Research Article
RAEF: An Imputation Framework Based on a Gated Regulator Autoencoder for Incomplete IIoT Time-Series Data
Table 2
Imputation performance on UAQ in MRE%.
| MP | MR (%) | Baseline | Ours | BM (%) | KNN | LIME-LSTM (%) | BRITS (%) | RFL (%) | RFG (%) | RFLR (%) | RFGR (%) |
| CM | 15 | 39.88 | — | 15.62 | 12.32 | 11.44 | 11.64 | 9.79 | 10.33 | 10 | 29.24 | — | 14.16 | 10.58 | 10.02 | 10.51 | 9.32 | 9.71 | 5 | 18.80 | — | 12.69 | 9.44 | 9.26 | 9.40 | 9.11 | 9.27 |
| UM | 15 | 45.28 | 32.10% | 13.29 | 11.70 | 10.04 | 10.78 | 9.12 | 9.37 | 10 | 32.47 | 23.84% | 12.00 | 10.10 | 8.86 | 9.48 | 8.61 | 8.79 | 5 | 19.35 | 15.88% | 11.46 | 8.85 | 8.35 | 8.61 | 8.40 | 8.26 |
|
|