Fracture aperture is the key parameter to evaluate the quality and productivity of tight clastic rock reservoir. The well logs are always used to predict the fracture aperture, but the logging response characteristics of tight reservoir are complex and irregular due to the heterogeneity of macroscopic spatial structure and microscopic structure. Therefore, it is difficult to accurately predict the fracture aperture using the conventional logging interpretation method or single machine learning model. To solve this problem, we proposed a fracture aperture prediction method based on improved committee machine (CM) model. The CM model is improved by using analytic hierarchy process and joint neural network model, and based on the prediction performance of each expert network, the hierarchical expert committee machine (HECM) model is formed by adding the hierarchical network module adaptively. The practical application shows that this method can fully excavate the fracture aperture information contained in the logging data and effectively quantitatively characterize the fracture aperture. It provides a reliable geophysical data information for the comprehensive evaluation of tight clastic rock reservoir.

This content is only available via PDF.
You can access this article if you purchase or spend a download.