-
oa Predicting Open-Hole Logs Using Artificial Neural Network Technique on Pulsed Neutron Data (On the Example of the Field in Dniepr-Donets Basin)
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 17th International Conference Monitoring of Geological Processes and Ecological Condition of the Environment, Nov 2023, Volume 2023, p.1 - 5
Abstract
Open-hole logs are crucial for formation evaluation and developing a field model. However, poor borehole conditions or financially restricted rig times can make open-hole logging impractical or impossible.
This paper presents a novel approach to predict open-hole log responses (neutron porosity and compressional slowness) using cased-hole pulsed neutron logging data on the example of one of the fields in the Dniepr-Donets basin (DDb). The emulation technique uses a feedforward, back-propagation-based artificial neural network (ANN) model and pulsed neutron data from four-detector CRE tool (Weatherford).
The presented data are from a study of two wells located in the same field for which open-hole and cased-hole logging were performed. Results demonstrate good match and reliable correlations between emulated and actual neutron porosity and compressional slowness curves (R2>0.7). It is shown that the proposed technique is a valid alternative to acquire emulated open-hole data when additional open-hole information is required after the well has been completed or wellbore conditions present risks for open-hole data acquisition. The presented ANN model can be used for the emulation of open-hole curves in other wells of the field or wells of nearby fields located in the central axial part of DDb.
