Research Article

A Comparison of Hour-Ahead Solar Irradiance Forecasting Models Based on LSTM Network

Table 9

Performance comparison between the proposed LSTM-MLP model and the general machine learning method.

ModelRMSE (W/m2)nRMSE (%)MAEMBER

BPNN76.927239.921533.2872−4.77980.9601
RNN77.844440.411540.0704−16.28640.9602
Random forest70.380836.536930.5618−0.95930.9659
SVM77.038439.993150.9564−1.67760.9593
General LSTM71.843437.296230.6385−1.95230.9649
The proposed LSTM-MLP62.161832.270226.6538−0.45470.9737