Eric B. Flynn

Maximum-likelihood estimation of damage location in guided-wave structural health monitoring

This paper describes the formulation of a maximum-likelihood estimate of damage location for guided-wave structural health monitoring (GWSHM) using a minimally informed, Rayleigh-based statistical model of scattered wave measurements. Also introduced are two statistics-based methods for evaluating localization performance: the localization probability density function estimate and the localizer operating characteristic curve. Using an ensemble of measurements from an instrumented plate with stiffening stringers, the statistical performance of the so-called Rayleigh maximum-likelihood estimate (RMLE) is compared with that of seven previously reported localization methods. The RMLE proves superior in all test cases, and is particularly effective in localizing damage using very sparse arrays consisting of as few as three transducers. The probabilistic basis used for modelling the complicated wave scattering behaviour makes the algorithm especially suited for localizing damage in complicated structures, with the potential for improved performance with increasing structure complexity.

A mobile-agent-based wireless sensing network for structural monitoring applications

A new wireless sensing network paradigm is presented for structural monitoring applications. In this approach, both power and data interrogation commands are conveyed via a mobile agent that is sent to sensor nodes to perform intended interrogations, which can alleviate several limitations of the traditional sensing networks. Furthermore, the mobile agent provides computational power to make near real-time assessments on the structural conditions. This paper will discuss such prototype systems, which are used to interrogate impedance-based sensors for structural health monitoring applications. Our wireless sensor node is specifically designed to accept various energy sources, including wireless energy transmission, and to be wirelessly triggered on an as-needed basis by the mobile agent or other sensor nodes. The capabilities of this proposed sensing network paradigm are demonstrated in the laboratory and the field.

Optimal Placement of Piezoelectric Actuators and Sensors for Detecting Damage in Plate Structures

We propose a novel approach for optimal actuator and sensor placement for active sensing-based structural health monitoring. Of particular interest is the optimization of actuator—sensor arrays making use of ultrasonic wave propagation for detecting damage in thin plate-like structures. Using a detection theory framework, we establish the optimum configuration as the one that minimizes Bayes risk. The detector incorporates a statistical model of the active sensing process that accounts for both reflection and attenuation features, implements pulse-echo and pitch-catch actuation schemes, and takes into account line-of-site. The optimization space was searched using a genetic algorithm with a time-varying mutation rate. For verification, we densely instrumented a concave-shaped plate and applied artificial, reversible damage to a large number of randomly generated locations, acquiring active sensing data for each location. We then used the algorithm to predict optimal subsets of the dense array. The predicted optimal arrangements proved to be among the top performers when compared to large sets of randomly generated arrangements.

Enhanced detection through low-order stochastic modeling for guided-wave structural health monitoring

Guided wave structural health monitoring offers the potential for efficient defect detection and localization on large plate-like structures. The aim of this article is to demonstrate that the performance of guided wave structural health monitoring on complex structures can be quantified, predicted, and enhanced through basic stochastic modeling and application of a likelihood-based detector. These are necessary steps in enabling such structural heath monitoring techniques to be used more robustly on safety-critical structures. Extensive ensembles of measurements were taken using a sparse array of transducers on an aluminum plate with geometric complexities (stiffeners), before and after introducing several sizes of damage at several locations, and in the presence of large noise processes. This enabled a full statistical treatment of performance evaluation using a modification of the standard receiver operating characteristic curve. The incorporation of stochastic modeling produced on average a 44% reduction in missed detections among the damage modes tested, with up to a 95% reduction in some cases.

Statistically-based damage detection in geometrically-complex structures using ultrasonic interrogation

This article introduces a novel approach to damage detection in the context of a structural health monitoring system. A statistical model of the ultrasonic guided wave interrogation process was developed and used to formulate a likelihood ratio test. From the likelihood ratio test, the optimal detector for distinguishing damaged and undamaged states of the structure was derived and found to be a metric related to signal energy. That result was confirmed using a data-driven approach based on using receiver-operating characteristic curves to compare the energy metric to other metrics found in the literature. The data in this study were generated from instrumenting two separate, geometrically complex structures with ultrasonic sensor–actuators. A three-story bolted-frame structure was constructed in the laboratory to test the approach for connections that produce highly uncertain wave paths, with damage being introduced through local impedance changes and bolt loosening. The second test structure was a section of a fuselage rib taken from a commercial aircraft in which holes and cracks were introduced to provide a testbed with a high degree of realism. The detection performance in both structures was quantified and presented. Finally, different sensor fusion strategies were implemented and the ability of such techniques to increase the statistical distinction between damaged and undamaged cases was quantified.

