Multitransform of Seismic Attributes to Petrophysical Properties Using Committee Fuzzy Inference System

This paper presents an intelligent model based on fuzzy systems for making a quantitative formulation between seismic attributes and petrophysical data. The methodology consists of two main steps. In the first step, petrophysical data including water saturation (Sw) and porosity are predicted from seismic attributes using fuzzy inference systems (FIS) including the Sugeno (SFIS), Mamdani (MFIS) and Larsen (LFIS) fuzzy inference systems. In the second step, a committee fuzzy inference system (CFIS) is constructed using a hybrid Genetic Algorithms-Pattern Search (GA-PS) technique. The inputs of the CFIS model are the outputs and average of the fuzzy inference systems. Each of them has a weighting factor showing its contribution to the overall prediction. For this paper, 3D seismic data and petrophysical data from 11 wells of the Iranian Offshore Oilfield in Persian Gulf Basin are used. The performance of the CFIS model is compared to that of a probabilistic neural network (PNN). The results show that the CFIS method performs better than a neural network, the best individual fuzzy model and a simple averaging method.