# A Review of Vibration-Based Damage Detection in Civil Structures: From Traditional Methods to Machine Learning and Deep Learning Applications

**Authors:** Onur Avci, Osama Abdeljaber, Serkan Kiranyaz, Mohammed Hussein, Moncef Gabbouj, Daniel J. Inman  
**arXiv:** [2004.04373](https://arxiv.org/abs/2004.04373)  
**Date:** 2020-04-09  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2004.04373v1/

## Key points

- Monitoring structural damage is crucial for maintaining the service life and safety of civil structures
- Machine Learning (ML) and Deep Learning (DL) algorithms have emerged as feasible and accurate tools for vibration-based structural damage detection
- This paper aims to bridge the gap between traditional methods and ML/DL approaches in vibration-based structural damage detection
- Continuous monitoring is important to detect anomalies or potential damage caused by environmental or human-induced factors
- Vibration-based methods utilize the response of a structure's vibrations to assess its condition and identify structural damage
- ML and DL algorithms have shown elegant performance and rigorous accuracy in this field
- The authors provide an overview of traditional methods before discussing recent applications of ML/DL algorithms for vibration-based structural damage detection
- These advanced techniques can enhance monitoring capabilities, improve early warning systems, and contribute to sustaining the longevity and safety of civil structures

## Summary

Monitoring structural damage is crucial for maintaining the service life and safety of civil structures. Traditional methods have been used to assess the health, integrity, and performance of structures by monitoring the occurrence and propagation of damage. However, with advancements in computing power and sensing technology, Machine Learning (ML) and Deep Learning (DL) algorithms have emerged as feasible and accurate tools for vibration-based structural damage detection. This paper aims to bridge the gap between traditional methods and ML/DL approaches by providing a comprehensive review of recent applications in vibration-based structural damage detection. The authors highlight the importance of continuous monitoring to detect anomalies or potential damage caused by environmental or human-induced factors. Vibration-based methods utilize the response of a structure's vibrations to assess its condition and identify structural damage. ML and DL algorithms have shown elegant performance and rigorous accuracy in this field. Despite multiple review studies on vibration-based structural damage detection, there has not been a study that specifically describes and discusses the transition from traditional methods to ML/DL techniques. The authors address this gap by presenting an overview of traditional methods before delving into the most recent applications of ML/DL algorithms for vibration-based structural damage detection in civil structures. This comprehensive review provides valuable insights into how these advanced techniques can enhance monitoring capabilities, improve early warning systems, and contribute to sustaining the longevity and safety of civil structures. Overall, this paper contributes to the existing body of knowledge by highlighting the potential benefits of incorporating ML/DL algorithms into vibration-based structural damage detection practices. By understanding these advancements, researchers and practitioners can make informed decisions regarding monitoring strategies for civil structures.