A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Reliability Analysis of Reinforced Slope Combining Random Forest Algorithm and Meshless SPH Algorithm
Abstract This paper presents a framework combining a mesh-free method called Smoothed Particle Hydrodynamics (SPH) algorithm with Random Forest (RF) algorithm to evaluate the failure probability for reinforced slope. The reinforced slope model is established in Slope/w and Monte Carlo Simulation is implemented using Win-Batch to obtain the failure samples yielding reinforced slope failure with factor of safety < 1. The run-out distances corresponding to failure samples are modeled using SPH and these distances are based to build a new limit state function of reinforced slope failure. Then, RF algorithm is used to predict the state of the reinforced slope to enhance the computation efficiency. The proposed framework is illustrated through a roadside reinforced slope of Harbin to Jiamusi Railway to investigate the influence of retaining wall height on the failure probability of slope. The computation results show that RF algorithm exhibits a relative high accuracy in predicting the state of the reinforced slope according to its factors of safety, retaining wall heights and the corresponding run-out distances. It is found that as the retaining wall height increases, the failure probability of the reinforced slope decreases significantly. The proposed framework and the research outputs provide much insight into risk assessment and risk mitigation for reinforced slopes.
Reliability Analysis of Reinforced Slope Combining Random Forest Algorithm and Meshless SPH Algorithm
Abstract This paper presents a framework combining a mesh-free method called Smoothed Particle Hydrodynamics (SPH) algorithm with Random Forest (RF) algorithm to evaluate the failure probability for reinforced slope. The reinforced slope model is established in Slope/w and Monte Carlo Simulation is implemented using Win-Batch to obtain the failure samples yielding reinforced slope failure with factor of safety < 1. The run-out distances corresponding to failure samples are modeled using SPH and these distances are based to build a new limit state function of reinforced slope failure. Then, RF algorithm is used to predict the state of the reinforced slope to enhance the computation efficiency. The proposed framework is illustrated through a roadside reinforced slope of Harbin to Jiamusi Railway to investigate the influence of retaining wall height on the failure probability of slope. The computation results show that RF algorithm exhibits a relative high accuracy in predicting the state of the reinforced slope according to its factors of safety, retaining wall heights and the corresponding run-out distances. It is found that as the retaining wall height increases, the failure probability of the reinforced slope decreases significantly. The proposed framework and the research outputs provide much insight into risk assessment and risk mitigation for reinforced slopes.
Reliability Analysis of Reinforced Slope Combining Random Forest Algorithm and Meshless SPH Algorithm
Liu, Xu (author) / Li, Liang (author) / Wang, ShangShang (author) / Chen, Fu (author) / Zhai, Ming (author) / Yang, Zhengquan (author) / Gao, Yuan (author)
2021
Article (Journal)
Electronic Resource
English
BKL:
57.00$jBergbau: Allgemeines
/
38.58
Geomechanik
/
57.00
Bergbau: Allgemeines
/
56.20
Ingenieurgeologie, Bodenmechanik
/
38.58$jGeomechanik
/
56.20$jIngenieurgeologie$jBodenmechanik
A Kriging-based hybrid optimization algorithm for slope reliability analysis
| Online Contents | 2012
A Kriging-based hybrid optimization algorithm for slope reliability analysis
| British Library Online Contents | 2012
Reliability Analysis of Spatially Variable Soil Slope Using Deep Learning Algorithm
| British Library Conference Proceedings | 2023