An Evolving Neural Network Using an Ant Colony Algorithm for a Permeability Estimation of the Reservoir

Abstract

An ant colony optimization algorithm (ACA) has the powerful ability to search for a globally optimal solution, and the back-propagation (BP) algorithm features rapid convergence on local optima. A proper hybrid of the two algorithms (ACA-BP) may accelerate the evolution of neural networks and improve their forecasting precision. An ACA-BP scheme adopts an ACA to search for the optimal combination of weights in the solution space and then uses a BP algorithm to obtain the accurate optimal solution quickly. The ACA-BP and BP algorithms were applied to predict the permeability of Mansuri Bangestan reservoir located in Ahwaz, Iran, utilizing available geophysical well log data. Experimental results showed that the proposed ACA-BP scheme was more efficient and effective than the BP algorithm.

Keywords:

• ANN
• ant colony algorithm
• back-propagation
• permeability
• reservoir