Anna Stroisz

Stress-Induced Fracturing of Reservoir Rocks: Acoustic Monitoring and μCT Image Analysis

Stress-induced fracturing in reservoir rocks is an important issue for the petroleum industry. While productivity can be enhanced by a controlled fracturing operation, it can trigger borehole instability problems by reactivating existing fractures/faults in a reservoir. However, safe fracturing can improve the quality of operations during CO2 storage, geothermal installation and gas production at and from the reservoir rocks. Therefore, understanding the fracturing behavior of different types of reservoir rocks is a basic need for planning field operations toward these activities. In our study, stress-induced fracturing of rock samples has been monitored by acoustic emission (AE) and post-experiment computer tomography (CT) scans. We have used hollow cylinder cores of sandstones and chalks, which are representatives of reservoir rocks. The fracture-triggering stress has been measured for different rocks and compared with theoretical estimates. The population of AE events shows the location of main fracture arms which is in a good agreement with post-test CT image analysis, and the fracture patterns inside the samples are visualized through 3D image reconstructions. The amplitudes and energies of acoustic events clearly indicate initiation and propagation of the main fractures. Time evolution of the radial strain measured in the fracturing tests will later be compared to model predictions of fracture size.

Elastic Dispersion Derived from a Combination of Static and Dynamic Measurements

Utilization of laboratory tests for calibration and interpretation of data from seismic surveys requires knowledge about elastic dispersion in the range from seismic to ultrasonic frequencies. Data on such dispersion are hard to obtain because it requires specially designed equipment and also relies on simplifying assumptions about rock symmetry. A new method for estimation of dispersion in this frequency range is presented here. This method requires only standard rock mechanical equipment with ultrasonic velocity measurements, and is based on comparison of static and dynamic data. A key element in this method is a procedure for elimination of strain amplitude as a source for differences between static and dynamic moduli. High-quality data is necessary, but the required accuracy is not extreme. Application of the method on one partly saturated shale and two dry sandstone samples indicates that dispersion increases with clay content, and decreases with stress.

A simple discrete-element-model of Brazilian test

Abstract We present a statistical model which is able to capture some interesting features exhibited in the Brazilian test of rock samples. The model is based on elements which break irreversibly when the force experienced by the elements exceed their own load capacity. If an element breaks the load capacity of the neighboring elements are decreased by a certain amount, assuming weakening effect around the defected zone. From the model we numerically investigate the stress-strain behavior, the strength of the system, how it scales with the system size and also its fluctuation for both uniform and Weibull distribution of breaking thresholds in the system. To check the validity of our statistical model we perform few Brazilian tests on Sandstone and Chalk samples. The stress-strain curve from model results agree qualitatively well with the lab-test data. Also, the damage profile right at the point when the stress-strain curve reaches its maximum is seen to mimic the crack patterns observed in our Brazilian test experiments.