What Do Hydraulic Fractures Look Like in Different Types of Reservoirs: Implications from a Series of Large-Scale Polyaxial Hydraulic Fracturing Experiments from Conventional to Unconventional

Abstract

Fracture geometry is always the focus of stimulation evaluation, especially for unconventional reservoirs. Though treating pressure calibration, hydraulic fracture modeling and microseismic monitoring throw some lights on the fracture characteristics, operators still are far from understanding the real fractures for its apparent complex process. This paper presents a series of 1m³ large block fracturing tests in different type of rocks from corresponding formation outcrops to help acquire some common senses on fracture propagation.

Hydraulic fracturing polyaxial tests were conducted on large scale homogeneous sandstone, carbonate, coal, shale and highly laminated tight oil rocks. To investigate the hydraulic fracture propagation mechanism and decisive factors in different rocks in normal condition, we use integrated analysis including fracturing pressure analysis, acoustic emission monitoring, rock splitting and tracing the dyed fracturing fluid. We also evaluate some industry-recognized factors affecting the fracture propagation such as vertical stress differences or weak planes induced fracture containment, fluid viscosity, natural fractures, horizontal beddings, horizontal stress contrast and orientation. Field microseismic monitoring results were also presented with same rock fabrics and geomechanics.

In homogeneous sandstone and carbonate, after breaking down, the pressure in the fracture keeps declining until breaking out of the rock border. Acoustic monitoring results show that hydraulic fracture propagates forward and upward simultaneous and the complete dynamic fracture propagation process is visualized. The dyed fluid on the rock confirms traditional bi-wing fractures and the geometry is correspondent with the monitoring result. This scenario can be used to represent the fracture breakdown process or fracturing large homogeneous interval. While fracturing highly laminated tight rocks, the hydraulic pressure keeps increasing steadily which represent the severe fracture containment or high process-zone stress. Rock splitting and acoustic emission mapping results also proved that fracture initiated and propagated along the horizontal-bedding. Coal and shale fracturing results proved that the rock heterogeneity and natural fracture play a decisive role in fracture propagation. Vertical stress differences is a major factor of fracture containment and low viscosity fluid also limit the fracuture height. Horizontal bedding severely confined and hindered the fracture propagating vertically, and the laboratory tests showed that induced bedding-parallel stimulation was performed. Laboratory tests and field microseismic monitoring results both indicate that the real fracture geometries are much more complicated than we predicted.

Even though there are big differences between laboratory tests and field stimulation in the aspect of rock, scale, geomechanics, fluid and pump rate. However, they still have proved some inherent mechanisms of fracture propagation for stimulation engineers. Rock fabric such as horizontal bedding, natural fracture are often overlooked and unquantifiable factors when considering the fracture geometry, which may even play a decisive role in fracture propagation.