A novel approach to estimate sand particle-size using convolutional neural network with acoustic sensing

Sung, Yeongho; Gyun Lim, Hae; Keon Kim, Jang; Won, Jongmuk; Choi, Hangseok

doi:10.1016/j.catena.2023.107639

Cited by 3 publications

(2 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering that the resonant frequencies of force signals obtained from strain gauges are more accurate than those of velocity signals recorded using an accelerometer, the integration of velocity signals can influence DCR estimates (Kim et al, 2022). Furthermore, even before being subjected to any impact, the accelerometer may produce signals representing noise; integrating the noise present in the subsequently acquired acceleration signals then introduces an error in the velocity signals, which increases with the integration time (Thong et al, 2004). Thus, in this study, the DCRs at each depth were evaluated using the F method, with t 1 selected as the upper bound for the integration.…”

Section: F I G U R Ementioning

confidence: 99%

“…Specifically, Lim et al (2023) monitored the dewatering process of fine-containing mine tailings using a one-dimensional convolutional neural network. Moreover, Sung et al (2024) suggested an approach to predict the particle size of sands using convolutional neural networks combined with acoustic sensing. However, no studies thus far have attempted to use ML to estimate DCR.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated signal‐based evaluation of dynamic cone resistance via machine learning for subsurface characterization

Aregbesola,

Byun

2024

Computer aided Civil Eng

View full text Add to dashboard Cite

Dynamic cone resistance (DCR) is a recently introduced soil resistance index that has the unit of stress. It is determined from the dynamic response at the tip of an instrumented dynamic cone penetrometer. However, DCR evaluation is generally a manual, time‐consuming, and error‐prone process. Thus, this study investigates the feasibility of determining DCR using a stacked ensemble (SE) machine learning (ML) model that utilizes signals obtained from dynamic cone penetration testing. Two ML experiments revealed that using only force signals provides more accurate predictions of DCR. In addition, the SE technique outperformed the base learning algorithms in both cases. Overall, the findings suggest that ML techniques can be used to automate the analysis of DCR with force and acceleration signals.

show abstract

Section: F I G U R Ementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automated signal‐based evaluation of dynamic cone resistance via machine learning for subsurface characterization

Aregbesola,

Byun

2024

Computer aided Civil Eng

View full text Add to dashboard Cite

show abstract

Internal pipe corrosion assessment method in water distribution system using ultrasound and convolutional neural networks

Sung,

Jeon,

Kim

et al. 2024

npj Clean Water

View full text Add to dashboard Cite

Internal pipe corrosion within water distribution systems leads to iron oxide deposits on pipe walls, potentially contaminating the water supply. Consuming iron oxide-contaminated water can cause significant health issues such as gastrointestinal infections, dermatological problems, and lymph node complications. Therefore, non-destructive and continuous monitoring of pipe corrosion is imperative for water sustainability initiatives. This study introduces a dual-mode methodology utilizing advanced ultrasound technology and convolutional neural networks (CNN) to quantify pipe corrosion. Scanning acoustic microscopy (SAM) employs high-frequency ultrasound to generate high-resolution images of pipe thickness, indicating iron oxide accumulation. SAM also captures internal pipe data to measure iron oxide concentration in the water. This data, analyzed by CNN, achieves an impressive 95% accuracy. This dual-mode system effectively assesses both the extent of pipe corrosion and water contamination, exemplifying the successful integration of SAM and CNN for precise and reliable monitoring.

show abstract

Quantitative Assessment of Sand Particulates in Gas-Water Slug Flow Using Deep Learning

Wang,

Tian,

Chang

et al. 2024

SPE Journal

View full text Add to dashboard Cite

Summary The weak collision response excited by micrometer-scale sand particulates is prone to overmixing with strong slug noise, significantly reducing the characterization and monitoring accuracy of sand particulate information in slug flows. Therefore, we developed a quantitative assessment method for sand particulates in slug flow that combines triaxial vibration monitoring and deep learning. First, a migration behavior characterization method of sand particulates is proposed combining nonlinear statistics, multifrequency coherence, and multiscale time frequency. The multifrequency response characteristics corresponding to the multiscale flow behavior of the sand-carrying slug flow were successfully characterized on the 2D time-frequency plane, namely, the mixed sand migration behavior [Intrinsic Mode Function 1 (IMF1)], liquid slug sand carrying (IMF2), forward liquid film and Taylor bubble sand carrying (IMF3), and reflux liquid film sand carrying (IMF4). Furthermore, the influence mechanism of gas superficial velocity (1.5–3.5 m/s), liquid superficial velocity (0.95–2.14 m/s), and sand content (0–20 g) on the triaxial vibration response of slug particulate flow with different migration behaviors is elucidated. Finally, a convolutional neural network (CNN)-gated recurrent unit (GRU)-self-attention mechanism (SATT) model for sand content assessment is developed based on the characterized multiscale migration behavior information and achieves an average recognition accuracy of 95.55% for data sets representing different sand migration behaviors in slug flow. This provides a new method for precisely identifying and monitoring sand production information of multiphase pipe flow.

show abstract

A novel approach to estimate sand particle-size using convolutional neural network with acoustic sensing

Cited by 3 publications

References 49 publications

Automated signal‐based evaluation of dynamic cone resistance via machine learning for subsurface characterization

Automated signal‐based evaluation of dynamic cone resistance via machine learning for subsurface characterization

Internal pipe corrosion assessment method in water distribution system using ultrasound and convolutional neural networks

Quantitative Assessment of Sand Particulates in Gas-Water Slug Flow Using Deep Learning

Contact Info

Product

Resources

About