基于混合遗传算法估计van Genuchten方程参数

Parametric estimation of the van Genuchten's equation based on hybrid genetic algorithm

van Genuchten(VG)方程是最常用的土壤水分特征曲线方程,其参数取值的精度直接影响到土壤水分运动方程计算的精度。该文建立了VG方程参数的优化模型,构建遗传算法与Levenberg-Marquardt算法相结合的混合遗传算法对其进行求解,并进行了数值试验。结果表明采用混合遗传算法比普通遗传算法不但提高了收敛效率,而且收敛迭代次数也大大减少;采用混合遗传算法估计参数的精度比非线性单纯形法和阻尼最小二乘法要高一些,而且不需要给参数初值。因此,混合遗传算法可以作为估计VG方程参数的一种新方法。

The van Genuchten equation is the most commonly used soil characteristic curve equation, and its parameter value precision directly affects the soil water movement equation computation precision. In this paper, the parameter optimized model for van Genuchten equation was established, and the hybrid genetic algorithm which combines genetic algorithm with I_even- berg-Marquardt optimization algorithm was used to solve it. A series of numerical experiments were conducted to verify the model. The results show that the hybrid genetic algorithm has the higher convergence efficiency than the ordinary genetic algo- rithm, moreover the number of convergence iteration also reduces greatly. The hybrid genetic algorithm has the higher estimate parameter precision than the nonlinear simplex method and the damped least squares method, moreover, do not need the initial parameter value. Therefore, the hybrid genetic algorithm can be can be used to estimate van Genuchten equation parameters as a new method.