Journal of Engineering Mechanics

Propagating solid elastic waves forms the basis of multiple nondestructive evaluation (NDE) and structural health monitoring techniques. The majority require direct sensor contact with the material or structure being tested. Conventional NDE methods such as ultrasonic testing have rigorous coupling requirements and limited ranges for damage or defect detection. These requirements and limitations make inspection of large areas or complex geometries slow and cumbersome or altogether impractical. Removing the requirement for contact between the sensor and structure could greatly relax the geometric and size limitations of current elastic wave–based NDE techniques. Laser-based excitation and detection of solid elastic waves are demonstrated capabilities; however, existing laser-based excitation techniques lack the precision and flexibility required by NDE methods that target specific wave modes or frequencies. This proof-of-concept study investigates using a pulsing laser for frequency controlled excitation of elastic waves. Low-frequency experiments demonstrate the principle of matching the excitation laser-pulse frequency to resonances of the tested component. An analytical thermoelastic analysis confirms that low-frequency results extend to higher-frequency ranges commonly used for NDE. The pairing of frequency-tuned, noncontacting laser-based excitation with existing noncontacting laser-based detection equipment could offer new possibilities for NDE of large areas and complex geometries.

The traditional experimental method to determine the transfer length in pretensioned concrete members consists of measuring concrete surface strains before and after detensioning with a mechanical strain gauge. This method is prone to significant human error and inaccuracies. In addition, because it is a time-consuming and tedious process, transfer lengths are seldom if ever measured on a production basis for product quality assurance. A rapid noncontact method for determining transfer lengths in pretensioned concrete railroad ties has been developed. The new method uses laser-speckle patterns that are generated and digitally recorded at various points along the pretensioned concrete member. A prototype was fabricated as a portable self-contained unit for field testing, which incorporates a unique modular design concept that has several preferable features. These include flexible adjustment of the gauge length, easy upgradability to automatic operation, robustness, and higher accuracy. A laser-speckle strain sensor was applied to the transfer length measurements of typical pretensioned concrete railroad ties in a railroad tie plant. These prestressed concrete tie members are expected to withstand repeated axle loadings of 290 kN, totaling 250 million gross tons annually and occurring at speeds in excess of 110 km/h. The technique achieved a microstrain resolution comparable to what could be obtained using mechanical gauge technology. Surface strain distributions were measured on both ends of 12 ties, and their associated transfer lengths were subsequently extracted. The measurements of the transfer length using the laser-speckle strain sensor were unprecedented because it was the first time that the laser-speckle technique had been applied to pretensioned concrete inspection, and particularly for use in transfer length measurements of concrete railroad ties. It was also demonstrated that the technique was able to withstand the harsh manufacturing environment, making transfer length measurements possible on a production basis for the first time.

A road is considered safe when it provides adequate friction between the tire and pavement surface interaction to avoid accidents. The current manual methods of evaluating surface friction of roads are not only dangerous for the inspectors and motorists on the road but are very time-consuming and subject to the inspector’s judgment. This paper confirms the possibility of monitoring pavement macrotexture from acoustic measurements underneath the body of a moving vehicle. Currently, macrotexture is quantified by a mean texture depth (MTD) index. In the present work, MTD is estimated from the sound generated by the tire-pavement interaction in a moving vehicle. To establish this approach, experiments were performed at the National Center for Asphalt Technology (NCAT), Auburn, Alabama. Signal processing techniques were used to identify a frequency band that led to an accurate correlation between integrated acoustic pressure and MTD collected by NCAT. This frequency band depends on the speed of the vehicle. The effect of microphone locations and the speed of the vehicle on macrotexture monitoring are studied, and conclusions are presented. Microphones that are close and oriented toward the contact patch of the rear tire have the highest correlation to MTD. It is observed that raising the microphones by 33 cm decreased the correlation by ∼13%. It was shown that higher vehicle speed has higher correlation values. Accurate estimates of MTD were obtained for road surfaces having MTD values in the range of 0.5–2.5 mm, with vehicle speeds in the range of 32–80 km/h (20–50 mi/h).

This paper describes the development of new air-coupled ultrawideband ground-penetrating radar (GPR) for highway pavement and bridge deck inspections that can achieve high spatial resolution and high inspection performance while operating on vehicles driving at regular highway speeds. The major design features include dual-frequency band operation, 8 gigasamples per second high-speed real-time data acquisition, high-speed large volume data transmission and storage, and customized signal-processing algorithms for GPR image enhancement and feature extractions. For design validations, steel reinforcing bar detection experiments in various setups are conducted. The measurement results show good agreement with the subsurface feature configurations.

