Scholarlink

The application of machine learning is now recognized as an essential tool in several aspects of water resource engineering. Consequently, in the current study, SVM_Puk and M5Rules models are utilized to predict how much energy would be dissipated by a stepped spillway under skimming flow conditions. Different input parameters such as critical flow depth (yc), slope (θ), width (w), height of the spillway (Hspl), step height (h), and number of steps (N) are taken for prediction of energy dissipation. Overall, 218 experimental datasets were collected from the literature, with 75% of those datasets being used for training and 25% being used for testing the model. To get the optimum result, several combinations of input parameters are also assessed. This study revealed that both models performed extremely well in predicting the energy dissipation of stepped spillways. However, as per statistical analysis, the SVM_Puk model has a superior performance in predicting energy dissipation of stepped spillway. The predicted values of SVM_Puk model show a correlation coefficient (CC) of 0.98, root-mean-square-error (RMSE) of 0.0445, and mean-absolute0error (MAE) of 0.0322 during testing.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Energy Dissipation in Skimming Flow of Stepped Spillway by Using Machine Learning Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ritusnata Mishra,
  • C. S. P. Ojha

摘要

The application of machine learning is now recognized as an essential tool in several aspects of water resource engineering. Consequently, in the current study, SVM_Puk and M5Rules models are utilized to predict how much energy would be dissipated by a stepped spillway under skimming flow conditions. Different input parameters such as critical flow depth (yc), slope (θ), width (w), height of the spillway (Hspl), step height (h), and number of steps (N) are taken for prediction of energy dissipation. Overall, 218 experimental datasets were collected from the literature, with 75% of those datasets being used for training and 25% being used for testing the model. To get the optimum result, several combinations of input parameters are also assessed. This study revealed that both models performed extremely well in predicting the energy dissipation of stepped spillways. However, as per statistical analysis, the SVM_Puk model has a superior performance in predicting energy dissipation of stepped spillway. The predicted values of SVM_Puk model show a correlation coefficient (CC) of 0.98, root-mean-square-error (RMSE) of 0.0445, and mean-absolute0error (MAE) of 0.0322 during testing.