Genetic Algorithm Optimized Neural Network Prediction of the Friction Factor in a Mobile Bed Channel

摘要

The problem of predicting flow resistance in alluvial channels with sufficient accuracy is of great interest to hydraulic engineers. The difficulty arises because the flow boundary in alluvial channels is not fixed but continually undergoes changes in its characteristic geometry and dimensions through mutual interaction between the flow and bed. As the process is extremely complex, getting deterministic or analytical form of process phenomena is too difficult. Data mining, which is particularly useful in modeling processes about which adequate knowledge of the physics is limited, is presented here as a complimentary tool to predict the complex non-linear relationship between the friction factor of an alluvial channel and its influencing factors. In this paper, genetic algorithm optimized back propagation neural network has been proposed to model the friction factor. Contributions of different parameters have been ranked by analyzing the weights of neural network.