This paper presents the results of studies assessing waveguided acoustic emission techniques for remotely sensing the condition of structures operating in extreme environments. A particularly acute challenge to sensing in extreme conditions is that the sensor and associated electronics are not sufficiently robust to withstand the structural test conditions. Instead, it is necessary to use signal transduction methods to carry the signal from the structure to a remote and more hospitable sensor location. In the studies reported here an inductively coupled plasma torch creates a simulated environment for the testing of hypervelocity vehicle heat shield materials. Stresses within the material generate acoustic emissions (AEs) that propagate as elastic waves. Waveguides coupled to the sample material and AE sensors allow sample material tests to be monitored remotely. The data were analyzed using multiple statistical methods. The results show that various testing conditions produce repeatable differences in the AE data set, providing evidence that AE testing is sensitive to changes in the thermal degradation of the sample material. AE testing currently offers the potential for real-time in situ monitoring of thermal degradation in laboratory testing conditions and represents progress toward a deployable diagnostic system for aerospace applications.

This study is intended to facilitate damage detection in concrete bridge girders without the need for visual inspection while minimizing field measurements. Beams with different material and cracking parameters were modeled using ABAQUS finite-element analysis software to obtain stiffness values at specified nodes. The resulting database was then used to train an artificial neural network (ANN) model to inversely predict the most probable cracking pattern. The aim is to use the ANN approach to solve an inverse problem where a unique analytical solution is not attainable. Accordingly, simple span beams with three, five, seven, and nine stiffness nodes and a single crack were modeled in this work. To confirm that the ANN approach can characterize the logic within the databases, networks with geometric, material, and cracking parameters as inputs and stiffness values as outputs were created. These networks provided excellent prediction accuracy measures ($R^2 \text{values}>\mathrm{99\%}$). For the inverse problem, the noted trend shows that better prediction accuracy measures are achieved when more stiffness nodes are used in the ANN modeling process. It was also observed that providing some outputs to the ANN as inputs, thus decreasing the number of required outputs, immensely improves the quality of predictions provided by the ANN. An experimental verification program will be conducted to qualify the effectiveness of the method proposed. This test program is described in details in the present paper.

The field application of three different nondestructive tests (NDTs)—air-coupled impact echo (IE), infrared (IR) thermography, and sounding (chain drag)—are evaluated in this paper, where an actual in-service concrete bridge deck is tested. Two different contactless IE test equipment sets are deployed as part of an effort to develop new rapid measurement methods. The IE data are presented as two-dimensional frequency maps, and the IR data are presented as temperature maps over the tested area. Sounding (chain-drag) result maps are also presented. For verification of the location of near-surface delamination damage, eight drilled core samples were extracted from the test area. The results obtained from each of the individual NDT methods show reasonably good agreement with the drilled cores in terms of locating near-surface delamination. Finally, the NDT methods are compared across general performance criteria, considering accuracy, testing practicality, and costs. The analysis shows that all of the evaluated NDT methods are comparable, and the chain-drag method is not more accurate and reliable for detection of shallow delamination in the deck.

The deteriorating national infrastructure demands improved nondestructive evaluation (NDE) and structural health monitoring methods for existing concrete structures. Vibration resonance tests offer an efficient NDE method to identify and characterize shallow (near-surface) delamination defects that afflict RC structures. However, efficient implementation of effective modal analysis methods for this purpose is hindered by practical testing limitations. This paper studies vibration resonance data from square, rectangular, and circular near-surface delamination defects in concrete using two testing configurations: a coupled source-receiver set configuration (driving point type), which is analogous to the impact-echo (IE) test, and a fixed-source moving sensor configuration, which represents a conventional modal analysis test. All data were collected using contactless air-coupled sensors, which enable efficient data collection from large structures from prepared laboratory samples. An approach to self-normalize each IE data set using surface wave pulse information in the time signal is proposed. The self-normalized multipoint IE data are presented in the form of modal images, where overlap of the first few natural mode shapes accurately predicts the areal size of underlying delamination defects. The mode shapes obtained with the corrected multipoint IE testing configuration, which provides a significant advantage for the application in NDE tests because damage location, size, and shape need not be known in advance, compare favorably with those from conventional modal testing.

