For rapid civil infrastructure assessment following natural and man-made emergencies, the utilization of minimally invasive and cost-effective drone deployable sensor packages has the potential to become a valuable tool. Although compact sensors with wireless data transfer capabilities have proven effective in monitoring the structural dynamics of infrastructure, these systems require data processing to occur externally, frequently off-site. These extra steps impede the high-speed assessment of a structure’s state. Difficulties can arise when the transmission is unfeasible due to degraded communication links during natural or man-made emergencies. Additionally, off-site data processing can add unneeded interruptions to actions that can be taken by emergency personnel after infrastructure damage. To enhance the effectiveness of sensor packages in expediting infrastructure assessment, incorporating real-time data analysis through embedded edge computing techniques emerges as a promising solution. The objective of this work is to demonstrate on-device data processing for frequency-based structural health monitoring techniques using drone-deployable sensors. This approach advances the effectiveness of drone-deployable sensors in rapid infrastructure assessment by mitigating their susceptibility to errors or delays in data communications. The proposed approach computes the frequency components of vibration measurements taken from a structure of interest, for example, the monitoring of a bridge immediately following a damaging event such as a flood. This work presents contributions in terms of outlining a methodology that emphasizes the hardware-based implementation of edge computing algorithms and examines the required on-device performance and resource utilization for structural health monitoring at the edge. The execution time for the sensor’s edge computing functions was profiled, resulting in an additional 9.77 seconds per test, an advancement over traditional transmit and analyze methods.
Electronic assemblies are subjected to damaging impact and shock loadings in various scenarios, including aerospace, automotive, and military applications. In safety-critical situations, the online detection, quantification, and localization of damage within the electronic assembly would enable intelligent systems to take corrective actions to mitigate or circumvent the effects of damage within the electronic assemblies. This preliminary work investigates a reduced-order model-based method for online damage detection, quantification, and localization of printed circuit boards (PCBs). The local eigenvalue modification procedure (LEMP) is used to accelerate the computational processing time of the model, thereby enabling its use in online damage detection during an impact or shock event. The proposed method tracks changes in the model’s state using an error minimization technique in the frequency domain. A baseline state is established by creating and simulating a numerical model that accurately represents a healthy PCB response. Potential reduced-order models with varying stiffness matrices are developed online and compared to the system’s current state. These reduced-order models introduce a single change in stiffness to the system. LEMP calculates the overall change in the system to obtain the new system-level dynamic response. Incorporating LEMP within the frequency-based analysis demonstrates the potential for effective damage detection on PCBs. This work validates the proposed methodology using a rectangular PCB with induced damage. The PCB is modeled pinned at each corner, and its dynamic response is simulated using ABAQUS and processed with the generalized eigenvalue procedure. LEMP is used to update a single change in the system while obtaining a 587 times speed up when compared to the generalized eigenvalue approach. The LEMP algorithm performance and reliability for updating the model state are discussed in the paper.
KEYWORDS: Sensors, Education and training, Data modeling, Vibration, Structural health monitoring, Performance modeling, Accelerometers, Systems modeling, Signal to noise ratio, Signal attenuation
For rapid assessment of infrastructure, the use of minimally invasive sensors that can be deployed remotely using autonomous vehicles is gaining popularity. Such systems are favorable for their ease of deployment and cost-effectiveness. Utilizing electropermanent magnets or adhesives to mount the sensors temporarily forms a barrier between the sensor and the structure being examined. This barrier creates undesirable nonlinearities and transmissibility losses that introduce errors into structural damage detection algorithms. Post-processing of signals using continuous modeling techniques from classical control theory can be applied to the collected signals to remove this error. However, post-processing creates additional analysis steps that require the signal to be taken off device. Processing the data at-the-edge prior to saving it to memory or transmitting it to a base station enables rapid assessment of infrastructure. With minimal time from signal detection to prognostics, such systems can be used in damage forecasting and infrastructure failure prevention. This preliminary work aims to develop a non-linear machine-based compensation technique that is resource and power efficient enough to be processed on-device. The proposed long short-term memory (LSTM) error-compensating network demonstrated potential by increasing the SNRdB by 9.3% and improving RMSE by approximately 20% while widening the usable lower limit of the sensor’s bandwidth from 2.78 to 1.34 Hz. The progress described in this report focuses on setting the framework for the proposed method and paves the way for a full-scale hardware implementation in the near future
The potential of levee failures poses significant risks to populations living behind them. Levee monitoring using ground velocity measurements obtained from geophones has been demonstrated with the simultaneous deployment of wired geophone arrays. However, the scale of levees makes their monitoring with wired sensors a challenging task. This work reports on the development of a stand-alone geophone monitoring system for levees constructed of earthen embankments. The newly developed open-source sensor package can simultaneously measure ground velocity, conductivity, and temperature in addition to ambient atmospheric pressure and humidity. The system is fully independent of processing, power management, sensors, and data storage all contained within a single instrument. This work reports the initial experimental validation of the proposed system using a granular earthen levee in a flume under controlled erosion conditions. Data is collected and post-processed for anomaly detection; sensing capabilities, and the effect of sensor noise are discussed. To the knowledge of the authors, this is the first open-source stand-alone geophone system developed and tested for the monitoring of earthen levees.
