Due to the high sensitivity of coda waves to the smallest structural alterations such as strain, humidity or temperature changes, ultrasonic waves are a valid means to examine entire structures employing networks of ultrasonic transducers. In order to substantiate this ex ante assessment, the viability of measuring ultrasonic waves as a valid point of reference and inference for structural changes is to be further scrutinized in this work. In order to investigate the influence of mechanical strain on ultrasonic signals, a four-point bending test was carried out on a reinforced concrete beam at Ruhr University Bochum. Thus, measurements collected from a network of selected transducer pairings arranged across the central, shear-free segment of the test specimen, were correlated to their respective strain fields. Detected ultrasonic signals were evaluated employing Coda Wave Interferometry. Such analysis comprised the initial non-cracked state as well as later stages with incremental crack depth and quantity. It was to ascertain that the test specimen can in fact be qualitatively compartmentalized into areas of compression and tension identified via Relative Velocity Changes presented in Attribute Maps. However, since results did not entail a zero crossing, i.e., neither positive nor negative values were to be calculated, only relative changes in this work displayed staggered over the height of the object under test, are discussed. Under the given methodological premises, additional information is currently required to make quantitative assertions regarding this correlation of ultrasonic and strain results. This holds true for the comparability of the ultrasonic and strain results for both non-cracked and even the cracked state.

No abstract

The integral collection of information such as strains, cracks, or temperatures by ultrasound offers the best prerequisites to monitor structures during their lifetime. In this paper, a novel approach is proposed which uses the collected information in the coda of ultrasonic signals to infer the condition of a structure. This approach is derived from component tests on a reinforced concrete beam subjected to four-point bending in the lab at Ruhr University Bochum. In addition to ultrasonic measurements, strain of the reinforcement is measured with fiber optic sensors. Approached by the methods of moment-curvature relations, the steel strains serve as a reference for velocity changes of the coda waves. In particular, a correlation between the relative velocity change and the average steel strain in the reinforcement is derived that covers 90% of the total bearing capacity. The purely empirical model yields a linear function with a high level of accuracy (R2 = 0.99, RMSE≈90μstrain).

Ultrasonic Coda Wave interferometry has the potential to detect minute changes in scattering materials like concrete. By permanently installing ultrasonic transducers in concrete, DFG Research unit CoDA aims to develop methods for concrete damage assessment in Germany's ageing infrastructure. To test the methods developed in simulations and laboratory experiments on a large scale, we have implemented several ultrasonic transducers, both embedded and externally attached at the Gänstorbrücke Ulm, one of Germany's most monitored road bridges. Since fall 2020 we are monitoring parts of the centre of the Bridge, as well as an abutment, and compare the results to the commercial monitoring system. All data is recorded with a self-made data collection device, the so-called W-Box, and analyzed with different coda wave-based algorithms to detect signal and volumetric velocity changes for several transducer combinations. The long-term measurements show that the influence of temperature changes on strains and therefore ultrasound velocity changes calculated with coda waves can be monitored. Maps of velocity change show that parts of the bridge react differently to environmental heating or cooling which indicates material property differences comparing those areas. The capabilities and limitations of the coda wave-based monitoring system are tested in a controlled experiment. Static loading using a truck with varying load at several positions allows the calibration of the system to improve the detectability of possibly damaging loads and changes induced by this loading versus the influence of the environment.

No abstract