Abstract
Fracrure density distribution along four bore-holes drilled by JNC, Tono Geoscience Center were analyzed by using ANN (Artificial Neural Network) and information entropy. The objective of this study is the development of technique that enables to predict the density in the deeper section of bore-hole. The obtained results are as follows.
(1) Fracture density in deeper part can be predicted by ANN model.
(2) The accuracy of prediction could be increased by using “entropy values” in ANN model.
(3) It is also found that geological structures, rock type and etc. must be well described before apply the tectonics because the stochastic property of fractures density in the function of geological condition.
