Evaluation of Multi-Cluster Fracturing Placement with Multi-Arm Caliper

With the increasing application of horizontal well drilling, multistage hydraulic fracturing has become widespread in unconventional resources. The production data often reveal that the production contribution of each cluster is highly non-uniform, indicating a corresponding inconsistency in fluid and proppant distribution. Understanding the distribution is crucial to completion designs aiming for a more even cluster contribution. In our study, we developed a new approach for evaluating the distribution of fracturing materials using multi-arm caliper data. Most perforations are performed perpendicular to the casing, and perforation holes are eroded as fracturing fluid flows through them. Because casing fluid flow is unidirectional, the erosion is asymmetric. Modern multi-arm caliper tools exhibit enough resolution to accurately resolve casing inside-diameter profiles. Our systematic workflow, perforation asymmetry analysis (PAA), assesses erosion and correlates it with material distribution. First, the peak in the inside-diameter profile representing a perforation hole is identified. Subsequent analysis of the profile’s asymmetry indicates the degree of erosion. Based on the analysis of individual perforation holes, the material entry for a cluster and the material distribution among clusters are modeled. Well C-2H, drilled in the Beetaloo sub-basin, Northern Territory, Australia, was stimulated with 21 hydraulic fracturing stages using the plug-and-perf technique. After well cleanup, a 24-arm caliper was run in the hole with the objective of understanding casing ovality after stimulation. The analysis encompassed different combinations of cluster numbers, perforation holes, treating fluids, and pumping rates. The results helped to iterate the completion and treatment design for an upcoming well. This new method can evaluate perforation hole erosion due to fracturing materials, which can be subsequently translated into insights about fracturing placement distribution. Compared with other measurements, this is an indirect method but exhibits good data availability and cost-effectiveness. Initially, it can be used and evaluated together with other solutions as a step toward subsequent precise standalone analysis.