Table 5 Performance metrics of different models (Lanzhou).
Dataset | Model | RMSE /(µg/m3) | MAE /(µg/m3) | SMAPE /(%) | PCC | DA | MBE |
|---|---|---|---|---|---|---|---|
Lanzhou | CNN | 11.9206 | 7.4553 | 21.65% | 0.6921 | 0.4058 | 1.1967 |
Elman | 11.4293 | 6.5316 | 18.66% | 0.7205 | 0.4746 | -0.2394 | |
LSTM | 11.5811 | 6.9034 | 19.73% | 0.7193 | 0.4167 | 1.2016 | |
BiLSTM35 | 11.3894 | 6.4826 | 18.63% | 0.7231 | 0.4275 | -0.4265 | |
GRU | 11.6285 | 6.6551 | 18.80% | 0.7148 | 0.4094 | -0.1408 | |
MSAGRU | 11.2284 | 6.5533 | 18.94% | 0.7309 | 0.4275 | 0.0151 | |
CNN-GRU56 | 11.7766 | 7.3760 | 21.55% | 0.6970 | 0.4565 | 0.6116 | |
CEEMDAN-GRU | 7.1517 | 4.2856 | 13.67% | 0.9003 | 0.7174 | -0.2237 | |
STL-MSAGRU | 5.2494 | 3.2411 | 9.97% | 0.9475 | 0.6957 | -0.1645 | |
STL-LSTM | 6.0625 | 3.3259 | 9.64% | 0.9302 | 0.7428 | -0.0977 | |
STL2-LSTM | 3.8915 | 2.4367 | 8.16% | 0.9722 | 0.7790 | -0.0390 | |
3D CNN-GRU36 | 10.5887 | 6.8007 | 19.80% | 0.7749 | 0.5797 | 1.6954 | |
EEMD-ALSTM69 | 5.6154 | 4.0844 | 14.74% | 0.9432 | 0.7246 | -1.3517 | |
FSR-MSAGRU | 0.5246 | 0.1875 | 0.63% | 0.9995 | 0.9928 | -0.0936 |
