Signal-to-noise ratio enhancement for 3C downhole microseismic data based on the 3D shearlet transform and improved back-propagation neural networks

Geophysics ◽  
2019 ◽  
Vol 84 (4) ◽  
pp. V245-V254 ◽  
Author(s):  
Xintong Dong ◽  
Hong Jiang ◽  
Sheng Zheng ◽  
Yue Li ◽  
Baojun Yang
Keyword(s):  
Neural Networks ◽  
Signal To Noise Ratio ◽  
Random Noise ◽  
Back Propagation ◽  
Bp Neural Networks ◽  
Signal To Noise ◽  
Shearlet Transform ◽  
Microseismic Data ◽  
Noise Ratio ◽  
Selection Of

As the seismic responses of unconventional hydraulic fracturing, downhole microseismic signals play an essential role in the exploitation of unconventional oil and gas reservoirs. In geologic structure interpretation and reservoir development, high-quality downhole microseismic data are necessary. However, the characteristics of downhole microseismic signals, such as weak energy and high frequency, bring great difficulty to signal-to-noise ratio enhancement. How to suppress the random noises in 3C downhole microseismic signals becomes problematic. To solve this problem, the 3D shearlet transform is introduced into downhole microseismic data processing. Different from the 2D shearlet transform, the correlation among the 3C of downhole microseismic signals is fully considered in the 3D shearlet transform, which enables the 3D shearlet transform to suppress random noise more effectively. In addition, for accurate selection of 3D shearlet coefficient, the back-propagation (BP) neural network is applied to the selection of coefficients. Unlike conventional threshold functions, BP neural networks can achieve optimal results by repeated training. At the same time, a new weight factor is proposed to improve the misconvergence of BP neural networks. Experimentally our method has been used to process synthetic and real 3C downhole microseismic signals, with results indicating that, compared with conventional methods, our new algorithm exhibits better performance in valid signal preservation and random noise suppression.

Signal-to-noise ratio enhancement for downhole microseismic data based on 3D shearlet transform

Geophysics ◽  
2019 ◽  
Vol 84 (3) ◽  
pp. V207-V218
Author(s):  
Juan Li ◽  
Yuan Li ◽  
Shou Ji ◽  
Yue Li ◽  
Zhihong Qian
Keyword(s):  
Noise Suppression ◽  
Signal To Noise Ratio ◽  
Random Noise ◽  
Great Difficulty ◽  
Threshold Function ◽  
Signal To Noise ◽  
Shearlet Transform ◽  
Microseismic Data ◽  
3D Matrix ◽  
Noise Ratio

Downhole microseismic data are characterized for their high frequency and small amplitude, which bring great difficulty for noise suppression. We present a random noise attenuation method for downhole microseismic data based on the 3D shearlet transform (3DST). In contrast to the 2D shearlet, 3DST takes into account the correlation among three components of downhole microseismic. With the help of correlation among the data, downhole microseismic data are reassembled into a new 3D matrix and then transformed to the shearlet domain. After the analysis of the coefficients’ energy and the high-order cumulant on each scale, an efficient threshold function is proposed. We apply a small threshold to the coefficients associated with the signal’s scales, and a large threshold is chosen for the scales of the noise. Experimental results indicate that the algorithm significantly improves the signal-to-noise ratio of the microseismic data and effectively preserves a valid signal.

