Table 6 Comparison of data preprocessing performance
From: Machine learning methods for predicting residual strength in corroded oil and gas steel pipes
Number | Source | Year | Model | Performance |
|---|---|---|---|---|
①| 2023 | BRANN | R2 = 0.9962; MSE = 0.4743 | |
② | 2023 | KPCA-BRANN | R2 = 0.9966; MSE = 0.6365 | |
③ | 2023 | LLE-BRANN | R2 = 0.9993; MSE = 0.0886 | |
④ | 2022 | Neural network | R2 = 0.9551; MAPE = 7.070%; RMSE = 1.6060 | |
⑤ | 2022 | PCA + Neural network | R2 = 0.9843; MAPE = 4.596%; RMSE = 0.9510 | |
⑥ | 2021 | ANFIS without PCA | R2 = 0.9516; MAE = 1.7423; RMSE = 2.3513 | |
⑦ | 2021 | ANFIS-PCA | R2 = 0.9919; MAE = 0.6917; RMSE = 0.9883 |
