AN ARTIFICIAL NEURAL-NETWORK FOR THE PREDICTION OF IMMISCIBLE FLOOD PERFORMANCE

Despite several decades of artificial neural network reserach in other engineering disciplines, only recently work has been reported on its use as a prediction tool in petroleum engineering applications. Existi ng methods for the prediction of fluid now in porous medium include nu merical simulation techniques and laboratory core flood experiments. B oth of these methods are generally expensive and time consuming. Howev er, neural networks, once successfully trained, can be used to predict reservoir performance in a short time with a personal computer. An ar tificial neural network was developed using data obtained from fine-me sh numerical simulation to predict the breakthrough oil recovery of im miscible displacement of oil by water in a two-dimensional vertical cr oss section. The network is able to predict the results of the fine-me sh numerical simulations without actually performing these simulation runs. Various neural network connections were investigated using the b ack-propagation with momentum algorithm for error minimization. This p aper describes the design, development, and testing of the neural netw ork.