Machine learning to improve the ground model for road settlement prediction
Machine learning to improve the ground model for road settlement prediction
The article discuss data processing and development of predictive ground model on the basis of data from extensive engineering geological survey. Basic model inputs are spatial coordinates (X, Y, elevation). Ground model with assigned soil groups, deformation properties and ground water level is created with input of spatial coordinates. Basic statistical and machine learning algorithms (regression, classification, clustering) were used in Python script to develop predictive ground model. Predictions are compared with the ground model developed by engineering geologist. ground model was used to calculate the settlement of the embankment subgrade in the highway route D1 for international and domestic transport.
M. Brcek; Jakub Panuska; Miroslav Kopecky
18th European Conference on Soil Mechanics and Geotechnical Engineering (ECSMGE2024)
A - New developments on structural design
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