Feasibility of an evolutionary artificial intelligence (AI) scheme for modelling of load settlement response of concrete piles embedded in cohesionless soil

• Published in: Ships and offshore structures Vol. 13; no. 7; pp. 705 - 718
• Main Authors: Jebur, Ameer A., Atherton, William, Al Khaddar, Rafid M.
• Format: Journal Article
• Language: English
• Published: Cambridge Taylor & Francis 03.10.2018; Taylor & Francis Ltd
• Subjects: Artificial intelligence; Aspect ratio; Bearing strength; Carrying capacity; Concrete; Concrete piles; Evolutionary algorithms; Feasibility studies; Levenberg-Marquardt MATLAB algorithm; Load carrying capacity; Mathematical models; Modelling; Pile friction; Pile load tests; pile load-settlement; Pile settlement; pile slenderness' ratio; Piles; Ratios; Rigidity; sand relative density; sandy soil; Sandy soils; Soil; Soil investigations; Statistical analysis; Statistical methods; Tests
• ISSN: 1744-5302; 1754-212X
• DOI: 10.1080/17445302.2018.1447746

This investigation aimed to examine the load carrying capacity of piles embedded in sandy soil of various densities, and to develop a predictive model to determine pile settlement using a novel artificial intelligence (AI) method. Experimental pile load tests were conducted using three concrete piles, with aspect ratios of 12, 17 and 25. Evolutionary Levenberg-Marquardt MATLAB algorithms, enhanced by T-tests and F-tests, were used in this process. According to the statistical analysis and the relative importance study, pile length, applied load, pile flexural rigidity, pile aspect ratio and sand-pile friction angle were found to play a key role in pile settlement. Results revealed that the proposed optimum model algorithm precisely characterized pile settlement. There was close agreement between the experimental and predicted data (Pearson's R = 0.988, P = 6.28 × 10 -31 ) with a relatively insignificant root mean square error of 0.002.