Oriel — AI & technology worth noticing, every day.

Generative Reconstruction of Full-Field Sensor States for Safety-Critical Digital Twins

Paper · https://arxiv.org/abs/2611.08499

arXiv:2611.08499v1 Announce Type: new Abstract: Accurate reconstruction of incomplete sensor fields is critical for state awareness in safety-critical digital twins (DTs). Thermocouples in high-temperature and radiation environments may drift, fail, or become intermittently unavailable, causing partial observability. This study develops a sparse-sensor reconstruction framework for Oregon State University's High Temperature Test Facility (HTTF) and evaluates six methods under intra- and inter-experiment protocols: three conditional generative models (i.e., a diffusion model [DM], variational autoencoder [VAE], and generative adversarial network [GAN]), two deterministic recurrent neural-network models, and ridge regression. For intra-experiment reconstruction within the PG{-}29 depressurized conduction cooldown (DCC) test, the models combine graph-based spatial encoding with gated recurrent units. The graph neural network (GNN)--gated recurrent unit (GRU) VAE achieves the lowest error, with a mean absolute error (MAE) of $11.73,^\circ$C, a root mean square error (RMSE) of $20.54,^\circ$C, and an $R^2$ of 0.98. The deterministic GNN--GRU and plain GRU obtain RMSE values of $25.26,^\circ$C and $35.22,^\circ$C, respectively, demonstrating the benefit of spatial modeling. All three generative models reduce RMSE by more than half compared with ridge regression. The DM maintains an $R^2$ of 0.94 when known sensors decrease from 91 to 45. For inter-experiment reconstruction, models trained on PG{-}27 pressurized conduction cooldown (PCC) are evaluated on PG{-}29 using section-context conditioning. The section-context DM achieves the best transfer performance, with an RMSE of $42.84,^\circ$C and an $R^2$ of 0.91. These results demonstrate the value of spatiotemporal generative modeling for missing-sensor reconstruction in safety-critical DT applications.

First seen by Oriel on .

AI summary

This study evaluates six methods for reconstructing missing thermocouple fields in safety-critical digital twins, finding that spatiotemporal generative models provide accurate intra- and inter-experiment recovery.

  • The framework targets incomplete measurements from Oregon State University's High Temperature Test Facility, where thermocouples can drift, fail, or become intermittently unavailable.
  • For intra-experiment PG-29 reconstruction, the GNN–GRU VAE achieves 11.73°C MAE, 20.54°C RMSE, and an R² of 0.98.
  • Spatial modeling improves deterministic performance: GNN–GRU and plain GRU reach RMSE values of 25.26°C and 35.22°C, respectively.
  • The diffusion model maintains an R² of 0.94 when known sensors are reduced from 91 to 45, while generative models halve ridge regression's RMSE.
  • For transfer from PG-27 to PG-29, section-context conditioning produces the strongest result, reaching 42.84°C RMSE and an R² of 0.91.

#digital twins #sensor reconstruction #generative models #graph neural networks

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
arXiv AI Papersoai:arXiv.org:2611.08499v1https://arxiv.org/abs/2611.08499 Read discussion