بهینه سازی یک بعدی مدلسازی جریان ناپایدار برای کانال های باز
Abstract: The objective of this research is to implement an optimized numerical one-dimensional open-channel unsteady flow model, and to evaluate the effect of optimization on the performance of the model. An implicit Preissmann-type finite-difference numerical scheme is implemented to solve the de St. Venant flow equations. Besides geometrical, initial and boundary condition data, the model necessitates descriptors that approximate energy dissipation along the channel. The parameter most used in hydraulic engineering is the empirical Manning\'s n value, which was originally developed for uniform flow situations. In this study two approaches are implemented. First, the model uses a set of n values that describe friction roughness along the channel, but remain unchanged throughout the simulation. In the second approach, n values change with discharge according to a pre-specified formula of variation. In order to optimize the model, a genetic algorithm is developed. The objective of the algorithm is to find the global optimal solution among many possible ones within a specified range of parameter values. To evaluate the performance, a specific case is modeled. The target application is a river reach located in the St. Louis harbor area. The simulation predicts discharges and water stages that can be compared to measured values by means of a difference function, whose value is minimized by the optimization process. After testing the algorithm with different strategies, the numerical results are encouraging and show that an optimal solution is obtained when the unsteady variation of n is introduced.
Keywords: Applied sciences, Unsteady flow, Open channels