PERMEABILITY DETERMINATION USING NEURAL NETWORKS IN THE RAVVA FIELD, OFFSHORE INDIA

This paper describes the use of the backpropagation neural network (BP NN) technique to predict reservoir permeability using conventional wel l log data. The technique is demonstrated with an application to the R avva oil and gas field, offshore India. The Ravva field reservoirs are middle Miocene age nearshore marine sandstones that are often laminat ed to thinly interbedded shale. The use of conventional permeability-p orosity crossplots to predict permeability in this field was not succe ssful. The BPNN permeability prediction model (''RAVVANET'') was devel oped from a data set consisting of core permeability and well log data from two early development wells. The model was blind tested with dat a from a third well, which was withheld from the modeling process. The results of this study show that BPNN model permeability predictions a re consistent with core analysis results.