Special Oil & Gas Reservoirs ›› 2020, Vol. 27 ›› Issue (5): 118-124.DOI: 10.3969/j.issn.1006-6535.2020.05.018

• Reservoir Engineering • Previous Articles     Next Articles

Classification and Evaluation of Tight Oil Reservoirs Based on Fuzzy C-Means Clustering and Bayes Discrimination

Wang Wei, Kang Shengsong, Gao Feng, Guo Fenzhuan, Zhang Liang   

  1. Shaanxi Yanchang Petroleum (Group) Co., Ltd., Xi'an Shaanxi 710075,China
  • Received:2019-12-18 Revised:2020-07-08 Online:2020-10-25 Published:2022-02-18

Abstract: In order to solve the problem that the conventional tight oil reservoirs are difficult to classify and evaluate due to the lack of core analysis data, the optimal type of tight oil reservoirs was classified by fuzzy C-means algorithm through optimizing attribute parameters; then Bayes discriminant analysis was applied to establish the relationship between reservoir types and conventional well logging attributes, and the well logging attributes of conventional oil production wells were utilized to distinguish the reservoir types. Field applications show that the reservoir types of Chang7 II6 and Chang7 I2 perforated sand bodies in Zhidan area of the Ordos Basin are consistent with the oil test results. The coincidence rate of reservoir classification of 203 wells in the study area is 89.7%. This study has a certain guiding significance for the efficient development of tight oil reservoirs.

Key words: tight oil, reservoir classification, fuzzy C-means clustering, Bayes discrimination, reservoir evaluation, Ordos Basin

CLC Number: