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

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,^○C, a root mean square error (RMSE) of 20.54,^○C, and an R2 of 0.98. The deterministic GNN–GRU and plain GRU obtain RMSE values of 25.26,^○C and 35.22,^○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 R2 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,^○C and an R2 of 0.91. These results demonstrate the value of spatiotemporal generative modeling for missing-sensor reconstruction in safety-critical DT applications.