Principal Component Analysis Based Dynamic Fuzzy Neural Network for Internal Corrosion Rate Prediction of Gas Pipelines

Table 3

The RMSE, MAPE and TIC of the prediction values for the internal corrosion rate performance of models on the testing dataset.

Evaluation metrics	ANN	FNN	D-FNN	PCA-D-FNN
RMSE	0.6863	0.6273	0.5464	0.4232
MAPE (%)	12.44	9.26	7.11	5.91
TIC	0.3471	0.3248	0.2784	0.2352