Wellbore fluid flow velocity calculation based on distributed acoustic sensing data

Abstract Monitoring fluids within the wellbore is essential for preventing hazardous events and optimizing production parameters to enhance output. Over the past decade, distributed acoustic sensing has been extensively used in underground monitoring due to its capabilities for real-time, full-depth monitoring and resistance to corrosion and interference. One of the key applications is monitoring injection or production profiles underground. There are two primary methods for monitoring injection or production profiles using distributed acoustic sensing. The first method determines the relative proportion of flow at each location based on the energy of flow-induced noise. The second method calculates the flow velocity profile inside the well, which can track the plug interface or be implemented using the Doppler effect. Currently, methods based on the Doppler effect are primarily implemented by calculating the two-dimensional Fourier transform of the data. However, this approach has limitations in spatial resolution. We propose a method combining cross-correlation and clustering to accurately calculate flow velocity with fewer channels, effectively identifying multiple velocities within a segment. The superiority of this method in calculating flow velocity using 30 channels has been verified through actual data comparison. Using this method to calculate fluid flow velocity within a smaller depth interval improves the spatial resolution of the production or injection profiles and providing a basis for a more accurate assessment of injection or production conditions.