A Computational Algorithm for Calculating Fracture Index of Core Runs

Abstract

<p>Fracture Index (FI), which represents the count of fractures over an arbitrary length of core with similar intensity of fracturing, provides insight into the fracture state of rock masses. Manual interpretation of core is not only time-consuming but also prone to human error. To address these challenges, this study develops a computational algorithm that automatically calculates the FI of core runs based on fracture spacing data. The variance of fracture spacing values is selected as the key evaluating indicator. The algorithm automatically groups fractures appearing in core imagery based on uniformity of fracture spacing. Upon counting the number of fractures in each group over a certain length of core run obtained from the optimal grouping scheme, the algorithm outputs the FIs automatically. The algorithm’s performance has been effectively demonstrated on synthetic data and three real-life core datasets. Algorithm results show good agreement with those produced by logging geologists and listed in standard human-generated reports. The occasional discrepancies are attributed to the inherent subjective nature associated with the manual logging.<br></p>