Optimal control of build height utilizing optical profilometry in cold spray deposits

Chakraborty, Abhijit; Shishkin, Sergey; Birnkrant, Michael J.

doi:10.1117/12.2259997

Cited by 6 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The focus of our work is to develop a framework for feedback control of nozzle motion that minimizes the material wastage while achieving desired surface profile. While detailed CS models are highly complex, for our control development purposes it suffices to use a reduced order model described in Chakraborty et al (2017) which captures the impact of nozzle motion/spray behavior on the dynamics of deposition process at a macroscopic scale. Let p(t), h(t) and α(t) be the spray nozzle's, position and orientation as a function of time, respectively, see Fig.…”

Section: Cold Spray Control Problemmentioning

confidence: 99%

“…max and ω min ≤ ω t ≤ ω max , respectively. This optimal control problem can be solved using model predictive control (MPC), see Chakraborty et al (2017). However, since typically N d 1 to resolve the surface profile sufficiently, such an approach can be computationally demanding for real time deployment.…”

Section: Cold Spray Control Problemmentioning

confidence: 99%

“…The need for in-situ monitoring, machine learning and closed loop control has been recognized as a key enabler to improve process repeatability/reliability Tapia and Elwany (2014);Everton et al (2016); Qi et al (2019). Specifically, in the CS context, Chakraborty et al (2017) apply a model predictive control approach in simulations. We use a similar model/setup but employ a RL framework, and provide an experimental demonstration.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Guided Policy Search Based Control of a High Dimensional Advanced Manufacturing Process

Surana¹,

Reddy²,

Siopis³

2020

Preprint

View full text Add to dashboard Cite

In this paper we apply guided policy search (GPS) based reinforcement learning framework for a high dimensional optimal control problem arising in an additive manufacturing process. The problem comprises of controlling the process parameters so that layer-wise deposition of material leads to desired geometric characteristics of the resulting part surface while minimizing the material deposited. A realistic simulation model of the deposition process along with carefully selected set of guiding distributions generated based on iterative Linear Quadratic Regulator is used to train a neural network policy using GPS. A closed loop control based on the trained policy and in-situ measurement of the deposition profile is tested experimentally, and shows promising performance.

show abstract

Section: Cold Spray Control Problemmentioning

confidence: 99%

Section: Cold Spray Control Problemmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Guided Policy Search Based Control of a High Dimensional Advanced Manufacturing Process

Surana¹,

Reddy²,

Siopis³

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In the literature, examples of in-situ CS process monitoring and anomaly detection methods exist. Particle velocimetry [9], surface profilometry [10] , thermal imaging [11] ,…”

Section: Introductionmentioning

confidence: 99%

Aero-acoustic monitoring of cold spray additive manufacturing

Arkhipov

View full text Add to dashboard Cite

Cold spray, or supersonic particle deposition, is an emerging additive manufacturing process used to develop material on target substrates. The process continues to evolve toward full-fledged, metal 3D-printing applications. As these developments are made, there is a growing need for non-destructive process monitoring. Cold spray is dictated by numerous process parameters. Process consistency is vital for the development of highstrength coatings with predictable and repeatable microstructural, thermo-mechanical, and 3D printing properties (e.g., predictable shape). Detection of unexpected parameter variations, anomalous events, and spray failures is critical for identifying process inconsistencies, minimizing machine downtime, and maximizing cost efficiency. In this study, airborne acoustic (i.e., aero-acoustic) emissions were monitored during the cold spray process to examine acoustically detectable spray parameters. Characteristics of aero-acoustic waves that emanate from free and impinging supersonic jets were investigated. The viability of using these characteristics as real-time metrics for cold spray additive manufacturing applications was explored. Results showed that changes in nozzle gas supply pressure, nozzle gas inlet temperature, and main gas flow rate were detectable in free-jet experiments via frequency spectrum analyses. The influence of robotic arm motion and coating buildup on airborne acoustic signal trends was evaluated during cold spray deposition procedures. Band power and other signal features were extracted from recorded audio signals to compare temporal behavior. During deposition experiments, powder feeding rate (PFR) and standoff distance (SOD) variations were identified by acoustic characteristics. Nozzle wear experiments revealed the aero-acoustic detectability of nozzle wear conditions and rates. Delamination failures during spray procedures were identified in comparisons between normal and anomalous cases. An alarm program was developed for delamination detection through the acoustic monitoring of multiple frequency ranges. Audio signal analyses indicated the feasibility of acoustic event detection as a real-time process monitoring technique for the cold spray process.

show abstract

Reinforcement Learning: An Industrial Perspective

Surana

2021

Studies in Systems, Decision and Control

View full text Add to dashboard Cite

Optimal control of build height utilizing optical profilometry in cold spray deposits

Cited by 6 publications

References 5 publications

Guided Policy Search Based Control of a High Dimensional Advanced Manufacturing Process

Guided Policy Search Based Control of a High Dimensional Advanced Manufacturing Process

Aero-acoustic monitoring of cold spray additive manufacturing

Reinforcement Learning: An Industrial Perspective

Contact Info

Product

Resources

About