Post Treatment Analysis of Real-Time Data

Theories of hydraulic fracturing and all fracture design models and programs assume the created fracture follows a planar path, that the fracture is in opening mode (simple tensile fracture), and that it consists of a single fracture extending on both sides of the borehole. In reality hydraulic fractures never satisfy these requirements. Because of anisotropy and inhomogeneiety of most reservoir rocks the hydraulic fracture is always mixed mode (mostly tensile, but also including some shearing and sliding mode fractures), and often with branches. Well completion details and treatment design and details of execution further complicate the fracture growth. Because of its irregular and random growth path and pattern such a fracture is called “Off-balance”. As recently discussed (Off-Balance Growth: A New Concept in Hydraulic Fracturing) an off-balance fracture does not occupy a single plane, its growing tip is not necessarily at its farthest point from the wellbore and moves randomly around the fracture, fracture has an irregular and rough surface, and where fracturing process includes tensile as well as significant shear fracturing and branching. Under these conditions the movement of the proppant inside the fracture is also not in a piston-like manner and it occupies the fracture in a random distribution dictated by the irregular growth of the fracture tip. In off-balance fracturing, the proppant nearest the wellbore may be the segment first pumped into it! In other words, last in is not necessarily the first out!

Off-balance fractures usually have narrower widths, shorter created and propped lengths, and often even less height than computed by simple fracturing models.

Off-balance growth has significant impact on production increase, proppant flowback, as well as borehole/casing stability during and after fracturing.

Details of well completion and job design contribute to its off-balance growth.

In the past the state-of-the-art technology was unable to use the actual treatment data and compute the geometry of the created fracture such that the actual and computed fracture behaviors exactly matched each other. Post job analysis has usually been limited to qualitative description of events during the job.

A unique and proprietary history matching technique has been developed by Daneshy Consultants International. This simulator, called SAGE-Frac (Simulation, Analysis, and Geometry Evaluation) computes the geometry of the created fracture such that there is a complete match between the actual and the simulated data at every point of the treatment. The new technology is based on the 3D mathematical formulation of the growth of a fracture that is obstructed from its normal growth. These obstructions can be created by multiple fractures, shear fractures, branches, or flow of proppant inside the fracture.

SAGE-Frac uses actual job data for the simulations. It allows for changes in rate, viscosity, pressure, and other job parameters as the job progresses. The new 3D model tracks fracture growth and detects fracture branching. The new 3D analysis method can be used for vertical as well as horizontal (penny-shaped) fractures. It can handle fractures in multi-layer zones with different formation properties and in-situ stresses.

SAGE-Frac provides two types of output;

Simulation of real-time job progress. In this mode the frac data is read sequentially in exactly the same manner as during the actual job. The program displays the onset of obstructions to proppant movement, branching, location and severity of obstructions, and how long their effect lasts.
Computations, and a plot, of created and propped fracture lengths, as well as the maximum created fracture width versus time. This output is subsequently used as a guide for design changes to future treatments.

SAGE-Frac is used with three companion programs called FLFric and SLFric, and LO-Calc.

FL-Fric (Fluid Friction Program) uses the actual job data to compute the frictional pressure of the fracturing fluid in the actual frac tubulars. The statistical model used in this program computes the variations of clean fluid friction pressure versus the injection rate. Since the output of this program is based on the actual job data, by comparing the actual friction pressure with the laboratory data, it can additionally provide a quality analysis measure for the consistency and viscosity of the actual frac fluid. We have seen cases of excellent agreement between the lab and actual data, and cases where the difference between the two was larger than “reasonable”.

SL-Fric (Slurry Friction Program) is based on our proprietary technology for the computation of the slurry friction pressure using actual job data and frac tubulars. To our knowledge, all existing slurry friction pressure equations use approximate correlations developed in the laboratory. Downhole pressures computed with SL-Fric are therefore more accurate and better reflect the actual fracture behavior. Using this data, we are also able to determine if perforations create an obstruction to movement of proppant from the wellbore into the fracture.

LO-Calc (Leak-Off Calculation) computes the leak-off coefficients of the fracturing fluid. Its main difference with existing programs is that it is linked with SAGE-Frac and considers the effect of branching on computation of leak-off coefficients. This difference is quite substantial and gives us the ability to avoid the common mistake by service companies and others who mis-interpret branching as high leak-off and unnecessarily increase the pad volume to offset it. In addition to better fracturing treatments, this correction also results in substantial cost savings for the Operator.

V-Frac (Vertical Fracture Design). This is our basic program for the design of vertical hydraulic fractures. The input and output of this program are basically the same as in many other design programs in use by the industry today. The main difference is that we use our own proprietary width equation that we have derived based on our experience with SAGE-Frac. We also include branching in the design of the fracture.

H-Frac (Horizontal Fracture Design). This is the basic program for the design of horizontal fractures. All comments made for V-Frac also apply to H-Frac. Again, the main difference between this and other programs in use by the industry is that we use our own proprietary width equation which we have derived based on our experience with SAGE-Frac. We also include branching in the design of the fracture.

Both of the above have routines for computation of temperature distribution inside the fracture and correct the fluid viscosity based on this temperature.

Using these programs we have been able to accomplish the following;

Better job treatments by identifying the problem areas, and changing job design and execution to address them.
Lower job costs by;

Proper selection of job volume, proppant amount, rate, viscosity, etc.
Establishing the useful part of the fracture treatment and eliminating the part that does not add value.

Determining the limits imposed by the formation on the treatment, such as fracturing out of zone, branching caused by formation inhomogeneiety, etc.
Completion-related problems interfering with successful fracturing treatments. For example, we have been able to detect fracturing problems caused by creation of multiple fractures, poor cement jobs, problems caused by downhole tools (such as interference from gravel pack tools used in frac-pack treatments), etc.
Treatment problems resulting from poor fluid quality.
Treatment problems resulting from poor equipment performance.
Treatment problems related to poor job design (rate, proppant concentration, job volume, proppant amount, etc.).