Discussion on artificial neural network technology used to predict the underground water content

摘要

In this paper, we made Ximazhuang water resources in Shijiazhuang city as proving ground and a number of local agro-wells known as the starting point,and made use of induced polarization method and resistivity sounding and other surface geophysical methods, based on the parameters such as apparent resistivity, polarization rate, half-bad and the decay rate, etc. to the relevance of water inflow, adopted artificial neural network techniques to build predictive models of underground water content, by the rules of mean-variance test shows that using surface geophysical methods to establish the model of the exploration area for quantitative prediction of water content is feasible, and it can save the cost of digging wells greatly, in this further we will put forward and establish the new concepts, new models to predict the ground aquifer water content with the very good promotional ,which based on artificial neural network techniques and integrated geophysical methods.