News Release

AI-based approaches for predicting buried pipeline stability in cohesive-frictional soil under inclined forces

Peer-Reviewed Publication

KeAi Communications Co., Ltd.

Configuration of Pipes Embedded in Cohesive-Frictional Soil.

image: 

Configuration of Pipes Embedded in Cohesive-Frictional Soil.

view more 

Credit: KHAMNOY KOUNLAVONG, ET AL.

Uplift resistance is a critical indicator for assessing the stability of buried pipelines under varying load conditions. To date, most studies on uplift resistance have focused on pipelines buried in granular soils like sand. However, buried pipelines are often installed in more complex soil conditions, such as cohesive-frictional soils that exhibit distinct mechanical properties due to the combined effects of cohesion and internal friction. This situation necessitates more sophisticated assessment methods for uplift resistance.

In a study published in the KeAi journal Journal of Pipeline Science and Engineering, researchers from Thailand and Vietnam assessed the uplift resistance of buried pipelines in cohesive-frictional soils using finite element limit analysis (FELA) integrated with AI techniques.

“A better understanding of the variables affecting uplift resistance is fundamental for effectively designing and installing buried pipelines, especially in regions with complex soil behavior,” explains co-corresponding author Suraparb Keawsawasvong.

To this end, the team examined the impacts of embedment depth ratios, load inclination angles, adhesion factors, internal friction angles, and soil strength ratios on uplift resistance. Their findings revealed that increasing embedment depth and load inclination angle enhanced uplift resistance by mobilizing larger volumes of surrounding soil, while higher soil strength ratios reduced uplift resistance. “Sensitivity analysis further identified embedment depth ratio as the most influential factor, followed by load inclination and adhesion, underscoring the intricate interplay of soil properties in determining pipeline stability,” adds Keawsawasvong.

In developing the predictive model, four machine learning (ML) techniques — deep neural network (DNN), recurrent neural network (RNN), long short-term memory (LSTM) network, and multivariate adaptive regression splines (MARS) — were combined to enhance predictive accuracy. The team then evaluated all proposed ML models by rank analysis, performance parameters, and scatter plots, with the DNN model exhibiting the highest predictive accuracy among the evaluated models, making it a reliable tool for predicting uplift resistance in varying soil conditions.

 “Our findings provide valuable insights into designing stable buried pipelines, particularly for offshore and seismic-prone environments,” says Keawsawasvong. “Future studies should incorporate field validations and larger datasets to further refine the predictive capabilities of machine learning models in pipeline stability assessment.”

###

Contact the author: Suraparb Keawsawasvong, Research Unit in Sciences and Innovative Technologies for Civil Engineering Infrastructures, Department of Civil Engineering, Thammasat School of Engineering, Thammasat University, Pathumthani, Thailand, ksurapar@engr.tu.ac.th

The publisher KeAi was established by Elsevier and China Science Publishing & Media Ltd to unfold quality research globally. In 2013, our focus shifted to open access publishing. We now proudly publish more than 200 world-class, open access, English language journals, spanning all scientific disciplines. Many of these are titles we publish in partnership with prestigious societies and academic institutions, such as the National Natural Science Foundation of China (NSFC).

 


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.