A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Using Novel Optimization Algorithms with Support Vector Regression to Estimate Pile Settlement Rates
To ensure structure safety, such as bridge structures, artificial intelligence ways are the most helpful way to predict failure factors such as pile settlement estimation. Various methods are taken into account for evaluating the movement of the piles, which helps to realize the future view of the project during the loading period. The most intelligent mathematical strategy to calculate the movement of the pile is applied. In this regard, support vector regression (SVR), a machine learning method, was used in this study, accompanying two optimizers to accurately determine the key SVR variables. The marine predator algorithm (MPA) and the grasshopper optimization algorithm (GOA) were combined with SVR to create the SVR-MPA and SVR-GOA frameworks. Additionally, several metrics were used to evaluate the model’s overall performance. The R2 for SVR-MPA in the training phase was found to be 0.997, showing a desirable modelling operation. At the same time, the RMSE was calculated 0.2843 mm and compared to the SVR-GOA, the differences are 1.16 and 106.38%, respectively, in favour of the former model. The comprehensive index of OBJ, including RMSE, MAE, and R2, was calculated at 0.2791 and 0.581 mm for models optimized by MPA and GOA, alternatively.
Using Novel Optimization Algorithms with Support Vector Regression to Estimate Pile Settlement Rates
To ensure structure safety, such as bridge structures, artificial intelligence ways are the most helpful way to predict failure factors such as pile settlement estimation. Various methods are taken into account for evaluating the movement of the piles, which helps to realize the future view of the project during the loading period. The most intelligent mathematical strategy to calculate the movement of the pile is applied. In this regard, support vector regression (SVR), a machine learning method, was used in this study, accompanying two optimizers to accurately determine the key SVR variables. The marine predator algorithm (MPA) and the grasshopper optimization algorithm (GOA) were combined with SVR to create the SVR-MPA and SVR-GOA frameworks. Additionally, several metrics were used to evaluate the model’s overall performance. The R2 for SVR-MPA in the training phase was found to be 0.997, showing a desirable modelling operation. At the same time, the RMSE was calculated 0.2843 mm and compared to the SVR-GOA, the differences are 1.16 and 106.38%, respectively, in favour of the former model. The comprehensive index of OBJ, including RMSE, MAE, and R2, was calculated at 0.2791 and 0.581 mm for models optimized by MPA and GOA, alternatively.
Using Novel Optimization Algorithms with Support Vector Regression to Estimate Pile Settlement Rates
Indian Geotech J
Sun, Lu (author) / Li, Tinghui (author)
Indian Geotechnical Journal ; 55 ; 79-91
2025-02-01
13 pages
Article (Journal)
Electronic Resource
English
Using Novel Optimization Algorithms with Support Vector Regression to Estimate Pile Settlement Rates
| Springer Verlag | 2025
Support Vector Machine to Predict the Pile Settlement using Novel Optimization Algorithm
| Online Contents | 2023
Design optimization using support vector regression
| British Library Online Contents | 2008