Pinpointing the key parameters that control the errors in estimating the total-load sediment flux using sand particles measured data in reservoir engineering

Sediment deposition is the principal problem affecting the useful life of reservoirs. Knowledge of both the rate and pattern of sediment deposition in a reservoir is required to predict the types of service impairments which will occur, the time frame in which they will occur, and the types of remedial strategies which may be practicable. The present analysis details the importance of physical properties to the total-load sediment fluxes using five equations, measured sand particles and suggests which properties have more control on the final result by providing insight into the relative strengths, weaknesses and limitations. The authors concentrated on available field and flume data sets gathered from different reliable sources, rather than focusing entirely on total-load equations. Artificial neural network (ANNs) method was used to validate this study. Several graphs and statistical analysis were presented to emphasize the influencing effect of those parameters that were detected by ANNs and are directly controlling the error in the total-load sediment flux using measured sand particle data sets.