JZUS - Journal of Zhejiang University SCIENCE

Journal of Zhejiang University SCIENCE A

Accepted manuscript available online (unedited version)

Predicting permeability coefficients of earth-rock material using improved GAN and explainable ensemble learning under small sample conditions

Author(s): Chengyu YU, Hongling YU, Xiaofeng QU, Baoxi LIU, Liangsi XU, Xinyu LIU, Xiangyu CHEN
Affiliation(s): College of Water Resources and Civil Engineering, China Agricultural University, Beijing 100083, China
Corresponding email(s): yuhongling@cau.edu.cn
Key Words: Permeability Coefficient Prediction; LightGBM; Wasserstein Conditional Generative Adversarial Nets; SHAP

Share this article to： More <<< Previous Paper \|Next Paper >>>

Chengyu YU, Hongling YU, Xiaofeng QU, Baoxi LIU, Liangsi XU, Xinyu LIU, Xiangyu CHEN. Predicting permeability coefficients of earth-rock material using improved GAN and explainable ensemble learning under small sample conditions[J]. Journal of Zhejiang University Science A,in press.Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/jzus.A2500127

@article{title="Predicting permeability coefficients of earth-rock material using improved GAN and explainable ensemble learning under small sample conditions",
author="Chengyu YU, Hongling YU, Xiaofeng QU, Baoxi LIU, Liangsi XU, Xinyu LIU, Xiangyu CHEN",
journal="Journal of Zhejiang University Science A",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/jzus.A2500127"
}

%0 Journal Article
%T Predicting permeability coefficients of earth-rock material using improved GAN and explainable ensemble learning under small sample conditions
%A Chengyu YU
%A Hongling YU
%A Xiaofeng QU
%A Baoxi LIU
%A Liangsi XU
%A Xinyu LIU
%A Xiangyu CHEN
%J Journal of Zhejiang University SCIENCE A
%P
%@ 1673-565X
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/jzus.A2500127"

TY - JOUR
T1 - Predicting permeability coefficients of earth-rock material using improved GAN and explainable ensemble learning under small sample conditions
A1 - Chengyu YU
A1 - Hongling YU
A1 - Xiaofeng QU
A1 - Baoxi LIU
A1 - Liangsi XU
A1 - Xinyu LIU
A1 - Xiangyu CHEN
J0 - Journal of Zhejiang University Science A
SP -
EP -
%@ 1673-565X
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/jzus.A2500127"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Accurate prediction of the permeability coefficient is crucial for evaluating the compaction quality of earthworks. However, during the compaction process, on-site testing is often time-consuming and expensive, leading to fewer samples, which affects prediction accuracy. Moreover, most current predictive models have limited capabilities and tend to be black-box models with poor explainability. To overcome these issues, in this study we propose a new method to predict the permeability coefficient of earth-rock material, based on an improved Generative Adversarial Network (GAN) and explainable OOA-HL-LightGBM. Firstly, by introducing the Was-serstein distance as the loss function into the Conditional Generative Adversarial Network (CGAN), the Wasserstein Conditional Generative Adversarial Network (WCGAN) has been proposed to generate high-quality data, addressing the issue of insufficient information caused by small samples. Furthermore, by incorporating material and compaction parameters as inputs, a high-accuracy permeability coefficient prediction model was developed using LightGBM with the Huber loss function and the Osprey Optimization Algorithm (OOA). Finally, the SHapley Additive exPlanations (SHAP) method was introduced into OOA-HL-LightGBM to analyze the specific roles of different features within the dataset to enhance the credibility of the prediction results. The proposed method was applied to a large-scale high-core rockfill dam in Southwest China to thoroughly verify its effectiveness and superiority.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference