Field Unit Enhances Fracture Investigation

Abstract

Summary A mobile research system that helps scientists investigate fracture growth in the field comprises data-acquisition and main-computer trailers and a production test unit. This paper describes equipment, procedures, field results, and laboratory procedures and results, and recommends a focus for future work. Introduction The Gas Research Inst. (GRI) sponsors a variety of research projects aimed at providing a stable supply of natural gas. An important part of the overall effort is stimulation of gas production from lowpermeability reservoirs. GRI has funded the development of a Mobile Testing and Control (T and C) Facility to measure, to monitor, and to record data during fracture treatments. These data, resident on a mainframe computer, help calculate the dimensions of the hydraulic fracture during the fracture treatment. Resource Engineering Systems Inc. has developed for GRI a 3D model that can be run in real time in the field. This model is used to predict both the shape and extent of the hydraulic fracture during a treatment. To define the shape of a hydraulic fracture properly, precise knowledge of the reservoir properties in the layer being fracture-treated plus the bounding layers above and below the pay interval is necessary. Important properties include the modulus of the layers and the in-situ stress value in each layer. To provide the detailed 3D evaluation of the formation, GRI funds research (such as the first Staged Field Experiment) to define the formation before a fracture treatment. We typically use a 3D hydraulic fracture model to history match actual fracture-treatment data. Data input consists of injection rates, rock properties, and fracture fluid properties. The 3D fracture propagation model uses these data to history match the actual pressures measured during the fracture treatment. By allowing the fracture shape and dimensions to vary with time, the model can he used to determine the most likely shape and extent of the hydraulic fracture. It is well understood that considerable research into the behavior of viscous crosslink polymer fluids at low shear rates is needed. In our research, we found that the analysis of actual field data becomes nonunique when both the viscous properties of the fluid and the fracture shape and extent can be altered by the reservoir modeler. In effect, the 3D reservoir model can predict several different types of fractures, depending on the viscous fluid properties used in the model. To improve the uniqueness of our computations, a method was needed to quantify the viscous behavior of fracture fluids better; therefore, the GRI rheology unit was built to obtain better information concerning the behavior of viscous, crosslinked, sand-laden slurries under simulated in-situ conditions. Every effort was made to simulate, as precisely as possible, time and shear conditions both on the surface and downhole as well as temperature profiles of the fluids tested. Obviously, we were not able to simulate pressure conditions actually seen in the downhole environment. It is generally accepted within the industry that pressure has a minimal effect on the rheological performance of these fracturing fluids. The data measured with this unit could then be entered directly into the 3D model. The result should be that the model has to vary only the fracture shape and extent during the history match.