Table 2 Model performance comparison.

From: Optimizing load demand forecasting in educational buildings using quantum-inspired particle swarm optimization (QPSO) with recurrent neural networks (RNNs):a seasonal approach

 

Traditional RNN

QPSO-RNN before optimization

QPSO-RNN after optimization

MAE

30.5

25.49

15.2

MSE

1250.65

1010.28

520.15

RMSE

35.36

31.78

22.8

R²

0.28 (28%)

0.34 (34%)

0.93 (93%)

MAPE

30.75%

26.22%

9.75%

sMAPE

25.50%

22.42%

8.30%

MASE

0.7

0.57

0.45

Back to article page