Gene expression programming for computing energy dissipation over type-B piano key weir

Abstract

Computation of energy dissipation over weir (or spillway) structures is vital to solving numerous engineering problems, notably during floods. Therefore, an accurate estimation of the energy dissipation at the base of the spillway is essential. The spillways must be designed to spill large amounts of water at high efficiency with high structural performances without causing any damage to the structure or its surroundings. Energy dissipation over weir structures is difficult to predict using conventional formulas based on empirical methods. Consequently, new and accurate techniques are still highly demanded. In this study, Gene Expression Programs (GEP) were used to estimate the relative energy dissipation at the base of type-B Piano Key Weirs (PKWs) by taking three non-dimensional parameters into account: headwater ratio, magnification ratio, and the number of cycles. The performances of the GEP model were compared with empirical equations based on statistical factors coefficient of determination (R2), and root mean square error (RMSE). The computed values of the relative residual energy using the proposed models are within ±5% of the observed ones. Results indicate that the proposed GEP model predicted the relative residual energy satisfactorily with the coefficient of determination (R2 = 0.9979 for training, 0.9980 for testing) and root mean square error (RMSE) of 0.0099, 0.0092 for training and testing datasets, respectively.