Classification of Buildings' Potential for Seismic Damage by Means of Artificial Intelligence Techniques

• URL: http://arxiv.org/abs/2205.01076v1
• Date: Wed, 6 Apr 2022 15:32:15 GMT
• Title: Classification of Buildings' Potential for Seismic Damage by Means of Artificial Intelligence Techniques
• Authors: Konstantinos Kostinakis, Konstantinos Morfidis, Konstantinos Demertzis, Lazaros Iliadis
• Abstract summary: A large training dataset is used for the implementation of the classification algorithms. The level of the seismic damage is expressed in terms of the Maximum Interstory Drift Ratio. The most significant conclusion which is extracted is that the Machine Learning methods can be used to solve some of the most sophisticated real-world problems.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Developing a rapid, but also reliable and efficient, method for classifying the seismic damage potential of buildings constructed in countries with regions of high seismicity is always at the forefront of modern scientific research. Such a technique would be essential for estimating the pre-seismic vulnerability of the buildings, so that the authorities will be able to develop earthquake safety plans for seismic rehabilitation of the highly earthquake-susceptible structures. In the last decades, several researchers have proposed such procedures, some of which were adopted by seismic code guidelines. These procedures usually utilize methods based either on simple calculations or on the application of statistics theory. Recently, the increase of the computers' power has led to the development of modern statistical methods based on the adoption of Machine Learning algorithms. These methods have been shown to be useful for predicting seismic performance and classifying structural damage level by means of extracting patterns from data collected via various sources. A large training dataset is used for the implementation of the classification algorithms. To this end, 90 3D R/C buildings with three different masonry infills' distributions are analysed utilizing Nonlinear Time History Analysis method for 65 real seismic records. The level of the seismic damage is expressed in terms of the Maximum Interstory Drift Ratio. A large number of Machine Learning algorithms is utilized in order to estimate the buildings' damage response. The most significant conclusion which is extracted is that the Machine Learning methods that are mathematically well-established and their operations that are clearly interpretable step by step can be used to solve some of the most sophisticated real-world problems in consideration with high accuracy.

Related papers

• Estimating Earthquake Magnitude in Sentinel-1 Imagery via Ranking [5.71478837100808]
  We propose to pose the estimation of earthquake magnitudes as a metric-learning problem. We train models to estimate earthquake magnitude from Sentinel-1 satellite imagery and to additionally rank pairwise samples. Our experiments show at max a 30%+ improvement in MAE over prior regression-only based methods.
  arXiv  Detail & Related papers  (2024-07-25T15:35:44Z)
• Controllable seismic velocity synthesis using generative diffusion models [4.742123770879715]
  We propose conditional generative diffusion models for seismic velocity synthesis. This approach enables the generation of seismic velocities that closely match the expected target distribution. We demonstrate the flexibility and effectiveness of our method through training diffusion models on the OpenFWI dataset.
  arXiv  Detail & Related papers  (2024-02-09T09:41:26Z)
• Predicted Embedding Power Regression for Large-Scale Out-of-Distribution Detection [77.1596426383046]
  We develop a novel approach that calculates the probability of the predicted class label based on label distributions learned during the training process. Our method performs better than current state-of-the-art methods with only a negligible increase in compute cost.
  arXiv  Detail & Related papers  (2023-03-07T18:28:39Z)
• A Comprehensive Study on Robustness of Image Classification Models: Benchmarking and Rethinking [54.89987482509155]
  robustness of deep neural networks is usually lacking under adversarial examples, common corruptions, and distribution shifts. We establish a comprehensive benchmark robustness called textbfARES-Bench on the image classification task. By designing the training settings accordingly, we achieve the new state-of-the-art adversarial robustness.
  arXiv  Detail & Related papers  (2023-02-28T04:26:20Z)
• Classification of structural building damage grades from multi-temporal photogrammetric point clouds using a machine learning model trained on virtual laser scanning data [58.720142291102135]
  We present a novel approach to automatically assess multi-class building damage from real-world point clouds. We use a machine learning model trained on virtual laser scanning (VLS) data. The model yields high multi-target classification accuracies (overall accuracy: 92.0% - 95.1%)
  arXiv  Detail & Related papers  (2023-02-24T12:04:46Z)
• Improving robustness of jet tagging algorithms with adversarial training [56.79800815519762]
  We investigate the vulnerability of flavor tagging algorithms via application of adversarial attacks. We present an adversarial training strategy that mitigates the impact of such simulated attacks.
  arXiv  Detail & Related papers  (2022-03-25T19:57:19Z)
• A Comparative Evaluation of Machine Learning Algorithms for the Prediction of R/C Buildings' Seismic Damage [0.0]
  The present paper attempts an extensive evaluation of the capability of various Machine Learning algorithms to predict the seismic response of R/C buildings. A large-scale comparison study is utilized by the most efficient Machine Learning algorithms. The experimentation shows that the LightGBM approach produces training stability, high overall performance and a remarkable coefficient of determination.
  arXiv  Detail & Related papers  (2022-03-25T05:10:14Z)
• Interpretability in Convolutional Neural Networks for Building Damage Classification in Satellite Imagery [0.0]
  We use a dataset that includes labeled pre- and post-disaster satellite imagery to assess building damage on a per-building basis. We train multiple convolutional neural networks (CNNs) to assess building damage on a per-building basis. Our research seeks to computationally contribute to aiding in this ongoing and growing humanitarian crisis, heightened by anthropogenic climate change.
  arXiv  Detail & Related papers  (2022-01-24T16:55:56Z)
• Towards advancing the earthquake forecasting by machine learning of satellite data [22.87513332935679]
  We develop a novel machine learning method, namely Inverse Boosting Pruning Trees (IBPT), to issue short-term forecast based on the satellite data of 1,371 earthquakes of magnitude six or above. Our proposed method outperforms all the six selected baselines and shows a strong capability in improving the likelihood of earthquake forecasting across different earthquake databases.
  arXiv  Detail & Related papers  (2021-01-31T02:29:48Z)
• Bayesian Optimization with Machine Learning Algorithms Towards Anomaly Detection [66.05992706105224]
  In this paper, an effective anomaly detection framework is proposed utilizing Bayesian Optimization technique. The performance of the considered algorithms is evaluated using the ISCX 2012 dataset. Experimental results show the effectiveness of the proposed framework in term of accuracy rate, precision, low-false alarm rate, and recall.
  arXiv  Detail & Related papers  (2020-08-05T19:29:35Z)
• RescueNet: Joint Building Segmentation and Damage Assessment from Satellite Imagery [83.49145695899388]
  RescueNet is a unified model that can simultaneously segment buildings and assess the damage levels to individual buildings and can be trained end-to-end. RescueNet is tested on the large scale and diverse xBD dataset and achieves significantly better building segmentation and damage classification performance than previous methods.
  arXiv  Detail & Related papers  (2020-04-15T19:52:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.