New Experimental Correlations to Predict Proppant Distribution Between Perforation Clusters Using Low Viscosity Fluids in a Horizontal Wellbore

Abstract

Abstract Hydraulic fracturing is a widely applied technology to increase recoverable reserves and accelerate production in low permeability reservoirs. In this technique, the hydraulic fractures are generated by pumping a fracturing fluid at a high flow rate and introducing proppant with the fluid to keep the fractures open after the pumping is halted. In horizontal well fracturing, one of the most important factors in this method is the proppant transport and distribution in the wellbore and among perforation clusters. Uneven proppant distribution between multiple clusters/fractures can cause a drop in the near-wellbore fracture conductivity, which in turn causes a negative impact on the production performance of the hydraulic fractures. This paper presents new experimental correlations that were developed and can be utilized to predict the proppant distribution between subsequent perforation clusters. The experimental correlations were developed using dimensional analysis from appropriate experimental data. The experimental data for this study was obtained from a horizontal wellbore apparatus with three perforation clusters at shot densities of 4 SPF with 90-degree phasing. Several experimental tests were conducted on 20/40 and 40/70 for white sand (SG of 2.65) and ultra-light weight ceramic (SG of 2.0) using freshwater fluids at three flow rates and over a wide range of proppant concentrations. Four types of experimental correlations were developed by experimentally investigating the effect of each parameter on the proppant distribution between the three perforation clusters. The first correlation type developed is based on the proppant concentration, while the second type is based on the flow rate and proppant concentration. The third correlation type includes the particle median diameter as another independent variable, along with the flow rate and proppant concentration. The fourth and final correlation developed consists of the proppant density along with the proppant median diameter, flow rate and proppant concentration as its independent variables. This fourth and final correlation was validated by comparing the predicted values to the laboratory values. The results of correlation analysis comparison to lab data show that the 20/40 and 40/70 ULW ceramic showed the lowest average error values at 2.92% and 1.56%, respectively, while the 20/40 and 40/70 white sand transport predictions were reasonable (average error range of 3.67% - 11.67%). The low error values indicate the high reliability of the developed correlations in predicting the proppant distribution between perforation clusters. To the authors’ knowledge, the developed correlations are the first of their kind to be based on experimental data. These correlations can help to determine the optimum flow rate that is required to attain even distribution of the proppant and provide more insight about the anticipated proppant distribution into and out of perforation clusters.