Model based damage detection of concrete bridge deck using Aadaptive neuro-fuzzy inference system

Damage detection of concrete bridge decks by measurement and monitoring variables related to vibration signatures such as acceleration signals is one of the main tasks of any bridge health monitoring system (bhms). generally damage puts some detectable/discoverable signs in the parameters of bridge vibration behavior. most of the existing methods of damage detection cannot be easily used in real-time diagnosis systems due to the need for some further calculations. these calculations are essential to find frequencies and mode shapes of bridge. in this paper an adaptive neuro-fuzzy inference system (anfis) is used to detect the possible damage location in a concrete bridge deck modeled by finite element method. a few damage scenarios are simulated for different locations of the deck and accelerations as representatives of response at some specific points are calculated. center of the deck is excited by an impact load. in the proposed anfis damage detection model, sampled (not complete) acceleration signals are used as inputs and the most possible region of the deck including damage is found. there is no need to calculate frequencies and mode shapes of the deck as vibration signatures. results show that this method can decrease the time and cost of visual inspections and be used as real-time damage detection caution system in practice.