ABSTRACT

This paper present with the development of realistic models for predicting porosity by applying machine learning concepts on petro-physical logs. This paper is motivated due to increased exploration of unconventional hydrocarbon resources. Hence development of realistic models will reduce the exploration costs. Oil well data is modeled using modified ANFIS, consisting of optimized membership functions and fine tuned FIS model. The modified ANFIS model was constructed and tested on data samples recorded from niger delta basin. The average root mean square deviation is calculated. The results reported in this paper indicate that proposed oil well neutron porosity model can lead to the construction of more reliable static reservoir models for oil well simulation frameworks.