Short-Term Power Load Forecasting Based on SAPSO-CNN-LSTM Model considering Autocorrelated Errors

Table 4

Comparison of model prediction results.

Prediction model	RMSE(1d)/kW	MAPE(1d)	RMSE(7d)/kW	MAPE(7d)
LSTM	182.14	2.16%	215.92	2.42%
CNN-LSTM	134.85	1.45%	182.58	1.76%
PSO-CNN-LSTM	101.53	1.15%	138.50	1.47%
SAPSO-CNN-LSTM	89.26	0.97%	134.86	1.34%
The proposed model	79.55	0.87%	122.03	1.20%