鉴于本构模型和土体参数确定上的困难,在有效应力原理和随机介质理论的基础上,建立采水区地面沉降时空预测的BP神经网络模型,所建模型具有分布参数模型的特征。运用所建模型对其他地面沉降监测点进行了计算和预测,研究表明,所建立的BP网络模型能准确地反映采水地面沉降的时空规律。
Because of the difficulties from the constitutive model and the parameters of soils, based on the effective stress principle and the stochastic medium theory, the BP network model for forecasting the temporal and spatial distribution of land subsidence caused by groundwater exploitation is established, which possesses the characteristics of distributed parameter model. Through using this model, the land subsidence of other points is simulated and predicted. The study shows that BP network model could nicely reflect the time and space characteristics of land subsidence caused by groundwater exploitation.