Lithofacies prediction from well log data using a multilayer perceptron (MLP) and Kohonen's self-organizing map (SOM) – a case study from the Algerian Sahara
28 Jun 2013
1Algerian Petroleum Institute, IAP, Boumerdes, Algeria
2LABOPHT, FHC, UMBB, Boumerdes, Algeria
Abstract. In this paper, a combination of supervised and unsupervised leanings is used for lithofacies classification from well log data. The main idea consists of enhancing the multilayer perceptron (MLP) learning by the output of the self-organizing map (SOM) neural network. Application to real data of two wells located the Algerian Sahara clearly shows that the lithofacies model built by the neural combination is able to give better results than a self-organizing map.