نوع مقاله : پژوهشی
نویسندگان
1 دانشکده عمران، دانشگاه صنعتی خواجه نصیرالدین طوسی
2 استادیار دانشکده مهندسی عمران/دانشگاه صنعتی خواجه نصیر الدین طوسی
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Damage detection methods are integral components of structural health monitoring systems. Identifying damage in structures using vibration-based methods has always been one of the most important and popular topics among researchers in structural health monitoring. Vibration-based damage identification includes extracting a feature that can be used to measure the minuscule changes caused by damage to the structure. In recent years, advances have been made in using chaotic excitation and representing damage-sensitive features based on the properties of the chaotic attractor. These types of damage-sensitive features try to measure the minuscule changes caused by structural damage by comparing the chaotic attractors obtained from the structural response. The high sensitivity of chaotic systems to small changes makes attractor-based features suitable for identifying structural damage. One of the most widely used attractor-based features is the Generalized Interdependence, which has a reasonable sensitivity to damage and relatively low computational complexity. Also, the comparative nature of this feature can help identify damage in the presence of environmental variables such as noise. However, this feature has limitations that make its use exclusive to particular instances. e.g., in structures where the exact location of the damage is known beforehand. In the damage identification method presented in this research, improvements like adding a damage sensitivity factor and applying controls over the operation, have been made to this feature to remove these limitations while preserving its exceptional properties in detecting damage in structures. In the structure examined in this research, where the generalized interdependence feature does not show the slightest decrease in dependence due to damage, the improved feature detects damage by showing about 20% better performance in finding a reduction in the dependence between two points of the structure. Two points of the structure are selected to be located at different distances from the damage. In other words, the improved feature can measure the different impacts due to damage on these two points.
کلیدواژهها [English]
)