New Insight Into Data Processing of Nuclear Magnetic Resonance Logging for Pore Structure Characterization in Tight Sandstone Reservoirs

The characterization of reservoir pore structure is crucial for the exploration and development of oil, gas, and water resources, as well as for carbon dioxide sequestration. Nuclear magnetic resonance (NMR) logging is an effective method for evaluating reservoir pore structure. The traditional NMR logging application requires the inversion of echo data to obtain relaxation time spectra, followed by the extraction of petrophysical parameters for reservoir pore structure evaluation. However, the uncertainty of the inversion results from NMR logging echo data with low signal-to-noise ratios (SNRs) can significantly impact the accuracy of NMR-based pore structure evaluation. In this article, a novel method based on integral transformation is proposed for NMR logging data processing, which aims to evaluate the reservoir pore structure while avoiding the influence of uncertainty in echo data inversion. First, six characteristic parameters directly extracted from the echo data using integral transformation are initially selected based on numerical simulations, and four parameters with stronger noise resistance ability are obtained. Then, based on the analysis of core experimental data, a secondary selection of the four integral transformation parameters is performed, resulting in two parameters that exhibit good correlation with capillary pressure curve (CPC) characteristics and petrophysical properties. Finally, a comprehensive index is obtained through parameter combination, which is used for pore structure characterization in tight sandstone samples and reservoirs. The effectiveness of the proposed NMR-based pore structure characterization method using integral transformation is verified through numerical simulations, core experiments, and well-logging data processing.