Research on the processing and interpretation methods of distributed fiber optic vibration signal logging injection profiles

Abstract

The oil and gas industry currently faces the dual challenges of improving extraction efficiency and reducing environmental impacts. Traditional well monitoring technologies are increasingly unable to meet the demands of modern oil and gas extraction due to technological limitations and environmental concerns. Distributed Acoustic Sensing (DAS) technology, which utilizes optical fibers as sensing elements, enables real-time and accurate monitoring of the CO2 injection process in wells. However, DAS technology encounters issues such as weak amplitude signals and noise interference in practical applications. In response, a comprehensive method for processing and interpreting DAS logging data for CO2 injection wells has been proposed. This method involves restoring amplitude signals through wavefront energy compensation and applying Gaussian filtering and wavelet threshold denoising algorithms for noise reduction. The processed results are classified using the K-Medoids clustering algorithm and the CO2 absorption volume in the injection layer is calculated using the integral area method. The accuracy and practicality of this method have been validated through comparative analysis with pulse neutron oxygen activation logging results. This approach aims to address the challenges faced by DAS technology in the oil and gas industry, thereby providing technical support for the efficient, environmentally friendly, and intelligent management of oil and gas fields.