Using Two-Step Fuzzy Ranking and Artificial Neural Network for Reservoir Characterization

Baijie Wang
Supervisor: Zhangxing (John) Chen


With the development of oil industry, the reservoirs characterization is more and more important, which is a process for quantitatively assigning reservoir properties using all the available field data. Right now there are several methods to deal with it, such as core analysis, well testing and regression analysis and so on. In terms of regression analysis, there are several limitations: the mutual interactions between well logs and reservoir properties are very complex; regression method only considers limited variables in well log; ignoring other important attributes could lead the result inaccuracy. So we apply two kinds of artificial intelligent methods (Fuzzy ranking and artificial neural network) to improve the performance of the reservoir characterization.