KEYWORDS: Field programmable gate arrays, Clocks, Neurons, Signal to noise ratio, Structural health monitoring, Sensors, Detection and tracking algorithms, Data acquisition, Damage detection, Algorithm development
Hard real-time time-series forecasting of temporal signals has applications in the field of structural health monitoring and control. Particularly for structures experiencing high-rate dynamics, examples of such structures include hypersonic vehicles and space infrastructure. This work reports on the development of a coupled softwarehardware algorithm for deterministic and low-latency online time-series forecasting of structural vibrations that is capable of learning over nonstationary events and adjusting its forecasted signal following an event. The proposed algorithm uses an ensemble of multi-layer perceptrons trained offline on experimental and simulated data relevant to the structure. A dynamic attention layer is then used to selectively scale the outputs of the individual models to obtain a unified forecasted signal over the considered prediction horizon. The scalar values of the dynamic attention layer are continuously updated by quantifying the error between the signal’s measured value and its previously predicted value. Deterministic timing of the proposed algorithm is achieved through its deployment on a field programmable gate array. The performance of the proposed algorithm is validated on experimental data taken on a test structure. Results demonstrate that a total system latency of 25.76 µs can be achieved on a Kintex-7 70T FPGA with sufficient accuracy for the considered system.
KEYWORDS: Sensors, Unmanned aerial vehicles, Signal to noise ratio, Electronic filtering, Structural health monitoring, Signal detection, Process modeling, Microelectromechanical systems, Magnetic sensors, Algorithm development
The rapid assessment of infrastructure following extreme weather or seismic events is important to ensure the stability of structures before their continued use. This work presents an amplitude compensation technique for accurate acceleration measurements formulated for unmanned aerial vehicle’s (UAV) deliverable sensor packages. These packages are designed for measuring the acceleration of structures, for instance, railroad bridges and power transmission towers. Current technology for structural health monitoring is expensive, stationary, and requires maintenance by certified personnel. These attributes prevent rapid assessment of remote and hard-to-reach structures. Low-cost, UAV-delivered sensor packages are an ideal solution due to their ability to be deployed on a large scale in a timely manner; cutting down on cost and the danger affiliated with structural health monitoring following extreme and hazardous events. One challenge to this approach is that the UAV deployable sensor package consists of several systems, including mounting hardware, embedded electronics, and energy storage that result in a loss of transmissibility between the structure and the package’s accelerometer. This work proposes a frequency response-based filter to isolate the structure’s vibration signature from interference caused by the sensor package itself. Utilizing an input-output relationship between the sensor package and a calibrated reference accelerometer, a model transfer function is constructed. Compensation is performed in the post-processing stage using the inverse transfer function model. This approach is shown to enhance the signal-to-noise ratio by 1.2 dB, an increase of 7.17%. This work investigates algorithm robustness and sensitivity to noise across the sensor package’s bandwidth of 6-20 Hz. A discussion on the limitations of the system is provided.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.