Research Article
Performance Analysis of FFBP-LM-ANN Based Hourly GHI Prediction Using Environmental Variables: A Case Study in Chennai
Table 1
The statistical performance measure of the FFBP-LM-ANN model during training and testing.
| Model/Input features | Performance index | Expression | Training/(2016, 2017, 2018) | Testing/(2019) |
| FFBP-LM-ANN/SZA, RH, T, CC, Pp | MSE | | 0.004306 | 0.005078 | RMSE | | 0.06562 | 0.07126 | rRMSE | | 6.562% | 7.21% | MAE | | 0.039332 | 0.042448 | rMAE | | 8.7336% | 9.4829% | MAPE | | 38.6397% | 44.432% | MBE | | 0.00004002301 | 0.000492 | R | | 0.961542 | 0.959541 | MRE | | 8.73336% | 9.5566% | NSE | | 0.95 | 0.94 |
|
|