Reservoir characterization and effective exploitation of low permeability reservoirs have always been cutting-edge topics in the world. Carbonate reservoirs in the Middle East are mainly porous carbonate reservoirs with low permeability, which are very different from fractured-cavity carbonate reservoirs. Such reservoirs are complex and diverse, differing markedly in geologic features and development modes. Therefore, it is necessary to conduct in-depth research on reservoir characterization and classification, so as to provide technical support for effective scale development of different types of low permeability carbonate reservoirs in the Middle East.
This paper presents a novel reservoir characterization and classification method based on principal component analysis (PCA) biplot and K-means clustering. By combining qualitative analysis with multivariate statistical analysis, eight characteristic parameters reflecting petrology, reservoir structure, reservoir physical property, interlayer, fluid property and heterogeneity were selected from 18 typical low permeability carbonate reservoirs in the Middle East, and the data matrix was constructed. The covariance matrix was completed with the method of singular value decomposition (SVD), and the PCA biplot was built up. The classified evaluation was carried out in combination with the K-means clustering method.
Reservoir samples, characteristic parameters and the interaction patterns between them were interpreted in the biplot as a graphical display. Results indicate that microscopic pore structure, reservoir thickness and heterogeneity were of great significance to the reservoir classification. Low permeability carbonate reservoirs in the Middle East could be divided into three types: thin reservoirs, thick layered reservoirs, and giant thick massive reservoirs. The irregular distribution of thief zone in thin reservoirs led to serious water out performance of horizontal wells. The vertical variation of reservoir physical property resulted in low sweep efficiency in thick layered reservoirs. The giant thick massive reservoirs were characterized by the coexistence of multiple pore systems, subtle barriers and baffles, and strong heterogeneity, which posed great challenges to the commingled water injection.
The traditional methods for classification of low permeability reservoirs simply based on permeability or fluid mobility convey little information on the internal attributes of reservoirs, thus lacking of adaptability to specific low permeability reservoirs. This methodology provides a graphical representation of interaction patterns which allows the response of each characteristic parameter in each reservoir to be directly interpreted. Regularities in the same type of reservoirs and the essential differences among various reservoirs are quickly identified by the method, which lays a theoretical foundation for the establishment of development modes of diversified low permeability carbonate reservoirs in the Middle East.