To understand and model the effects of natural hurricane winds on residential buildings, full-scale field testing and analysis are needed. Toward this goal, a group of researchers at Florida Tech, in collaboration with the Florida Coastal Monitoring Project team, have deployed sensors in a real storm scenario and collected roof wind pressure and local wind data. Because the behavior of a hurricane is subject to vagaries of nature, the current research effort is oriented to develop a rational basis for analyzing the data. To investigate this, the variable time averaging is applied to the data collected by the Florida Coastal Monitoring Project from three storms: Hurricane Gordon, Hurricane Isidore, and Hurricane Rita. The wind data correspond to a 10-m-high tower in different exposures at a 10-Hz sampling rate. In the wind load specifications worldwide, a wide range of averaging times have been used, 3–60 s to define gusts and 10–60 min to define mean speeds. This is seemingly more heuristic for the local hurricane wind field for which ergodicity and stationarity are not well established. In this study, variable averaging times from 1 to 1,000 s were applied to the wind speed time series data. Then three-dimensional statistical maps of the mean, RMS, skewness, and kurtosis distributions are obtained as functions of the averaging period. The effect of the averaging time on the mean is evidenced by a large variance for short averaging periods and a small variance for long averaging periods for all three hurricanes. The variation of the mean reaches a minimum at approximately 300 s. Therefore, all the higher-order moments are also calculated with respect to the 300-s mean. The skewness values are different (nonzero) for different storms, indicating a non-Gaussian behavior for the fluctuations, although the kurtosis values range mostly from 2.5 to 3.5.

The wind loads including forces and moments on bridge decks are evaluated in the current study using the velocity fields and their derivatives. Based on the conservation of momentum and momentum moment, the integral over the control volume is converted to the surface integral over the boundary of the control volume using the divergence theorem. The wind forces and moments can be calculated based on the obtained velocity fields and their derivatives around bridge decks, and the pressure terms in the equations are canceled out. Velocity fields are obtained from computational fluid dynamics (CFD) simulations and particle image velocimetry (PIV) wind tunnel experiments. Results from different turbulence models are compared with each other and wind tunnel experimental results in the literature. The calculated wind loads from numerical integrations of velocity fields from PIV experiments or CFD numerical simulations are compared with the results from a direct integration of the CFD-simulated surface pressure and the results from force-balance measurements in the wind tunnel. The results from several cases with different sizes of control volumes and space resolutions for both the square cylinder section and the twin-box girder section are discussed.

For a slender long-span bridge, the global aeroelastic and aerodynamic phenomena, such as flutter stability and buffeting response, induced by wind turbulence significantly affect the safety and lifetime performance of the bridge. To investigate these phenomena, a wind tunnel experimental technique with bridge section models is typically required to identify some critical aerodynamic coefficients, such as flutter derivatives, which are dependent on the specific profile (shape) of a bridge cross section and the functions of reduced frequencies. Obviously, this practice is based on the assumption that the shape of the cross section of a long-span bridge does not vary over time. Such an assumption may not remain valid when the stochastic traffic on the bridge is considered, because of the simple fact that the presence of vehicles changes the profile of the bridge cross section. The current study aims to provide some insights through experimental assessment of traffic impacts on flutter derivatives of the modified bridge cross section because of the presence of traffic. A bridge section model with scaled vehicle models distributed on the bridge deck is tested in the wind tunnel following the simulated stochastic traffic flow. Several scenarios are tested to provide insights on the variations of the flutter derivatives over different sections along the bridge and at different time instants for the same location. In addition, some extreme situations, such as under evacuation or serious congestion, are also studied.

Flexible bridges could experience vortex-induced and buffeting excitations below their design wind speeds and divergent self-excited oscillations (flutter) beyond a critical wind speed. For regular straight-line winds that are stationary or weakly stationary, a frequency domain formulation is routinely used for flutter analysis of bridges. However, when evaluating performance of flexible bridges subjected to gusty winds that are transient in nature, the frequency-domain formulation is not applicable, and hence time-domain methods should be used. In this paper, time-domain formulations were used to predict aeroelastic loads acting on a rigid bridge deck section model subject to stationary and gusty straight-line winds based on the knowledge of upstream wind speed and model displacement measurements. In this procedure, rational functions and indicial functions are used to formulate self-excited forces and buffeting forces, respectively. The functions used here were recently obtained from wind tunnel tests performed on a streamlined bridge deck section model with a smaller geometric scale. The results of the validation using a larger section model of this bridge deck subject to a ramp-type gust are presented here.

