Damage Identification Based On Digital Image Correlation And Modal Strain Energy

In order to prevent civil engineering structures from being damaged by natural or man-made factors in the course of service,many identification methods have been proposed to detect the damage of existing structures,and then to repair and reinforce the structures as early as possible.Among them,the modal strain energy identification method is one of the most widely used indicators at present.It considers the influence of stiffness and vibration mode at the same time,and has certain advantages over the identification method using only one modal parameters.This paper derives the modal strain energy by Timoshenko beam theory,then introduces the damage localization mechanism and identification index of modal strain energy,and establishs the numerical model of a space steel truss to verify the damage location effect of the index.Additionally,research about the index is focused on the effect of damage degree,damage location and number of damage elements.The numerical analysis results show that different types of bars have different recognition effects,and the recognition effect of the middle bars is better than that of the end bars.When the number of damaged bars is small,the recognition results are more accurate;Due to the different degrees of damage,multiple damage cases is easily affected by the surrounding elements,and it is prone to misjudgment.Structural damage identification methods based on vibration information commonly use sensors to measure dynamic signals in practical applications,and it uses some mathematical processing means to construct the damage index to find the damage,locate the damage and estimate the damage.Although the measurement method has high sensitivity and accuracy,the number of measuring points is numerous when it is measured in a complicated structure,and it may not be easily arranged.The problem of sensor arrangement is a huge obstacle of current damage identification field.This thesis uses the Digital Image Correlation(DIC)method,which has the greatest advantage of being able to perform full field measurements.Taking the space steel frame model as experiment reseach object,Digital Image Correlation(DIC)method is used to measure modal parameters for calculating the modal strain energy as the damage identification index of three different damage conditions.Moreover,the identification results of using DIC method are compared with those of the acceleration sensor method.Finally,the effectiveness of this measurement method in damage identification is verified.Since the measured signal in the experiment is greatly disturbed by noise,this paper adopts the wavelet threshold de-noising technology to reasonably reduce noise of the acceleration and displacement signal,which achieves a satisfactory de-noising effect.The identification after de-nosing is more accurate than that before de-noising and the interference degree of surrounding elements is reduced.However,the damage of the two members of the space steel truss structure is complicated,and although the degree of interference is reduced after noise reduction,it is prone to misjudgment.In general,in engineering measurement,wavelet threshold noise reduction can be used as an effective signal processing tool for damage identification work to improve the accuracy of DIC measurement.