Abstract
Improving cluster efficiency is critical for economic and efficient multi-cluster per stage fracturing in unconventional shale & tight horizontal well completion. This paper highlights the findings from a field trial to test different perforation design variables which contribute to cluster efficiency. The goal was to optimize perforation design parameters and improve cluster efficiency for a given stage, and thus the well in its entirety.
A two well trial was conducted across the same bench formation on a single pad in Midland Basin. In all, eight perforation designs were created using two set points (high and low) across three key perforation design variables: 1) perforation phasing & orientation, 2) perforation diameter, and 3) perforation friction. Each design was repeated eight times (i.e. eight stages) to allow for a meaningful number of data points. After stimulation operations were conducted an acoustic imaging technology was utilized to assess the perforation dimensions for all perforations post-fracture for all stages as well as various sets of pre-fracture perforations.
In total, the trial was conducted across 64 stages (8 perforation designs × 8 stages per perforation design) using a Design of Experiments (DoE) method to assign low or high set points for each perforation design to best ascertain the impact of each test variable on the response variable as well as test for multicollinearity across the test variables. The uniformity index metric was used as a proxy for cluster efficiency and was calculated using two methods (a) eroded perforation area increase, and (b) post frac perforation area. Based upon the results obtained from the acoustic imaging data set and the subsequent data analysis, the uniformity index improved with a perforation design that had higher average perforation friction, smaller perforation hole shot size and a 0 degree in-line perforation orientation.
The field trial results of uniformity index provided high quality statistical quantification of optimum perforation design parameters and its impact on cluster efficiency.