Understanding Strain Signals from Low-Frequency Distributed Acoustic Sensing In-Well Measurements: Experimental and Numerical Modeling

Abstract

Summary Over the past few decades, the potential of distributed measurements using in-well fiberoptic data has been demonstrated. In particular, low-frequency distributed acoustic sensing (LF-DAS) has experienced rapid growth toward monitoring wells for integrity and production purposes. Despite technological advancements, applications in this space have been hindered by a limited understanding of the data. The richness of detail in the data is a double-edged sword; while it provides substantial amounts of information to guide the operator, it simultaneously complicates the task of deciphering the underlying signals. To address this issue, it would be advantageous to devise a straightforward approach to understand the nature of common signals due to strain encountered during routine well operations. We present an experimental method for interpreting tubing strain signals using a long spring that is fixed at both ends, mimicking the tubing in a well. Tracking the position of coils over time allows us to record the displacement when a force is applied to the spring. We show that the displacement observed from such an experiment is similar to what we observe in LF-DAS data from wells in operation. Typical signals, such as the pistoning effect on a valve or strain caused by fluid flow, are compared with experimental data. More complex phenomena, such as stick/slip friction and thermal expansion, are modeled using a mass-spring system and compared with wellbore examples. Developing a fundamental understanding of the signals will allow for real-time identification of events, facilitated by fiber-optic data, substantially enhancing operational outcomes by preempting integrity issues and promoting production optimization.