data-driven technique

Discharges in hydrometric stations are estimated by converting the stage values to the discharge using a stage-discharge relationship or by multiplying mean velocity with flow cross-sectional area. Estimation of mean velocity in hydrometric stations, especially during flood events, is not easily possible. Therefore, the method of estimating mean velocity by converting the maximum velocity to mean velocity using a conversion factor is a desirable method. The velocity convert factor estimation in stations without enough valuable measured discharge data is a challenging issue. Present study develops a method for determining the conversion factor by combining the entropy velocity profile and a data-driven technique (genetic programming) by knowing only one mean velocity value, and thus develops a method to determine discharge at the weak gauging sites. The advantage of the method introduced in this study is the simplicity of application and the use of parameters that can be easily measured to estimate the mean velocity values. The performance of the method was evaluated by comparing the computed and the observed mean velocity values and the Root Mean Square Error and the Mean Absolute Error were found to be 0.05 m s-1 and 0.04 m s-1, respectively. The results showed that the introduced method estimates a suitable conversion factor compared to similar methods and is applicable for stations without measurement.