Prediction of Fractal Dimension in Shale CT and its Robustness to Interference Based on Convolutional Neural Networks

Abstract

[Purposes] The development of shale oil and gas often requires a thorough understanding of the internal pore-fracture distribution patterns within shale reservoirs to optimize development strategies and enhance production capacity. In this context, the fractal dimension holds significant importance for reflecting the distribution patterns of pores and fractures within shale formations. [Methods] In this study, a convolutional neural network-based method for predicting the fractal dimension of shale Computed Tomography (CT) images is proposed. An independent convolutional neural network model is constructed, specifically designed for oil shale CT images. CT slices of oil shale samples treated with different temperatures, along with their corresponding fractal dimensions, are employed as the dataset and labels. The constructed convolutional neural network is trained and utilized for prediction to realize, effectively extracting fractal dimensions from shale CT images. [Findings] The trained model is applied to various practical scenarios and compared with the box-counting method. The results demonstrate a high degree of similarity between the predicted fractal dimensions of shale CT images by using the convolutional neural network and those computed through the box-counting method, with a difference of approximately 0.01. Additionally, the convolutional neural network method exhibits robustness against interference while also significantly accelerating the computation process compared with the box-counting method. Therefore, it can be concluded that the proposed method effectively captures the structural characteristics of images, allowing for reliable estimation of image fractal dimensions with notable resilience to noise and artifacts.

Keywords

• shale
• fractal dimension
• machine learning
• convolutional neural network
• shale ct