Search for: All records

Transportation infrastructure experiences distress due to aging, overuse, and climate changes. To reduce maintenance costs and labor, researchers have developed various structural health monitoring systems. However, the existing systems are designed for short-term monitoring and do not quantify structural parameters. A long-term monitoring system that quantifies structural parameters is needed to improve the quality of monitoring. In this work, a novel Transportation Rf-bAsed Monitoring (TRAM) system is proposed. TRAM is a multi-parameter monitoring system that relies on embeddable backscatter-based, batteryless, and radio-frequency sensors. The system can monitor structural parameters with 3D spatial and temporal information. Laboratory experiments were conducted on a 1D scale to evaluate and examine the sensitivity and reliability of the monitored structural parameters, which are displacement and water content. In contrast to other existing methods, TRAM correlates phase change to the change in concerned parameters, enabling long-term monitoring. 

Concrete exhibits time-dependent long-term behavior driven by creep and shrinkage. These rheological effects are difficult to predict due to their stochastic nature and dependence on loading history. Existing empirical models used to predict rheological effects are fitted to databases composed largely of laboratory tests of limited time span and that do not capture differential rheological effects. A numerical model is typically required for application of empirical constitutive models to real structures. Notwithstanding this, the optimal parameters for the laboratory databases are not necessarily ideal for a specific structure. Data-driven approaches using structural health monitoring data have shown promise towards accurate prediction of long-term time-dependent behavior in concrete structures, but current approaches require different model parameters for each sensor and do not leverage geometry and loading. In this work, a physics-informed data-driven approach for long-term prediction of 2D normal strain field in prestressed concrete structures is introduced. The method employs a simplified analytical model of the structure, a data-driven model for prediction of the temperature field, and embedding of neural networks into rheological time-functions. In contrast to previous approaches, the model is trained on multiple sensors at once and enables the estimation of the strain evolution at any point of interest in the longitudinal section of the structure, capturing differential rheological effects. 

Abstract Structural health monitoring (SHM) is the automation of the condition assessment process of an engineered system. When applied to geometrically large components or structures, such as those found in civil and aerospace infrastructure and systems, a critical challenge is in designing the sensing solution that could yield actionable information. This is a difficult task to conduct cost-effectively, because of the large surfaces under consideration and the localized nature of typical defects and damages. There have been significant research efforts in empowering conventional measurement technologies for applications to SHM in order to improve performance of the condition assessment process. Yet, the field implementation of these SHM solutions is still in its infancy, attributable to various economic and technical challenges. The objective of this Roadmap publication is to discuss modern measurement technologies that were developed for SHM purposes, along with their associated challenges and opportunities, and to provide a path to research and development efforts that could yield impactful field applications. The Roadmap is organized into four sections: distributed embedded sensing systems, distributed surface sensing systems, multifunctional materials, and remote sensing. Recognizing that many measurement technologies may overlap between sections, we define distributed sensing solutions as those that involve or imply the utilization of numbers of sensors geometrically organized within (embedded) or over (surface) the monitored component or system. Multi-functional materials are sensing solutions that combine multiple capabilities, for example those also serving structural functions. Remote sensing are solutions that are contactless, for example cell phones, drones, and satellites. It also includes the notion of remotely controlled robots. 

Multistatic GPR has the advantages of reducing survey time and leverages more comprehensive data collection. Traditionally in multistatic GPR data processing, the 2D Bscan image obtained from each receive antenna are simply stacked for 3D image reconstructions. However, such approach is typically inadequate as the multistatic GPR receivers are mounted with spatial offsets, causing back-scattering signals from the same target to have differing time of arrivals. For proper fusion of multistatic GPR data, migration methods that consider the transmitters and receivers spatial offset and data variations among different receiving antennas may be employed. In this study, the back-projection algorithm (BPA) is investigated. The algorithm consists of determining the wave travel path and associated travel time, and projecting the corresponding signal value back into space domain. Furthermore, antenna radiation pattern is incorporated. The BPA enables scatter shape reconstruction and is prone to parallel computing. For validation, multistatic GPR 3D tomographic image reconstruction is successfully applied to laboratory data. 

Ground penetrating radar (GPR) is a remote geophysical sensing method that has been applied in the localization of underground utilities, bridge deck survey, localization of landmines, mapping of terrain for aid in driverless cars, etc. Multistatic GPR can deliver a faster survey, wider spatial coverage, and multiple viewpoints of the subsurface. However, because of the transmit and receive antennas spatial offset, formation of 3D GPR image by simple stacking of the acquired A-scans is inaccurate. Also, averaging of different receivers data may lead to destructive interference of back-scattered waves due to different time delays implied by the spatial offset, so averaging does not lead to higher SNR in general. Furthermore, the energy back-scattered by scatter points are spread in hyperbolas in the GPR raw data. Migration or imaging algorithms are employed to increase SNR by focusing the hyperbolas. This focusing process also leads to better accuracy in target localization. In this paper, a computationally efficient synthetic aperture radar (SAR) imaging algorithm that properly integrates multistatic GPR data in both ground and air-coupled cases is presented. The algorithm is successfully applied on two synthetic datasets. 

The development of modern cities heavily relies on the availability and quality of underground utilities that provide drinking water, sewage, electric power, and telecommunication services to sus- tain its growing population. However, the information of localiza- tion and condition of subterranean infrastructures is generally not readily available, especially in areas with congested pipes, which impacts urban development, as poorly documented pipes may be hit during construction, affecting services and causing costly de- lays. Furthermore, aging components are prone to failure and may lead to resources waste or the interruption of services. Ground penetrating radar (GPR) is a promising remote sensing technique that has been recently used for mapping and assessment of under- ground infrastructure. However, current commercial GPR survey systems are designed with wheel-encoders or GPS for positioning. Wheel-encoder based GPR surveys are restrained to linear-route only, preventing the use of GPR for accurate localization of city wide underground infrastructure inspection. While GPS signal is degraded in urban canyons and unavailable in city tunnels. In this work, we present a new GPR system integration with augmented reality (AR) based positioning that can overcome the limitations of current GPR systems to enable arbitrary-route scanning with a high fidelity. It has the potential for automation of GPR survey and integration with AR smartphone applications that could be used for better planning in urban development. 

Ground penetrating radar (GPR) subsurface sensing is a promising nondestructive evaluation (NDE) technique for inspecting and surveying underground utilities in complex urban environments, as well as for monitoring other key infrastructure such as bridges and railroads. A challenge of such technique lies on image formation from the recorded GPR data. In this work, a fast back projection algorithm (BPA) for three-dimensional GPR image construction is explored. The BPA is a time-domain migration method that has been effectively used in GPR image formation. However, most of the studies in the literature apply a computationally intensive BPA to a two-dimensional dataset under the assumption that an in-plane scattering occurs underneath the GPR antennas. This assumption is not precise for 3D GPR image formation as the GPR radiation scatters in multiple directions as it reaches the ground. In this study, a generalized form for an approximation to determine the scattering point in an air-coupled GPR system is developed which considerably reduces the required computations and can accurately localize the scattering point position. The algorithm is evaluated by applications on GPR data synthesized using GprMax, a finite-difference time domain (FDTD) simulator.