The Application Of Improved Genetic Algorithms On The Optimal Design Of Tree Networks In Rural Drinking Water Project

The optimal design of tree networks is an important task of water supply engineering. It has been broadly and deeply studied by many domestic and foreign scholars with many representative optimal theories. The design methods have been advanced such as enumerating method, limits of flow method, linear programming method, nonlinear programming method, dynamic programming method, generalized reduction gradient (GRG) method, simulated annealing(SA) and genetic algorithms(GA) and so on.Firstly, this paper expatiates the content and sense about optimal design of tree networks, briefly introduces some optimal methods which have been advanced, analyzes these methods and points out their limitation. Based on improved genetic algorithms, this paper takes yearly expenditure converting value as target function and then sets up the genetic algorithms(GA) model on optimal design of the backbone of tree networks in rural drinking water project by means of taking some effective measures on selection operator, crossover operator, mutation operator and some parameters setting, after which it develops the module of design system of rural drinking water project planning. The module is operated easily with a friendly interface. And the running speed is fast with reasonable results, which meets the command in advanced and has a preferable practicability.Finally, the paper gives examples of engineering design optimization.