Intelligent Classification Method for Tight Sandstone Reservoir Evaluation

Strategic Insights -

Investigators at Northeast Petroleum University in Daqing, People's Republic of China, have proposed a novel intelligent classification method to improve the accuracy of reservoir evaluations in tight sandstone reservoirs. This method, called GA-XGBoost, integrates genetic algorithm optimization with extreme gradient boosting to enhance classification accuracy in small-sample scenarios. The research team analyzed the lithological, physical, and lithofacies characteristics of tight sandstone reservoirs and selected key evaluation parameters, including mineral composition, porosity, and logging data. The experimental results demonstrated that GA-XGBoost achieved an 88.8% classification precision, outperforming traditional algorithms in both efficiency and accuracy.

Key Takeaways:

  • The researchers proposed an intelligent classification method, GA-XGBoost, which integrates genetic algorithm optimization with extreme gradient boosting to enhance classification accuracy in small-sample scenarios.
  • The lithological, physical, and lithofacies characteristics of tight sandstone reservoirs were analyzed, and key evaluation parameters were selected, including mineral composition, porosity, and logging data.
  • The experimental results demonstrated that GA-XGBoost achieved an 88.8% classification precision, outperforming traditional algorithms in both efficiency and accuracy.
  • The research advances the standardization of intelligent reservoir evaluations and provides a more reliable classification approach for tight sandstone reservoirs.
  • The study also contributes to the integration of geological exploration and computational intelligence, offering new insights into the application of machine learning in geosciences.
  • The research was financially supported by the National Natural Science Foundation of China (NSFC), Key Research and Development Project of Hainan Province, Heilongjiang Provincial Natural Science Foundation Sponsored Project, Basic Research Fund Project of Heilongjiang Provincial Education Department, and Daqing City Guiding Science and Technology Plan Project.

Statistics:

  • The GA-XGBoost method achieved an 88.8% classification precision, comparing to traditional algorithms.
  • The study analyzed the characteristics of tight sandstone reservoirs, including mineral composition, porosity, and logging data.
  • The research was conducted by a team of 11 authors from Northeast Petroleum University, including Zongbao Liu, Zihao Mu, Chunsheng Li, Tao Liu, Kejia Zhang, Yuchen Yang, Liyuan Liu, Jiacheng Huang, Shiqi Zhang, and Haiwei Mu.

Sources:

  • Liu, Z., Mu, Z., Li, C., et al. (2025). Intelligent Classification Method for Tight Sandstone Reservoir Evaluation Based On Optimized Genetic Algorithm and Extreme Gradient Boosting. Processes, 13(5), 1379.
  • NewsRx. Study Results from Northeast Petroleum University Update Understanding of Technology (Intelligent Classification Method for Tight Sandstone Reservoir Evaluation Based On Optimized Genetic Algorithm and Extreme Gradient Boosting). Life Science Weekly, June 17, 2025; p 4612.