Comparison of two different application of neural network on Sudan soil profile

摘要

The objective of this paper is to compare between two studies that had been carried out using ANNs in prediction of Sudan soil profile. This importance in the design and implementation of all engineering projects which reduce cost and time. Artificial Neural Networks (ANNs) program is applied to realize this aim. The data of 1909 boreholes from 417 sites was used firstly for a single model and then divided to five zones to be used for localized models. The input data is the coordinate and depth and the output data is the soil classification and soil parameters. The result showed that ANNs can be used as a good decision support and source of information for soils profiles. It is more efficient tools to be used for small zones than all area.