What does this research mean for the field?

A physics-informed, multi-sensor machine learning approach utilizing arc-cycle feature extraction can predict process instabilities in wire arc additive manufacturing with over 95% accuracy and allows for partial transferability to complex geometries. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

September 6, 2026Journal of Intelligent ManufacturingOpen Access

Physical characterization and detection of process instabilities in wire arc additive manufacturing through arc-cycle feature extraction

View Full Paper

Ask AI

Bookmark

Authors

ARAndré Ramalho BBBenjamin Bevans DSDouglas Serrati

Discussion

Member takes

Overview

Experimental evaluation demonstrates machine learning detection of additive manufacturing instabilities using multi-sensor arc features, indicating high transferability across complex geometries.

Key Points

To characterize and detect wire stubbing and droplet overgrowth process instabilities in wire arc additive manufacturing using multi-sensor monitoring and physics-informed feature extraction.
Acquired electrical current, voltage, acoustic signatures, and high-speed melt pool imaging data during wire arc directed energy deposition.
Extracted six physical features across arcing, short circuit, and arc ignition phases to train a support vector machine model for distinguishing stable and unstable states.
Conducted ablation studies across sensing modalities and evaluated model transferability to complex geometries without recalibration.
The support vector machine model predicted the onset of process instability in thin walls with statistical accuracy exceeding 95%.
Ablation testing confirmed that acoustic signal information complements electrical signals for identifying unstable deposition states.
Transferring the uncalibrated model to complex geometries maintained detection accuracy, with false positive rates increasing from 2.3% to ~4.5% due to transient melt pool conditions.

Cite This Study

Ramalho et al. (2026) studied this question.

synapsesocial.com/papers/6a9d1e3328139818eab20df5 https://doi.org/10.1007/s10845-026-02967-4

View Full Paper

Ask AI

Bookmark

September 6, 2026Journal of Intelligent ManufacturingOpen Access

Physical characterization and detection of process instabilities in wire arc additive manufacturing through arc-cycle feature extraction

View Full Paper

Ask AI

Bookmark

Authors

ARAndré Ramalho BBBenjamin Bevans DSDouglas Serrati

Discussion

Member takes

Overview

Experimental evaluation demonstrates machine learning detection of additive manufacturing instabilities using multi-sensor arc features, indicating high transferability across complex geometries.

Key Points

To characterize and detect wire stubbing and droplet overgrowth process instabilities in wire arc additive manufacturing using multi-sensor monitoring and physics-informed feature extraction.
Acquired electrical current, voltage, acoustic signatures, and high-speed melt pool imaging data during wire arc directed energy deposition.
Extracted six physical features across arcing, short circuit, and arc ignition phases to train a support vector machine model for distinguishing stable and unstable states.
Conducted ablation studies across sensing modalities and evaluated model transferability to complex geometries without recalibration.
The support vector machine model predicted the onset of process instability in thin walls with statistical accuracy exceeding 95%.
Ablation testing confirmed that acoustic signal information complements electrical signals for identifying unstable deposition states.
Transferring the uncalibrated model to complex geometries maintained detection accuracy, with false positive rates increasing from 2.3% to ~4.5% due to transient melt pool conditions.

Cite This Study

Ramalho et al. (2026) studied this question.

synapsesocial.com/papers/6a9d1e3328139818eab20df5 https://doi.org/10.1007/s10845-026-02967-4

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark