Determination of yielding point by means a probabilistic method on acoustic emission signals for application to health monitoring of reinforced concrete structures

Abstract

Summary Reinforced concrete (RC) flanged beam specimens were tested under incremental cyclic load till failure in flexure, and simultaneously, the acoustic emission (AE) signals released by the specimens were recorded. To assess damage in RC structures, a previously published index of damage (ID) based on AE signals was used. This index, however, needs to know the yielding point of the specimen. In the present study, yielding point was identified with a probabilistic method known as Gaussian mixture modeling (GMM) applied to the AE signals, as compared with that obtained by means of the plastic strain energy. It was observed that yielding load obtained with both methodologies was almost same, thus validating the GMM method. This result permits to use the ID index for damage monitoring of RC structure in practical scenarios, by using only information hidden in the AE signals. The influence of loading rate, failure type (tensile and shear), RC beam depth, concrete compressive strength, and percentage of tensile steel reinforcement on ID were studied in this work.

Smart monitoring of strengthened beams made of ultrahigh performance concrete using integrated and nonintegrated acoustic emission approach

“…There are several applications of AE in concrete structures such as damage identification, damage monitoring, damage classification, etc 16 . Some of the recent studies such as damage detection of concrete beams by integrating AE and Weibull damage function, 17 AE characterization of fatigue crack growth, 18 determination of yielding point, 19 damage characterization of crumb rubber concrete, 20 analysis of the bond behavior of corroded reinforcement, 21 and fracture monitoring in carbon nanotube‐crumb rubber mortar highlight successful implementation of AE testing for failure characterization in concrete structures. Application of machine learning‐based algorithms 22,23 such as support vector machine and Gaussian mixture modeling are also studied by various researchers to identify local fracture mechanism, crack classification, and damage identification in reinforced concrete structures.…”

Damage and fracture law of coal samples with different joint angles and their characterisation by acoustic emission

“…They proposed the damage index ( D ), defined as D = N / N m , where N is the accumulated AE count released at any time during the damage monitoring of the test specimen and N m is the AE count accumulation at the time of yielding. However, these constitutive models can neither provide a calculation method for the AE count number of the yield point 28 nor characterise the instantaneous failure of the material. Nonetheless, their existing results lay a foundation for studying the influence of joints and other heterogeneities on damage evolution.…”

Structural health monitoring of building rock based on stress drop and acoustic‐electric energy release

“…AE and EMR are elastic waves and electromagnetic waves generated by the destruction of rock materials. They are closely related to the properties of rock and the deformation and failure conditions [9][10][11][12][13] and have been widely used in the monitoring and prediction of the stability of rock mass structures. Therefore, it is particularly important to study the energy release law during the stress drop generation process of rock failure and the transformation mechanism between stress drop and AE and EMR energy for the health monitoring of rock material stability.…”