Full text loading...
-
Applying Local Linear Neuro-fuzzy Model to Predict a Well Log from Other Well Logs
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 72nd EAGE Conference and Exhibition incorporating SPE EUROPEC 2010, Jun 2010, cp-161-00691
- ISBN: 978-90-73781-86-3
Abstract
We applied neuro-fuzzy model to predict a well log from other Well Logs. We used local linear neuro-fuzzy model tree (LOLIMOT) algorithm to train our model. This model uses well logs in a well as training data set. Trained neuro-fuzzy model is used to predict a well log or core measurement from relevant well logs. We used this method in a case study in an oil field in Persian Gulf. We predict the sonic log from some of the other well logs. Neuro-fuzzy model, Comparing with traditional neural network such as RBF, PNN and MLFN, shows better performance in predicting sonic log from other logs.