Advances in Civil Engineering, Год журнала: 2024, Номер 2024(1)

Опубликована: Янв. 1, 2024

Site investigation is crucial in geotechnical engineering. The cone penetration test (CPT) and the multichannel analysis of surface waves (MASWs) are widely used as geophysical methods, respectively. CPT offers high precision but requires a cost only provides soil information at limited locations. In contrast, MASW covers broad range has less accuracy compared to CPT. This study proposes novel ensemble prediction method that fuses both data overcome limitations using either dataset alone. employs random forest (RF) gradient boosting decision tree (GBDT) achieve transformation between shear velocity tip resistance ( V s – q c ) unknown Unlike traditional empirical regression models, this more accurate reliable predictions by leveraging complementary strengths MASW. proposed RF‐GBDT model validated from New Zealand Geotechnical Database. results show established outperforms simple models various popular machine learning predicting Specifically, integrating increases R 2 location CPT3 0.477 0.758, demonstrating can improve parameters areas with sparse data.

Язык: Английский