Application of artificial neural network on prediction reservoir sensitivity

People try to evaluate reservoir sensitivity and diagnose formation damage by performing experiments. However, it needs too much time, it isn't accurate enough either. And its replacement by computer is necessary. In this paper, the application of artificial neural network to predict reservoir sensitivity is studied and corresponding models are constructed: back-propagation neural network and adaptive resonance theory neural network. The former is used to evaluate reservoir sensitivity and the latter to diagnose formation damage. During the application process of artificial neural network to predict reservoir sensitivity, the original data are converted to the data needed in decision-making and the experience of specialists is used in diagnosis and decision-making. This minimizes the influence of uncertain factors on the problem and enables the model to be advanced, predominant and adaptive. The corresponding software based on the research of application of artificial neural network is programmed and the verification of the models constructed shows satisfactory reliability.