M.R Akbari - Faculty of Petroleum Engineering, Amirkabir University of Technology, Tehran, Iran
M.J. Ameri - Faculty of Petroleum Engineering, Amirkabir University of Technology, Tehran, Iran
M.H. Ghazanfari - School of Chemical and Petroleum Engineering, Sharif University of Technology, Tehran, Iran
Y. Motamedi - Faculty of Petroleum Engineering, Amirkabir University of Technology, Tehran, Iran

Acid fracturing is a classical treatment used in carbonate formations to improve well productivity. Acid fracture conductivity is an important parameter for designing a fracture job. A model of acid fracturing conductivity must accurately anticipate fracture conductivity versus closure stress. The fracture conductivity is substantially influenced by rock type. A serious challenge of recent studies has been to predict behavior of different formations under various closure stresses. In this study an artificial neural network model was developed to precisely predict fracture conductivity by incorporating experimental data from various formations, whereby resulting in a good match between model predictions and experimental data. The effects of rock type was investigated on fracture conductivity, and show that different formations have different responses under various closure stresses. There is an optimum point at which maximum fracture conductivity is achieved, but finding this point is difficult because it is distinct for different formations.

Acid fracture conductivity, Artificial neural networks, Lithology, Rock type