltage in the modeling procedure to improve the outputs prediction quality. Given a set of controllable input parameters, the produced coating specifications are influenced by the gun voltage measured during the spraying process. As the gun voltage can only be measured once the coating process has started, making predictions about the expected voltage is necessary to better select the process inputs that produce a coating with desired specifications. We suggest that the gun voltage is related to the status of the manufacturing equipment. Exploiting voltage information, we propose a modeling and configuration procedure that uses Gaussian process regression and Kalman filtering to reduce the impact of session-to-session equipment changes as well as in-session equipment wearing. We then demonstrate this approach in simulation and experiments, using an industrial atmospheric plasma spraying set-up to produce YSZ coatings.

show abstract

Plasma Spray Process Parameters Configuration using Sample-efficient Batch Bayesian Optimization

Cited by 7 publications

References 19 publications

Advanced Manufacturing Configuration by Sample-Efficient Batch Bayesian Optimization

Advanced Manufacturing Configuration by Sample-Efficient Batch Bayesian Optimization

Advanced Manufacturing Configuration by Sample-efficient Batch Bayesian Optimization

Spraying Parameters Selection Based on Predicted Equipment Status: A Study on Measured Voltage

Contact Info

Product

Resources

About