Multi-wave-mode, multi-frequency detectors for guided wave interrogation of plate structures

The detection and localization of damage using an array of closely spaced transducers is investigated theoretically and experimentally using single- and multiple-mode guided wave active sensing models. Detectors are derived using a generalized likelihood ratio approach assuming that amplitude, absolute phase, and source location of a scattered wave are unknown, while frequency, group velocity, and phase velocity are known. Theoretical detection performance for processing with each detector is derived and related to the energy-to-noise ratio of a scattered mode as a metric of determining when processing with multiple modes provides increased performance over processing with a single mode. Experimentally, detectors are implemented to detect scattering from a small mass glued to the surface of an aluminum plate with a 7 × 7 array of transducers. Relative detection and localization performance is compared through receiver operating characteristic curves and histograms of distance from true damage location for 1000 no-damage and damaged measurements. A single-mode, multi-frequency detector is shown to have the best detection and localization performance for the tested damage scenarios.

Wireless impedance device for electromechanical impedance sensing and low-frequency vibration data acquisition

This paper presents recent developments in an extremely compact, wireless impedance sensor node for combined use with both impedance method and low-frequency vibrational data acquisition. The sensor node, referred to as the WID3 (Wireless Impedance Device) integrates several components, including an impedance chip, a microcontroller for local computing, telemetry for wireless data transmission, multiplexers for managing up to seven piezoelectric transducers per node, energy storage mediums, and several triggering options into one package to truly realize a self-contained wireless active-sensor node for SHM applications. Furthermore, we recently extended the capability of this device by implementing low-frequency A/D and D/A converters so that the same device can measure low-frequency vibration data. The WID3 requires less than 60 mW of power to operate and is designed for the mobile-agent based wireless sensing network. The performance of this miniaturized device is compared to our previous results and its capabilities are demonstrated.

A High-Speed Dual-Stage Ultrasonic Guided Wave System for Localization and Characterization of Defects

This paper presents a joint approach to the detection, localization and characterization of structural defects. Our technique, termed Dual-Stage Structural Health Monitoring (DSSHM), combines an online, embedded ultrasonic structural health monitoring system with acoustic wavenumber spectroscopy data gathered from a scanning laser Doppler vibrometer. The combined system provides non-disruptive monitoring and high measurement fidelity necessary to drive critical decisions. Stage 1 is performed in situ (while the structure is in operation): an array of embedded ultrasonic transducers is used to excite and measure ultrasonic wave pulses in order to detect and locate irregularities within the test structure. In Stage 2, a subset of the same transducer array produces a steady-state ultrasonic response in the immediate vicinity of each structural irregularity. A laser Doppler vibrometer is then introduced to measure the full field response around that particular location in order to verify the presence of damage and provide high fidelity localization and characterization of the affected area. This document explores both the application of ultrasonic guided wave physics and the integration of sensing hardware and signal processing algorithms in implementing DSSHM.

Bayesian probabilistic structural modeling for optimal sensor placement in ultrasonic guided wave-based structural health monitoring

Many optimal sensor placement methods for structural health monitoring establish performance metrics based on the detection of a limited set of damage states and locations. In guided wave-based inspection, however, monitoring is carried out over a continuous region with a continuous distribution of possible damage locations, types, sizes, and orientations. Here, traveling waves are excited and then received by a set of transducers with the intent of detecting and localizing previously unobserved scattering sources that are associated with damage. To measure sensor network performance in this application, we implement a Bayesian experimental design approach by computing the total posterior expected cost of detection over the entire monitoring region. Since the optimization usually must be carried out using a computationally expensive meta-heuristic such as a genetic algorithm, efficient modeling of the interrogation process is key to solving this distributed sensor placement problem. In this work, we implement a previously developed semi-analytical modeling approach for wave scattering within our Bayesian probabilistic framework in order to optimally place active sensors for detecting cracks of unknown location, size, and orientation. This involves assuming a set of a priori probability distributions on the three unknowns and defining spatial distributions of cost associated with type I and type II detection error. These parameters are driven by the geometry, material, in-service structural loading, and performance requirements of the structure. Through a set of sensor placement examples, we demonstrate how changes in the probability and cost distributions will dramatically alter the optimal layout of the transducer network.

Damage detection on composite structures with standing wave excitation and wavenumber analysis

This paper describes the use of wavenumber filtering for damage detection with a signal-frequency standing wave excitation on composite structures. Using a single, fixed frequency excitation from a mounted piezoelectric transducer, the full steady-state wavefield could be obtained using a Laser Doppler Vibrometer with a mirror-tilting device. After completing the scanning, a wavenumber filtering is applied to determine dominant wavenumber components of the measured wavefield, which could be used for indicative of structural damage. Mapping processes based on local wavenumber filtering is then carried out for damaged area visualization. Also introduced are the comparison of two methods for damage identification and visualization: the local wavenumber mapping and acoustic wavenumber spectroscopy. To demonstrate the proposed techniques, several experiments are performed on composite structures with different types of damage, including debonding and delamination on composite plates. The results demonstrate that the techniques are very effective in localizing damage with the potential for the quick inspection of a variety of composite structural components.