Short-term Prediction of Flow Rate and Suspended Sediment Transport in a Tidal River Using Genetic Programming

Predicting flow rate and sediment transport are essential for water transportation, water resources management and drainage systems in lowland areas. This paper’s aim is to develop simple models for predicting flow rate and suspended sediment transport in the lower Chao Phraya River, which has complicated flow characteristics. The predicted flow rate equations were developed based on genetic programming (GP) and the relationships among flow rates, water levels and flow directions during spring and neap tide periods. The prediction results for the short term show accurate performance with correlation coefficient (r) values greater than 0.88 for the training and testing processes (with the flow rates of -3,000 m³/s to +3,000 m³/s). The flow rate and nearshore sediment transport rate using the GP model in spring tide show more accuracy than those in the neap tide. During spring tide, there is a change in the water level that had a great influence on both the flow rate and the flow velocity. The flow direction is one of the important parameters for the proposed GP model that can be used for practical engineering applications for predicting of flow rate and nearshore sediment transport rate for water management in the tidal river.