Analysis of neural networks efficiency when accounting for weight coefficients jitter, leading to reduction in output signal/noise ratio

2021 ◽  
Author(s):  
S.V. Zimina
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Output Signal ◽  
Signal To Noise Ratio ◽  
Signal To Noise ◽  
Useful Signal ◽  
Artificial Neural ◽  
Noise Ratio ◽  
Weight Coefficients

Setting up artificial neural networks using iterative algorithms is accompanied by fluctuations in weight coefficients. When an artificial neural network solves the problem of allocating a useful signal against the background of interference, fluctuations in the weight vector lead to a deterioration of the useful signal allocated by the network and, in particular, losses in the output signal-to-noise ratio. The goal of the research is to perform a statistical analysis of an artificial neural network, that includes analysis of losses in the output signal-to-noise ratio associated with fluctuations in the weight coefficients of an artificial neural network. We considered artificial neural networks that are configured using discrete gradient, fast recurrent algorithms with restrictions, and the Hebb algorithm. It is shown that fluctuations lead to losses in the output signal/noise ratio, the level of which depends on the type of algorithm under consideration and the speed of setting up an artificial neural network. Taking into account the fluctuations of the weight vector in the analysis of the output signal-to-noise ratio allows us to correlate the permissible level of loss in the output signal-to-noise ratio and the speed of network configuration corresponding to this level when working with an artificial neural network.

scholarly journals Collecting Microplastics in Gardens: Case Study (i) of Soil

2021 ◽  
Vol 9 ◽  
Author(s):  
Zahra Sobhani ◽  
Yunlong Luo ◽  
Christopher T. Gibson ◽  
Youhong Tang ◽  
Ravi Naidu ◽  
...  
Keyword(s):  
Signal To Noise Ratio ◽  
Point Of View ◽  
Signal To Noise ◽  
Statistical Point ◽  
Image Mapping ◽  
Noise Ratio ◽  
In The Beginning ◽  
Raman Image ◽  
Selection Of

As an emerging contaminant, microplastic is receiving increasing attention. However, the contamination source is not fully known, and new sources are still being identified. Herewith, we report that microplastics can be found in our gardens, either due to the wrongdoing of leaving plastic bubble wraps to be mixed with mulches or due to the use of plastic landscape fabrics in the mulch bed. In the beginning, they were of large sizes, such as > 5 mm. However, after 7 years in the garden, owing to natural degradation, weathering, or abrasion, microplastics are released. We categorize the plastic fragments into different groups, 5 mm–0.75 mm, 0.75 mm–100 μm, and 100–0.8 μm, using filters such as kitchenware, meaning we can collect microplastics in our gardens by ourselves. We then characterized the plastics using Raman image mapping and a logic-based algorithm to increase the signal-to-noise ratio and the image certainty. This is because the signal-to-noise ratio from a single Raman spectrum, or even from an individual peak, is significantly less than that from a spectrum matrix of Raman mapping (such as 1 vs. 50 × 50) that contains 2,500 spectra, from the statistical point of view. From the 10 g soil we sampled, we could detect the microplastics, including large (5 mm–100 μm) fragments and small (<100 μm) ones, suggesting the degradation fate of plastics in the gardens. Overall, these results warn us that we must be careful when we do gardening, including selection of plastic items for gardens.

Random denoising and signal nonlinearity approach by time-frequency peak filtering using weighted frequency reassignment

Geophysics ◽  
2013 ◽  
Vol 78 (6) ◽  
pp. V229-V237 ◽  
Author(s):  
Hongbo Lin ◽  
Yue Li ◽  
Baojun Yang ◽  
Haitao Ma
Keyword(s):  
Weight Function ◽  
Instantaneous Frequency ◽  
Seismic Data ◽  
Signal To Noise Ratio ◽  
Random Noise ◽  
Signal To Noise ◽  
Transform Method ◽  
Time Frequency ◽  
Frequency Peak ◽  
Noise Ratio

