Sediment transport modeling in open channels using neuro-fuzzy and gene expression programming techniques.

Area of Science:

• Civil Engineering
• Hydraulic Engineering
• Sediment Transport

Background:

• Sediment deposition poses significant economic and technical challenges in rigid boundary channels like sewers and irrigation systems.
• Self-cleansing criteria are used to prevent continuous sediment deposition, with modern designs incorporating more hydraulic parameters beyond single velocity values.

Purpose of the Study:

• To implement and compare neuro-fuzzy (NF) and gene expression programming (GEP) techniques for estimating the particle Froude number (Frp).
• To assess the non-deposition condition of sediment transport in rigid boundary channels using advanced computational methods.

Main Methods:

• Development of NF and GEP models using laboratory experimental data covering diverse sediment and pipe sizes.
• Performance evaluation of developed models against empirical equations using root mean square error (RMSE) and concordance coefficient (CC).
• Utilizing Taylor diagrams to validate the correlation between measured and predicted values.

Main Results:

• The precise NF model (NF4) demonstrated superior performance compared to the best GEP model (GEP1) and existing regression equations.
• NF models showed better accuracy in estimating the particle Froude number (Frp) for non-deposition conditions.

Conclusions:

• The neuro-fuzzy (NF) method exhibits high accuracy and applicability for estimating the particle Froude number (Frp) in sediment transport studies.
• NF models offer a reliable approach for designing self-cleansing rigid boundary channels, mitigating sediment deposition issues.