ABSTRACT
This paper presents a three-dimensional hydraulic-fracturing simulator permitting direct input of the expected net pressure at the end of a hydraulic fracture treatment – a parameter that can be readily derived from field measurements. The simulator can thus easily be used at the wellsite to generate field-calibrated fracture geometries and optimised pump schedules. It embodies a twodimensional description of fluid flow in the fracture. Rapid fracture growth (e.g. through negative stress barriers) is handled in a stable manner.
Propagation through layers with different Young's moduli is described using the Barenblatt approach, giving a realistic estimate of containment by modulus contrasts: commercially available models generally overestimate the containment effect. This is of importance for the simulation of drill-cutting reinjection, where stiffness layering may enhance possible containment by soft, permeable sand layers interbedding the thick shales. The model can also handle the continuous stiffness gradient in these shales.
Benchmark comparison with data from the GRI staged field experiment No. 3 shows that the new model compares well with existing models for this case. The performance of the model is illustrated for three field cases. The first case concerns the design of a massive hydraulic fracturing treatment offshore The Netherlands, in a reservoir consisting of five layers with negative stress barriers. The simulator was used to optimise the perforation location and the pump schedule, using an expected net pressure at shut-in of 500 psi. The observed onset of tip screen-out at the end of the treatment suggests that the schedule as it was pumped away was indeed (nearly) optimum. The treatment produced an initial transient productivity improvement factor of about five.
The other two field cases are examples of different North Sea drill cuttings reinjection scenarios. The possible importance of stiffness layering between the thick shales and thin interbedding permeable sands when injecting large volumes (e.g. from a dedicated injection well) is highlighted. The simulator is used to assess the relative merits of intermittent injection with long shut-in intervals (one month) and continuous injection.
