Evaluation of Deep Learning against Conventional Limit Equilibrium Methods for Slope Stability Analysis
Evaluation of Deep Learning against Conventional Limit Equilibrium Methods for Slope Stability Analysis
复制标题
DOI：
10.3390/app11136060
复制
发表时间：
2021-07-01
影响因子：
2.7
通讯作者：
Liu, Zhen (Leo)
中科院分区：
文献类型：
--
作者：
Azmoon, Behnam;Biniyaz, Aynaz;Liu, Zhen (Leo)
本文通过控制良好的数据对比研究，评价了两种新的深度学习方法及其与传统方法的关系和差异。我们实现了四种被广泛接受的极限平衡分析方法，并将它们的实现和结果与新提出的深度学习方法进行了比较。这将为工程师提供一个关于深度学习与传统方法相比如何工作的清晰参考。通过本文，读者可以很容易地看到新方法的潜力和技术优势。这是一个很好的例子，展示了传统的基于物理的方法和数据驱动的方法之间的比较，并展示了数据驱动的方法如何改变或补充传统的工程实践。这项工作将有助于弥合传统地质系统工程分析与先进工程信息学之间的差距，并为许多类似的工程应用（例如，机械或稳定性分析）探索“大数据”解决方案。本文在人工智能和传统极限平衡分析方法的基础上，对深度学习作为一种新的边坡稳定性分析方法进行了比较研究。为此，开发了计算机程序，采用Bishop简化法、Fellenius法、Janbu简化法和Janbu修正法四种极限平衡法计算安全系数。该代码在RocScience中针对Slide2进行了验证。随后，使用平均FS值来近似斜率的“真实”FS，用于标记用于深度学习的图像。使用此代码，创建了具有广泛几何形状和土壤属性的斜坡图像的综合数据集。平均FS值用于标记图像，以实现两种深度学习模型：多类分类和回归模型。训练后，使用深度学习模型预测独立斜坡图像集的FS。最后，将模型的性能与传统方法进行了比较。本研究发现，与传统方法相比，深度学习方法的准确率高达99.71%，计算效率提高了18倍以上。
Featured Application This paper presents a comparison study with well-controlled data to evaluate two new deep learning methods and their relationships and differences with traditional methods. We implemented four widely accepted limit equilibrium analysis methods and compared their implementations and results with the newly proposed deep learning methods. This will lend engineers a clear reference regarding how deep learning works in comparison with traditional methods. With this paper, readers can easily see the potential and technical advantages of the new methods. This presents a good example to show the comparison between traditional physics-based approaches and the data-driven approaches and demonstrate how data-driven approaches can change or complement the traditional engineering practices. The work will help bridge the gap between traditional engineering analysis of geosystems and advanced engineering informatics and explore "big data" solutions for many similar engineering applications (e.g., with mechanical or stability analysis). This paper presents a comparison study between methods of deep learning as a new category of slope stability analysis, built upon the recent advances in artificial intelligence and conventional limit equilibrium analysis methods. For this purpose, computer code was developed to calculate the factor of safety (FS) using four limit equilibrium methods: Bishop's simplified method, the Fellenius method, Janbu's simplified method, and Janbu's corrected method. The code was verified against Slide2 in RocScience. Subsequently, the average FS values were used to approximate the "true" FS of the slopes for labeling the images for deep learning. Using this code, a comprehensive dataset of slope images with wide ranges of geometries and soil properties was created. The average FS values were used to label the images for implementing two deep learning models: a multiclass classification and a regression model. After training, the deep learning models were used to predict the FS of an independent set of slope images. Finally, the performance of the models was compared to that of the conventional methods. This study found that deep learning methods can reach accuracies as high as 99.71% while improving computational efficiency by more than 18 times compared with conventional methods.