Time-frequency peak filtering (TFPF) may efficiently suppress random noise and hence improve the signal-to-noise ratio. However, the errors are not always satisfactory when applying the TFPF to fast-varying seismic signals. We begin with an error analysis for the TFPF by using the spread factor of the phase and cumulants of noise. This analysis shows that the nonlinear signal component and non-Gaussian random noise lead to the deviation of the pseudo-Wigner-Ville distribution (PWVD) peaks from the instantaneous frequency. The deviation introduces the signal distortion and random oscillations in the result of the TFPF. We propose a weighted reassigned smoothed PWVD with less deviation than PWVD. The proposed method adopts a frequency window to smooth away the residual oscillations in the PWVD, and incorporates a weight function in the reassignment which sharpens the time-frequency distribution for reducing the deviation. Because the weight function is determined by the lateral coherence of seismic data, the smoothed PWVD is assigned to the accurate instantaneous frequency for desired signal components by weighted frequency reassignment. As a result, the TFPF based on the weighted reassigned PWVD (TFPF_WR) can be more effective in suppressing random noise and preserving signal as compared with the TFPF using the PWVD. We test the proposed method on synthetic and field seismic data, and compare it with a wavelet-transform method and [Formula: see text] prediction filter. The results show that the proposed method provides better performance over the other methods in signal preserving under low signal-to-noise ratio.

Pitfalls in horizon autopicking

2015 ◽  
Vol 3 (1) ◽  
pp. SB1-SB4 ◽  
Author(s):  
Donald A. Herron
Keyword(s):  
Data Quality ◽  
Signal To Noise Ratio ◽  
Signal To Noise ◽  
Tracking Algorithms ◽  
Input Control ◽  
Noise Ratio ◽  
Selection Of

Interpreters use horizon autopicking in many seismic interpretations in the modern workstation environment. When properly used and with data quality permitting this technique enables efficient and accurate tracking of horizons but is not without its pitfalls. Four common pitfalls are improper selection of the input control or seed grid, not accounting for the “directional” behavior of tracking algorithms, attempting autopicking in areas with poor reflection continuity and/or low signal-to-noise ratio, and failing to recognize elements of geology that are not suitable for autopicking.

scholarly journals A Blind Spectrum Sensing Method Based on Deep Learning

Sensors ◽  
2019 ◽  
Vol 19 (10) ◽  
pp. 2270 ◽  
Author(s):  
Kai Yang ◽  
Zhitao Huang ◽  
Xiang Wang ◽  
Xueqiong Li
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Spectrum Sensing ◽  
Short Term Memory ◽  
Signal To Noise Ratio ◽  
Signal To Noise ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Noise Ratio

Spectrum sensing is one of the technologies that is used to solve the current problem of low utilization of spectrum resources. However, when the signal-to-noise ratio is low, current spectrum sensing methods cannot well-handle a situation in which the prior information of the licensed user signal is lacking. In this paper, a blind spectrum sensing method based on deep learning is proposed that uses three kinds of neural networks together, namely convolutional neural networks, long short-term memory, and fully connected neural networks. Experiments show that the proposed method has better performance than an energy detector, especially when the signal-to-noise ratio is low. At the same time, this paper also analyzes the effect of different long short-term memory layers on detection performance, and explores why the deep-learning-based detector can achieve better performance.

Accurate Measurement of Trace Elements Using an Innovative Fixed Goniometer for a Simultaneous Spectrometer

1994 ◽  
Vol 38 ◽  
pp. 691-698
Author(s):  
K. Kansai ◽  
K. Toda ◽  
H. Kohno ◽  
T. Arai ◽  
R. Wilson
Keyword(s):  
Trace Elements ◽  
Exposure Time ◽  
Background Subtraction ◽  
Signal To Noise Ratio ◽  
Detection Limits ◽  
Background Correction ◽  
Analysis Time ◽  
Signal To Noise ◽  
Noise Ratio ◽  
Selection Of

Advancements in trace clement analysis require improvements in both the signal-to-noise ratio and accurate background correction. With a sequential spectrometer, one can obtain detection limits of around 0.1 ppm for medium to heavy Z elements. Conditions can be individually optimized for each element, for example, selection of filters, collimators, crystals and background subtraction. The disadvantage is that the analysis time may become “long” if many elements are to be analyzed. This long exposure time can lead to the deterioration of some samples.

Sign in / Sign up

Export Citation Format

Share Document