Insights on drop mass systems to predict pile compressive resistance from dynamic load test energy measurements
Insights on drop mass systems to predict pile compressive resistance from dynamic load test energy measurements
We are currently immersed in the era of big data, where large volumes of data is generated and stored daily. However, the true challenge lies in transforming this data into meaningful and actionable information. To extract insights from these vast datasets, there is a growing dependence on machine learning techniques, which essentially build upon the foundation of traditional statistical methods. These techniques enable the creation of models that enhance our understanding of diverse subjects and facilitate informed decision-making. This study focuses on the establishment and exploration of a database derived from dynamic load tests on piles (DLT). In DLT on piles, impact hammers are employed, characterized by their potential energy or kinetic energy just before impact. Testing on high-capacity drilled deep foundations presents several challenges, including the need for sufficient energy to mobilize compressive static resistance of the pile. The investigation delves into understanding the correlation among various DLT test variables and uncovering potential relationships using supervised models, such as linear and non-linear regressions. The findings from this exploration have unveiled crucial insights, such as the influence of diameter on resistance and the existence of a non-linear relationship between resistance and the maximum energy transferred to the pile.
Jaime Santos; Jeniffer Viegas
18th European Conference on Soil Mechanics and Geotechnical Engineering (ECSMGE2024)
F - Future city world vision
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom