A hybrid method for solving automated production history match of reservoirs with strong heterogeneity

Key words:  strong heterogeneity reservoir  principal component analysis method  discrete cosine transform method  simultaneous perturbation stochastic approximation

Abstract:

Many sandstone oil and gas reservoirs are features with large reservoir span and strong heterogeneity, and long-term water flooding can make their geological condition more complicated. In this study, a hybrid method of automatic history matching is proposed for strong heterogeneity reservoirs. The new method consists of two parts. The first part includes a combination of principal component analysis (PCA) and a discrete cosine transform (DCT) method, in which the features of the prior models can be extracted with reduced data dimension and preserving the comprehensive geological characters and improving the robustness of history matching. In the second part, an artificial bee colony algorithm (ABC) is combined with a stochastic perturbation and stochastic approximation algorithm (SPSA), and the improved algorithm is applied to improve the accuracy and the speed of the solution. In a case study, the proposed method is applied to a strong non-homogeneous reservoir model. The results show that The PCA combined with DCT method can extract the main geological features of strong heterogeneous reservoir, which is close to the real geological model. The method can provide theoretical support for the study and identification of complex heterogeneous oil and gas reservoirs.