What question did this study set out to answer?

The aim is to develop a Physics-Informed Generative Adversarial Network (PI-GAN) for accurate thermal field data generation and anomaly detection in PBF-LB.

April 27, 2026Digital engineering.Open Access

Physics-informed GANs for thermal simulation and anomaly detection in PBF-LB

View Full Paper

Ask AI

Bookmark

Authors

SGShradha GhansiyalUniversity of Kaiserslautern SHSamarth HathwarUniversity of Kaiserslautern JPJacques PlatzUniversity of Kaiserslautern

Discussion

Member takes

Implication

Randomized trial shows improved anomaly detection and thermal simulations in PBF-LB, indicating better process quality management.

Key Points

The aim is to develop a Physics-Informed Generative Adversarial Network (PI-GAN) for accurate thermal field data generation and anomaly detection in PBF-LB.
Proposed a PI-GAN model integrating governing physical laws for thermal field data generation.
Trained an anomaly detection pipeline using the data generated from PI-GAN.
Focused on addressing limitations of traditional methods by reducing dataset size requirements.
Achieved realistic synthesis of thermal fields with limited training samples from PBF-LB data.
Enabled early detection of process deviations, allowing for timely interventions and quality control.
Reduced dependency on large experimental datasets while enhancing prediction accuracy.

Cite This Study

Ghansiyal et al. (2026) studied this question.

synapsesocial.com/papers/69eefd43fede9185760d4025 https://doi.org/10.1016/j.dte.2026.100110

View Full Paper

Ask AI

Bookmark

Physics-informed GANs for thermal simulation and anomaly detection in PBF-LB

Discussion

Member takes

Implication

Key Points

Cite This Study

Physics-informed GANs for thermal simulation and anomaly detection in PBF-LB

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider