基于Neuman理论的神经网络优化确定水文地质参数

Optimal Determination of Hydrogeological Parameters Using Artificial Neural Networks Based on Neuman Theory

针对利用非稳定流抽水试验资料确定潜水含水层参数传统方法的不足,系统分析考虑垂直分量和弹性释水的Neuman潜水井流模型解析解的基础上,利用实码加速遗传算法(RAGA)和自适应BP神经网络模型相结合对Neuman潜水井流模型解析解进行优化求解,提出确定潜水含水层水文地质参数的Neuman-BP法。以计算实例表明,Neuman-BP法不需分抽水时间———降深过程的前、后段分别进行参数确定,避免了前、后段所求导水系数T的不一致,既充分利用了抽水试验数据,又获得了较高精度的参数,简化了参数确定过程。

The traditional method for identifying the unconfined aquifer parameters has many disadvantages. Therefore, based on analyzing the analytical solution of the Neuman model that considers the effects of elastic storage and anisotropy of aquifers on drawdown behaviors, BP model of artificial neural networks have been built by samples which are produced by using Real number coded Accelerating Genetic Algorithm （RAGA） and the random number. BP model with the Gauss Integral to identify the unconfined aquifer parameters are combined and a method of Neuman-BP is also proposed. By application, this method can automatically identify the aquifer parameters, and the obtained parameters also have good accuracy.