Weld acoustic emission inspection of structural elements embedded in concrete

Abstract

After a catastrophic failure of the weld of the anchoring element of one cable in a stayed bridge, a non-destructive inspection was required to evaluate the weld condition of the 111 remaining anchoring elements to prevent future and similar failures. This examination was quite complicated since the anchoring elements are partially embedded in the reinforced concrete tower, and the weld is fully integrated into the concrete. Considering that direct access to the weld was not possible, acoustic emissions (AE) were a feasible alternative for these inspections. This study describes the inspection method, from laboratory tests simulating actual conditions for calibration to field tests for the method's tuning and evaluation. The AE inspection results are presented, and welds’ condition is classified according to the acoustic energy, measured through a severity index and graded from a zonal intensity plot. Two structural elements were selected for concrete demolition to expose the weld for penetrant and ultrasonic inspections to correlate measurements of the actual condition of the welds and their defect size. Because of the analysis, welds are identified for immediate repair and the rest for AE monitoring to evaluate defect evolution through the increase of the severity index.

An Experimental Study on Acoustic Emission of Broken Wires in Steel Strand

“…The noise rejection was done by simulating the noise signal and using the wavelet decomposition method, without specific in-depth characterization of the noise. Carrion [11] simulated the weld condition at the bridge cable joints in the laboratory and evaluated the classification of the weld condition according to the energy and assessed the degree of damage to the bridge ties by increasing the severity index of the acoustic emission signal. This paper focuses on the design of an acoustic emission signal acquisition test.…”

Preprocessing Acoustic Emission Signal of Broken Wires in Bridge Cables

“…In 2021, Ren [10] constructed a convolutional neural network for the classification and identification of wire broken signals by collecting AE signals from bridge cable pulling experiments in the laboratory and converting the AE signals into time-frequency image signals by wavelet transform. In 2022, Carrion [11] evaluated the classification of bridge cable damage by introducing the severity index of the AE signals.…”