The effects of upscaling fracture properties in unconventional reservoirs



Journal Title

Journal ISSN

Volume Title



Reservoir simulation is a tool that allows us to predict the behavior of reservoirs for better management. In unconventional reservoirs, geological and fracture properties are essential for reliable simulation models. However, highly detailed simulation models can be computationally expensive and difficult to simulate. Hence, reservoirs are often simplified, upscaled and homogenized, losing important formation characteristics. Likewise, fracture parameters are often simplified to uniform fracture properties. Simulations with such generalizations can introduce uncertainties, errors, and unreliable results. The effects of simplified reservoir models might not be disruptive in conventional reservoirs where the porosity and permeability are high. However, while simulating tight and ultra-tight reservoirs with nano-Darcy permeability the upscaling effects can create disparity in production results. The goal of this effort was to examine the effects of these simplifications for modeling of tight reservoirs. This work utilized reservoir simulation and data analytics software to quantify the change in simulated reservoir behavior, subject to varying fracture and reservoir properties. The models were categorized by the following fracture/reservoir parameters: fracture length, spacing, fracture conductivity, permeability, and porosity. A base model with uniform fracture properties was created. Non-uniform models with varying degrees of upscaling were then created which maintained the average value for all parameters across every simulation model. The results show that upscaling has noticeable effects when normalizing fracture length or conductivity. Furthermore, when upscaling, fracture clusters can create substantial errors. Since the clusters affect the change in flow regimes, the production interference among fractures must be considered while upscaling. By understanding the effects of upscaling on pivotal parameters, one can create simulation models that best represent the actual well behavior while minimizing the computational cost.



Reservoir, Simulation, Upscaling, Unconventional, Hydraulic fracturing, Nano-Darcy