Abstract

We combine active and passive acoustic measurements to improve the spatio-temporal imaging of hydraulic fracture growth performed under true triaxial confinement in the laboratory. 64 active piezo-electric transducers (54 P waves, 10 S waves) work in source (32) - receivers (32) mode to perform an acoustic survey at repetitive intervals (every 10 seconds) during a hydraulic fracture growth experiment. The analysis of the evolution of the active acoustic monitoring changes allow to image the evolution of the fracture front (via an inversion of the active acoustic waves diffracted by the fracture front). An additional 16 piezoelectric transducers are pre-amplified and work in passive mode continuously recording at 10MHz. We present a nearly-automatic passive signal processing, acoustic emission detection, and location algorithm. This allows to record, detect and to locate acoustic emissions associated with fracture initiation and growth in between the active acoustic measurement sequences. Using a hydraulic fracturing test performed in gabbro, we discuss how active and passive acoustic monitoring complements one another and bring different type of information on hydraulic fracture growth.

Introduction

Hydraulic fractures are a class of tensile fractures that propagate in a material as a result of fluid pressurization. Investigation of the growth of such fluid-driven fractures under controlled conditions at the laboratory scale plays an important role in order to validate theoretical predictions. We combine active acoustic imaging and passive acoustic emission (AE) monitoring to capture the spatio-temporal evolution of hydraulic fracturing in rocks at the laboratory scale.

Active acoustic imaging is based on a 4D seismic survey at the laboratory scale in the ultrasonic range: an acoustic survey is repeated at regular intervals during the experiment. Earlier studies [1, 2, 3, 4, 5] have shown its capability to obtain quantitative information on fracture growth during laboratory hydraulic fracture experiments. The active wave-field can be diffracted by the fracture tip (as well as the fluid front if a lag is present near the fracture tip) and also reflected by and transmitted through the fluid-filled fracture. The time-evolution of the transmitted waves notably allows to identify a dry region near the fracture tip (fluid lag) [1, 2]. Records of the evolution of the arrival times of the active wave-fields diffracted by the fracture tip have enable to estimate the evolving fracture tip position under the hypothesis of a horizontal radial fracture centered on the injection point [5, 6]. The fluid layer thickness in the fracture (i.e. its opening) can also be estimated by matching the spectrum of the transmitted signals with the transmission coefficient of a three layers model, considering attenuation and delay of transmitted waves [7]. The diffraction and transmission techniques have been shown to provide results in agreement with optical methods [8]. We have recently extended these active acoustic techniques increasing the number of sensors and improving the fracture front reconstruction algorithm [9].

This content is only available via PDF.
You can access this article if you purchase or spend a download.