Abstract: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.

PROCESSING OF LOW SIGNAL/NOISE RATIO NMR LOGGING DATA

A Noise Reduction Method for Non-Stationary Noise Model of SPECT Sinogram Based on Kalman Filter

A statistical learning perspective on the inversion of NMR relaxation data

A Novel 2-D Inversion Method for Low-Field NMR Working on Fluid Typing

A Generally Regularized Inversion for NMR Applications and Beyond

A machine learning framework for low-field NMR data processing

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

Combined Transformation of Ordering SPECT Sinograms for Signal Extraction from Measurements with Poisson Noise.

A novel denoising method for low SNR NMR Logging echo signal based on deep learning

NQR Signal Processing Based on Multi-Stage Wiener Filter

An Improved Dual Asymmetric Penalized Least Squares Baseline Correction Method for High-Noise Spectral Data Analysis

An Intelligent Denoising Method for Nuclear Magnetic Resonance Logging Measurement Based on Residual Network

An Inversion Method for Calculating Formation and Borehole Parameters in X-ray Lithology Density Logging

3-D Inversion of the Data-Normalized Time-Lapse Controlled-Source Audio-Frequency Magnetotelluric Method

Separate and Detailed Treatment of Absolute Signal and Noise Enables NMR Under Adverse Circumstances

Nonstationary signal inversion based on shaping regularization for random noise attenuation

Enhancing signal-to-noise ratio and resolution in low-field NMR relaxation measurements using post-acquisition digital filters

Multi‐trace basis‐pursuit seismic inversion for resolution enhancement

Deep Learning as a Method for Inversion of NMR Signals

The Discrete Newton Method of the Three Parameters Inversion in Electrical Logging

Stationary wavelet denoising of solid-state NMR spectra using multiple similar measurements