利用遗传算法对土壤模型的反演计算

Inversion Calculation of Soil Model by a Genetic Algorithm

为确定土壤分层结构模型,利用Wenner法测量了视在电阻率和极间距之间的关系,并引入了非线性目标函数将土壤参数求解问题转化为待求目标函数的极值问题。但由于传统的线性化或拟线性化方法求取目标函数存在易陷入局部最优的问题,提出了运用遗传算法优化计算分析土壤分层结构模型参数,用该方法可计算任意分层的土壤模型。计算结果表明,该法切实可行。

How to determine soil stratification structure model lays the foundation of designing the grounding system and simulation calculations of substation and power plants grounded network. The relationship between apparent resistivity and probe distance was used in a Wenner measurement method, through the introduction of nonlinear objective function, the solution of soil parameters will be transformed as an extremum question of objective function. Because the traditional linear approach or simulation linear approach has shortcomings such as getting in a partial optimization when ask for objective function. The authors introduce genetic algorithms to deal with the parameters optimization for multilayer structure models of soil, the methodology can calculate the arbitrary stratified soil models. Calculations show that the method is practical.