The reliable measurement of pressures on low-rise buildings in the atmospheric boundary layer (ABL) flow remains a challenge, as has been shown by the large discrepancies among results obtained in different wind tunnel facilities or even in the same wind tunnel. Two major causes of the discrepancies are the difficulty of simulating large-scale, low-frequency turbulent fluctuations uniformly across laboratories and the small scale of models in typical civil engineering wind tunnels. To address these issues, it was proposed that a simplified flow be used in laboratory simulations, rather than a conventional ABL flow. In the simplified flow the reference mean wind speed is larger than the mean wind speed of the ABL flow, and the low-frequency fluctuations present in the ABL flow are suppressed; that is, the peak energy of the missing low-frequency fluctuations is supplied in the simplified flow by the increment in the mean wind speed, which may be regarded as a flow fluctuation with zero frequency. High-frequency turbulent fluctuations, which typically affect flow reattachment, are approximately the same in the ABL and the simplified flow. Because, over small distances, low-frequency fluctuations are highly coherent spatially for small low-rise buildings with dimensions of up to approximately 20 m (e.g., single-family residential homes), the peak aerodynamic effects of the two flows may be hypothesized to be approximately the same. Preliminary experimental results obtained in University of Western Ontario’s ABL wind tunnel facility and Florida International University’s small-scale Wall of Wind facility are shown to support this hypothesis. The use of the proposed simplified flow is currently being tested by the authors for application to computational wind engineering (CWE) applications. Such use eliminates the need to simulate the lower frequency fluctuations of the boundary layer flow and thus makes it possible to achieve practical CWE calculations, and it is advantageous in experiments from the points of view of measurement accuracy, model scaling, repeatability of the simulations, and computational efficiency.

Based on previous studies, it has been determined that internal pressures can contribute as much (or more) as the external pressures to the net pressures occasioned for low-rise buildings with a single dominant opening. However, the role of internal pressures, as experienced for cases with realistic geometry and multiple openings, is rarely investigated. As is known, low-rise buildings are susceptible to multiple-opening failures during strong windstorms, and the resulting changes in internal pressures are critical measures for tracking how the failure progresses from a localized level to a catastrophic level. Using the Wind Tunnel Laboratory located at Louisiana State University, by testing a typical 5:12 gable-roof residential-building model with 17 openings, a full picture of the internal pressure changes experienced during progressive failure of multiple openings is obtained. The results indicated that (1) mean internal pressure coefficients estimated by two currently available, public hurricane-loss models were closer to the minimum internal pressure coefficient measurements than the mean internal pressure coefficient measurements at each stage of the multiple-opening failure; (2) ASCE guidelines underestimate the internal pressure for both partially enclosed and enclosed building types; and (3) Helmholtz resonance occurred near the theoretical resonance frequencies for two single opening cases for the current model with high damping. Based on these results, the flow resonance, dependent on the building geometry and the opening layout, should be considered to predict the potential wind-induced damage for low-rise buildings, and this is what is explored in this paper.

Extensive damage to residential wood-frame buildings caused by failures of roof-to-wall connections during extreme wind events underscores the need to improve their performance. Most of these connections use mechanical connectors, e.g., metal clips and straps (sometimes referred to as hurricane clips and hurricane straps). The allowable capacity of these connectors is based on results of unidirectional component tests that do not simulate multiaxial aerodynamic loading effects induced by high wind events. The objective of this research was to facilitate a better understanding of these loading effects on roof-to-wall connections of a typical low-rise gable roof residential structure subjected to combined impacts of wind and a potential breach of the building envelope. Large-scale experiments on a heavily instrumented building model generated multiaxial aerodynamic loading data on roof-to-wall connections for various wind angles of attack and internal pressure conditions. The results showed the severity of increased loading on connections in certain configurations of wall openings that could occur from the breach of the building envelope in windstorms. It was also shown that lateral components of the wind load acting simultaneously with the uplift may be significant, depending on wind angles of attack and internal pressure conditions. Aerodynamic test data were used to perform component-level triaxial-load testing on hurricane clips to determine their load capacities and compare results to those obtained using the traditional approach of uniaxial-load testing. Component level test results showed that current uniaxial testing methods tend to overestimate the actual load capacities of metal connectors. Neglecting triaxial loading effects in testing of connectors and in the design of connections could potentially cause the type of failures frequently documented in hurricane damage reconnaissance reports.

This paper presents the use of simple architectural elements such as aerodynamic mitigation devices for reducing high wind-induced suctions occurring at roof and wall corners of low-rise buildings where wind-induced building envelope failures usually initiate. The architectural elements considered in the current study include trellises (pergolas), roof extensions of gable ends (gable end ribs), ridgeline extensions (ridge rib), and sideways extensions of walls (wall ribs). A small-scale model of residential villa was tested in a boundary layer wind tunnel for two different roof geometries (gable and hip). Moreover, selected cases were investigated at the Wall of Wind, a large-scale testing facility, to investigate scale effects. The effectiveness of these architectural elements in reducing high suction (negative pressure) was assessed by comparing the wind-induced pressure measurements obtained before and after introducing the elements. Based on the results obtained for the worst wind angle of attacks, the peak suction was reduced after introducing the architectural elements by 65% at gable-end corners, 60% close to roof ridges, 45% at soffits, 35% at wall corners, and 25% at eaves. These simple architectural elements, which can be retrofitted to the stock of existing homes or incorporated into the design of new buildings, can be used as aerodynamic mitigation devices for reducing peak suction at critical locations of the building envelope.