Performance of artificial neural network and convolutional neural network on slope failure prediction using data from the random finite element method

Hsiao, Cheng-Hsi; Chen, Albert Y.; Ge, Louis; Yeh, Fu‐Hsuan

doi:10.1007/s11440-022-01520-w

Cited by 17 publications

(5 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The term 'deep' can refer to a structure of more than one hidden layer in NNs. A convolutional neural network (CNN) is a commonly used deep learning model, which is widely used in image and pattern recognition [63] in the computer field, and has been extended to other fields, such as foundation bearing capacity problems [15,64] and slope stability problems [65][66][67] in geotechnical engineering. Hidden layers in CNNs usually contain convolutional layers, pooling layers, and fully connected layers.…”

Section: Artificial Neural Network (Anns) Adaptive Neuro-fuzzy Infere...mentioning

confidence: 99%

“…Ahangari Nanehkaran et al [140] compared five different machine learning models for FOS predictions and verified the models using a confusion matrix and errors table to confirm the accuracy evaluation indexes. Hsiao et al [65] proposed a pre-trained model using an ANN and CNN to directly estimate the safety factor, the trace of the slope slip surface, and finally quickly predict the probability of failure. Jiang et al [141] trained a surrogate model with a gradient-boosting regression tree to predict the FOS under the effect of heavy rainfall.…”

Section: Prediction Of the Factor Of Safetymentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review

Xu,

He,

Shan

et al. 2023

Modelling

View full text Add to dashboard Cite

In traditional slope stability analysis, it is assumed that some “average” or appropriately “conservative” properties operate over the entire region of interest. This kind of deterministic conservative analysis often results in higher costs, and thus, a stochastic analysis considering uncertainty and spatial variability was developed to reduce costs. In the past few decades, machine learning has been greatly developed and extensively used in stochastic slope stability analysis, particularly used as surrogate models to improve computational efficiency. To better summarize the current application of machine learning and future research, this paper reviews 159 studies of supervised learning published in the past 20 years. The achievements of machine learning methods are summarized from two aspects—safety factor prediction and slope stability classification. Four potential research challenges and suggestions are also given.

show abstract

Section: Artificial Neural Network (Anns) Adaptive Neuro-fuzzy Infere...mentioning

confidence: 99%

Section: Prediction Of the Factor Of Safetymentioning

confidence: 99%

Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review

Xu,

He,

Shan

et al. 2023

Modelling

View full text Add to dashboard Cite

show abstract

“…To address these challenges, Ouyang et al [59] proposed a flexible solution for the analysis of laterally loaded piles, leveraging the power of machine learning. Machine learning, an emerging artificial intelligence technology, offers distinct advantages in intensive computation and universal approximation capabilities (He et al [37], Huynh et al [42], Hsiao et al [40], Zhang et al [83], Zhang et al [84], Nguen et al [57], [58]). It commonly utilizes neural networks, which are mathematical models with layered structures comprising linear transformations and nonlinear activation functions.…”

Section: Introductionmentioning

confidence: 99%

Analyzing Laterally Loaded Piles in Multi-layered Cohesive Soils: a Hybrid Bnwf Approach With Validation and Parametric Study

Gendy

2024

Preprint

View full text Add to dashboard Cite

Pile foundations frequently encounter lateral loads originating from various hazards. These types of foundations are commonly utilized in structures like bridges, retaining walls, and high-rise buildings. Analyzing laterally loaded piles presents a complex geotechnical problem that entails considering multiple interrelated design factors. It requires accounting for structural bending behavior, soil-structure interaction, soil nonlinearity, and optimizing for cost-effectiveness. In this paper, the commonly used approach beam on nonlinear Winkler foundation is developed. This methodology involves representing the pile using one-dimensional finite elements in the vertical direction, incorporating nonlinear bending stiffness. Additionally, soil deformation is determined using empirically derived P-y curves, which are obtained from full-scale field tests. By combining the pile stiffness with the soil stiffness considering the full interaction between the pile and the surrounding soil, the complete stiffness matrix of the single pile is formed, leading to a reduction in the number of equations that need to be solved. Both Euler and Timoshenko beams are considered, and the analysis is conducted using both finite elements and finite difference methods. The proposed hybrid approach is validated by comparing its results from analyzing laterally loaded piles in multi-layered soil profiles with those obtained from different models in existing literature and available field measurements. The well-known software ELPLA is equipped with the proposed hybrid technique. Furthermore, a parametric study investigates the behavior of laterally loaded pipe piles in soft and stiff clay, culminating in the presentation of dimensionless curves from this study.

show abstract

“…In order to help audiences to understand the following contents, the basic method of neural network (NN) 9,10…”

Section: Introductionmentioning

confidence: 99%

Analysis of medical insurance in China by neural network

Xiang

2023

International Conference on Computer Application and Information Security (ICCAIS 2022)

View full text Add to dashboard Cite

Medical insurance plays role in the wellbeing of population. However, the difficulty with the amount addition of participants increases dynamically. Especially, the increase of organizations including in this domain makes optimization of medical insurance complex. Thus, it is necessary to find the rule based on the information provided by the related to data as supplementary to support decision making. In order to finish this task, Neural Network, proved its robustness in data analysis, is included in the proposed data management frame. Input and cost as the most important attributes of medical insurance are set as target features. Then, the relationship between these two attributes must be considered. A skipping window is defined to adjust the proportion of the target features in training and testing stages. NN network with double direction of time window for medical insurance is given to foreseeing target attributes. Based on the simulation result, the largest average accuracy is 0.8333.

show abstract

Performance of artificial neural network and convolutional neural network on slope failure prediction using data from the random finite element method

Cited by 17 publications

References 40 publications

Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review

Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review

Analyzing Laterally Loaded Piles in Multi-layered Cohesive Soils: a Hybrid Bnwf Approach With Validation and Parametric Study

Analysis of medical insurance in China by neural network

Contact Info

Product

Resources

About