Scholarlink

<p>Spillways provide an environmentally friendly and sustainable solution for erosion control, energy dissipation and enhancing downstream riverbank protection. The present study performed physical modeling and soft computing analysis to evaluate energy dissipation through various gabion spillways. Nine physical models of Gabion stepped spillway with a triangular gabion sill placed over steps were investigated for wide range of discharge with downstream slopes of 1:1.5,1:1 and 1:0.5 (V: H) and compared with the standard gabion and rigid stepped spillway. Physical findings reveal that the gabion stepped spillway comprising sills improves energy dissipation compared to standard gabion and rigid stepped spillways. Findings from Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Extreme Learning Machine (ELM) models reveal that the ELM model outperformed both the ANN and ANFIS models, achieving lower root mean square error (RMSE) values of 0.0028 (training) and 0.0035 (testing) as well as higher correlation coefficient (R²) values of 0.9844 (training) and 0.9678 (testing).</p>

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

Experimental and Soft Computing Analysis of Hydraulic Performance in Gabion Stepped Spillways: A Comparative Study

  • 期刊论文

  • 发表日期：

    2026年5月25日

查看全文
  • Aniket Kumar Sharma,
  • Bharat Jhamnani

摘要

Spillways provide an environmentally friendly and sustainable solution for erosion control, energy dissipation and enhancing downstream riverbank protection. The present study performed physical modeling and soft computing analysis to evaluate energy dissipation through various gabion spillways. Nine physical models of Gabion stepped spillway with a triangular gabion sill placed over steps were investigated for wide range of discharge with downstream slopes of 1:1.5,1:1 and 1:0.5 (V: H) and compared with the standard gabion and rigid stepped spillway. Physical findings reveal that the gabion stepped spillway comprising sills improves energy dissipation compared to standard gabion and rigid stepped spillways. Findings from Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Extreme Learning Machine (ELM) models reveal that the ELM model outperformed both the ANN and ANFIS models, achieving lower root mean square error (RMSE) values of 0.0028 (training) and 0.0035 (testing) as well as higher correlation coefficient (R²) values of 0.9844 (training) and 0.9678 (testing).