A LSTM surrogate modelling approach for caisson foundations

OCEAN ENGINEERING

卷:204

Article

Zhang, Pin^1,2  Yin, Zhen-Yu¹  Zheng, Yuanyuan^3,4  Gao, Fu-Ping^5,6 

[1] Hong Kong Polytech Univ, Dept Civil & Environm Engn, Hung Hom, Kowloon, Hong Kong, Peoples R China

[2] Southern Marine Sci & Engn Guangdong Lab Guangzho, 1119 Haibin Rd, Guangzhou, Peoples R China

[3] Sun Yat Sen Univ, Sch Civil Engn, Guangzhou 510275, Peoples R China

[4] Southern Marine Sci & Engn Guangdong Lab Zhuahai, Zhuahai, Peoples R China

[5] Chinese Acad Sci, Inst Mech, Key Lab Mech Fluid Solid Coupling Syst, Beijing 100190, Peoples R China

[6] Univ Chinese Acad Sci, Sch Engn Sci, Beijing 100049, Peoples R China

关键词: Caisson foundation;    Failure envelope;    Smoothed particle hydrodynamics;    Long short-term memory;   

DOI  :  10.1016/j.oceaneng.2020.107263

来源: Elsevier

【 摘 要 】

This study proposes a hybrid surrogate modelling approach with the integration of deep learning algorithm long short-term memory (LSTM) to identify the mechanical responses of caisson foundations in marine soils. The LSTM based surrogate model is first trained based on limited results generated from the SPH-SIMSAND based numerical simulations with a strong validation, thereafter it is applied to predict the mechanical responses of soil-structure interaction and the failure envelope of unknown caisson foundations with various specifications as testing. The results indicate that the LSTM based model is more flexible than macro-element method, because it can directly learn the failure mechanism of caisson foundation from the raw data, meanwhile guarantees a high computational efficiency and accuracy in comparison with physical and numerical modelling. LSTM based surrogated model shows a great potential of application in engineering practice.

【 授权许可】

Free