Borehole Log Reconstruction using ANN: A Case Study from KBH-08, Koyna-Warna Region

Abstract

Well logs provide valuable information about the sub-surface geological formations. Several complications in data recording, like, instrument error, borehole conditions, incomplete logging, etc., often result in data loss which in turn causes difficulties in the interpretation of recorded logs. The present work emphasises on rebuilding the missing logs using artificial neural networks and the technique is applied to reconstruct the missing sonic log from the KBH-08 borehole in the Koyna-Warna region. The generated sonic log shows the fracture zones in the borehole that correlate well with the zones identified from other logs.