Explicit Simulation of Multiple Hydraulic Fractures in Horizontal Wells

Abstract

Abstract This paper presents explicit simulation of hydraulic fractures in horizontal wells to predict the fracture behaviour and post-fracture production profile leading to an optimum design and maximum production enhancement. The paper demonstrates the advantages of using explicit numerical simulation in contrast to analytical modeling. Conventionally, analytical methods and software are used to forecast post-fracture production rates to evaluate the profitability of fracturing. The availability of analytical software that is simple and fast has been the rationale for using analytical methods in the past. However, computer technology has enabled us to run numerical models with nearly the same speed. Although analytical methods have been continuously improving, there are a number of parameters and effects which are not fully taken into consideration by these methods and, therefore, result in unrealistic production forecasts. These factors include non-Darcy effects along the fracture, multiphase flow, condensate banking, flow convergence, and reservoir layering and geometry; they can be realistically simulated using explicit fracture modeling. The use of numerical modeling enables the user to utilize detailed reservoir properties and to simulate the flow from matrix to fracture as it occurs in the reservoir. Numerical simulation of fractures is even more essential for horizontal wells. In addition to removing skin and increasing the equivalent wellbore radius, fracturing in horizontal wells enhances the production by improving the vertical communication between the reservoir layers and by connecting all the layers to the wellbore. This can practically be modelled using explicit flow simulation. The above considerations are demonstrated in the simulation study conducted for multiple fracturing of a new horizontal well in the Dunbar Field, operated by TOTAL E&P UK PLC. The objectives of the study were to assess the feasibility of fracturing, to determine the optimum number of parallel fractures, and to forecast the production. Fracturing operations are very costly; precise assessments must therefore be carried out before the operation takes place. Introduction Hydraulic fracturing is a high-cost operation that, if successful, can significantly benefit the profitability of a new well. There are however high associated risks that are result of the large number of variables that must be considered in the planning and design process. In order to accurately forecast the post-production rates it is essential that they are calculated using a reliable methodology that adequately represents both the large range of influential reservoir parameters and their inherent uncertainty. The most common way of estimating post-fracture production has been to use analytical software such as nodal analysis software with fracture modeling capabilities or to use specialized hydraulic fracture analytical simulators. The advantage of this method is simplicity and speed. However, there are a certain aspects of the problem that are not considered when using these methods that may influence the results substantially. Production forecasting in horizontal wells suffers even more when using simplistic methods because the vertical communication caused by fractures is not properly modelled. In recent years, numerical simulation for hydraulic fracture modeling has been introduced to the industry. Numerical simulation allows detailed reservoir properties, layering, pressure-volume-temperature (PVT), and field and well geometry to be incorporated into the model. It is therefore possible to model the fluid flow from matrix to fracture and also flow along the fracture more realistically. However, complexities of building the model and numerical diversion have always been two problems in fracture modeling. To solve these problems, new numerical fracture modeling software or templates were developed. A template is a user-friendly interface between the user and the main simulator engine. Generating a fracture model using the templates is substantially faster. In addition, changing parameters and regenerating the model is straightforward, so that sensitivity analysis to various parameters can be carried out quickly. This paper demonstrates the differences between analytical and explicit techniques with a view to raising awareness and understanding of fracture modeling in the industry. The application of fracture modeling is demonstrated in the study carried out for TOTAL E&P UK PLC on the Dunbar Field.