Using Differential Evolution Algorithm to Optimize the Parameters of the Soil Characteristic Curve
OU Zhong-hui · Journal of Biomathematics · 2013
Differential Evolution is a simple and effective intelligent evolutionary algorithm in the continuous space global optimization problem.In this paper,we use different data sets in Unsaturated Soil Database and Differential Evolution algorithm to predict the soil characteristic curve parameters,obtain the parametric estimation for different environments(field,laboratory) and soil moisture(wet,dry) and effect of soil texture and sampling on parametric estimation and fitness.The results show that:parameteric estimation and characteristic values based on the dry soil data from the laboratory has high precision;The root mean square error of the clay soil is the smallest among four different-texture soils,its fitness about characteristic is better,the values of soil saturated water content and residual water content are higher.The parametric estimations of sand soil are otherwise;The size of sample has little effect on the fitting mean value of parametric estimation.