A novel method for NMR data denoising based on discrete cosine transform and variable length windows

Gu, Mingxuan; Xie, Ranhong; Xiao, Lizhi

doi:10.1016/j.petrol.2021.108852

Cited by 20 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 2 illustrates the training procedure of 1D DCGAN model. The binary cross-entropy loss function BCELoss() in the table is shown in equation (13). In each iteration, D is trained first, and the loss value of D contains real loss LossD r and fake loss LossD f .…”

Section: Network Trainingmentioning

confidence: 99%

“…Gao et al developed a gray-scale morphology filter method based on the morphological difference between the echo data and noise [12]. Discrete cosine transform (DCT) based on variable length windows was applied to the noise reduction of NMR echo data [13]. The above denoising methods are mainly based on signal processing such as Gabor transform, wavelet transform, DCT, SVD, and EMD, which have achieved relatively mature research results.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Liu

Cai²,

Jiang³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

The T2 spectrum obtained by inversion of the NMR logging (NMRL) echo signal can provide various petrophysical parameters to help technicians effectively identify the type of fluid. However, raw logging data with a low signal-to-noise ratio (SNR) can cause inversion results to deviate from the truth. Therefore, a deep learning denoising method is proposed to eliminate the limitations of traditional mathematical transform methods. A one-dimension deep convolutional generative adversarial networks (1D DCGAN) model is designed to fit the noise distribution of actual logging data, then generate simulation data. A multi-scale echo denoising network (MsEDNet) is designed to adaptively learn the multi-exponential decay characteristics of the signal and the optimal transform space. The denoising dataset consists of simulation data and logging data, which is applied to train MsEDNet and improve the generalization performance. With effectiveness analysis and various denoising experiments, it is validated that the proposed method has excellent denoising performance on simulation data, logging data, and water tank data.

show abstract

Section: Network Trainingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

Liu

Cai²,

Jiang³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…where k represents the sampling points in the neighborhood nearest to the sampling point [28], z 1 and z 2 are Gaussian filtering functions in the spatial domain and frequency domain of bilateral filtering functions, respectively, and their specific forms are shown as follows:…”

Section: Point Cloud Smoothness In Feature-rich Regionsmentioning

confidence: 99%

CAD-Aided 3D Reconstruction of Intelligent Manufacturing Image Based on Time Series

Zhang

Wang

et al. 2022

Scientific Programming

View full text Add to dashboard Cite

To improve the three-dimensional (3D) reconstruction effect of intelligent manufacturing image and reduce the reconstruction time, a new CAD-aided 3D reconstruction of intelligent manufacturing image based on time series was proposed. Kinect sensor is used to collect depth image data and convert it into 3D point cloud coordinates. The collected point cloud data are divided into regions, and different point cloud denoising algorithms are used to filter and denoise the divided regions. With the help of CAD, FLANN matching algorithm is used to extract feature points of time-series images and complete image matching. Three-dimensional reconstruction of sparse point cloud and dense point cloud is carried out to complete 3D reconstruction of intelligent manufacturing images. The experimental results show that the image PSNR of this method is always above 52 dB, and the maximum reconstruction time is 4.9 s. The 3D reconstruction effect of intelligent manufacturing image is better, and it has higher practical application value.

show abstract

“…Low-field NMR T 1 – T 2 instruments are commonly used in the NMR logging field, but the resolution of fluid in the T 1 – T 2 spectrum is poor. In addition, the signal-to-noise ratio (SNR) of field-collected echo data is low, and the inversion of NMR echo data based on the inverse Laplace transform is an ill-posed problem. − In the inverted T 1 – T 2 spectrum, different fluids overlap significantly, making it difficult to extract fluid features and calculate fluid saturation. Therefore, it is necessary to develop a saturation evaluation method to provide guidelines for geologists to calculate reserves and determine the perforation location in shale.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Fluid Saturation in Shale Using 2D Nuclear Magnetic Resonance

Xie

Guo

et al. 2023

Energy Fuels

Self Cite

View full text Add to dashboard Cite

Two-dimensional (2D) nuclear magnetic resonance (NMR) has become an indispensable tool for fluid saturation evaluation in shale reservoir. NMR spectrum derived from echo data with a low signal-to-noise ratio, however, cannot be effectively used for fluid feature extraction and saturation calculation. In this study, we proposed a fluid saturation evaluation method that combines morphology, nonnegative matrix decomposition, and fully constrained least squares. Moreover, an Akaike information criterion (AIC)-based method was proposed to determine the number of fluid types. A shale formation data set with varying fluid saturation was processed to validate the effectiveness of the proposed method. The results show that the proposed method can efficiently and accurately obtain the NMR T 1 −T 2 feature and saturation of different fluid types and has advantages over the comparable method in both computational efficiency and accuracy. Furthermore, the impact of the structure size was investigated, demonstrating that the proposed method is very fault tolerant.

show abstract

A novel method for NMR data denoising based on discrete cosine transform and variable length windows

Cited by 20 publications

References 26 publications

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

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

CAD-Aided 3D Reconstruction of Intelligent Manufacturing Image Based on Time Series

Evaluation of Fluid Saturation in Shale Using 2D Nuclear Magnetic Resonance

Contact Info

Product

Resources

About