Application of artificial neural network in Geology: Porosity estimation and lithological facies classification

Based on the relationship between porosity (or lithological facies) and other petrophysical properties, Artificial neural networks (ANN) are respectively trained for porosity estimation and lithological facies classification, using core porosity (CPOR) data and core lithological facies interpretation results of part of core interval together with some well logs (petrophysical properties). After the ANN were constructed, they were used for porosity estimation and lithological facies identification in both trained and untrained core intervals, for further analysis of errors or accuracies of the estimated results and ANN. After careful analysis, the error of estimated porosity is from −0.3 to 0.3, and the accuracy of lithological facies identification is 0.7, both showing high reliability of ANN. The constructed ANN can be confidently applied for other un-cored wells or intervals.