Damage Detection of Structures Considering Environmental Variations Based on the Deviation Patterns of Natural Frequencies

Abstract

The fundamental objective of structural damage detection is to ascertain the presence of damage in the structure under monitoring. Environmental variations (especially temperature) may also cause the aberration of dynamic properties. This could result in false alarms and pose a threat to the reliability of vibration-based damage detection. In this paper, a new approach toward damage detection under varying environmental conditions is investigated based on the deviation patterns of natural frequencies. The idea is derived from the fact that most damage occurs in a local area of a structure while environmental variations affect the whole structure. Natural frequencies deviate from their previous state in different patterns based on the causation of the deviation. The method uses simple mathematics to distinguish the two patterns using the first 10 natural frequencies of a structure. The proposed method was applied in two case studies including a numerical model with severe temperature variations simulated in ABAQUS and 1-year health monitoring data of a cable-stayed bridge in China as a real case study. By implementing the proposed method, the effects of environmental variations were eliminated in both case studies and damage was accurately detected before it was evident to the visual inspections of the bridge.