Journal of Intelligent Material Systems and Structures

A method for damage detection based on self-sensing piezoelectrics is presented. When a piezoelectric actuator is electronically excited, it causes stress in the structure to which it is attached. As the structure deformation is influenced by the presence of damage, so will be the response of the sensor. The goal of the method is to use the same piezoelectric element as both actuator and sensor, by analyzing the electromechanical response of the system in the frequency domain. In this way, damaged and undamaged structures can be distinguished and the extent of damage estimated. The distinction between the excitation and the response signals is achieved by an electronic bridge circuit. A comparison between this self-sensing method and the more common cross-talk method is performed, showing the similarity of the two methods and the validity of our technique. A method to differentiate among the spectrum changes due to actual damage and those due to temperature shifts is also presented. As the temperature rises, the PZT characteristics change and the resonance frequencies of the spectrum shift. However, the acquired signal can be processed to compensate for the temperature changes.

Search strategies are compared that can potentially identify single, rectangular, midplane delaminations in composite plates using the signals from surface-mounted piezoceramics. The strategies are demonstrated by repeatedly assuming a single delamination and then running a delaminated plate model, also described, that predicts transducer frequency responses. An objective function compares the responses to those for a hypothetical plate with unknown damage. When the objective function is minimized, the responses agree, and the assumed delamination is the best-estimate. The chosen objective function is well-behaved but has several local minima; therefore, rardom search strategies are used to find the global minimum. The most efficient strategy starts with assumed delaminations selected from a uniform random distribution at discrete points in the parameter space (i.e., a combinatorial search). Some of these points will be close to local minima. The strategy continues by using a random walk with shrinking step size across the parameter space. This latter search is made more robust by using probabilistic hill climbing acceptance tests, also known as simulated annealing. The overall technique systematically investigates several possible local minima and identifies a best-estimate delamination to a preset accuracy.

Health monitoring of civil infrastructure systems is cost-effective and necessary since these systems are generally the most expensive investments/assets in any country. In the U.S. these assets are estimated at $20 trillion. In addition, these systems have long service life compared with any other kinds of commercial product, and are rarely replaceable once they are erected. Yet the feedback and controls on the "state of health" of constructed systems are practically nonexistent. Nondestructive evaluation (NDE), sensing and smart structures are essential parts of this feedback and monitoring system for infrastructures. The NSF NDE initiative as well as workshops and related awards/projects are described in this paper.

A technique for monitoring vibration measurements to detect structural damage is presented. Frequency response functions from a healthy structure are measured for reference data, then cross-spectral densities between pairs of combined damage and external forces are computed to detect any occurring damage. The excitation forces are not measured, and can be uniform, random, and uncorrelated, or applied at a single point on the structure. The technique bounds the damage location between the closest sensors on the structure, only a small number of sensors are needed, the damage force is approximated, and no model of the structure is used. For bending vibrations, rotations must also be measured which in practice is often difficult. In a finite-element simulation, the method located small damage and approximated the damage force for a beam structure.

Structural health is directly related to structural performance and in this respect it is a governing parameter with regard to safety of operation. This aspect of structural health is particularly relevant for transportation systems and in this connection structural health monitoring is a safety issue. In aviation, and particularly in air transportation, structural health monitoring has been developed to a rather satisfactory level albeit using rather unsophisticated methods of inspection involving considerable man-hour requirements as well as very special skills and attitudes on the part of the inspector. Undoubtedly structural reliability of aircraft has increased considerably also as a result of the implementation of damage tolerant design practices and of automated fabrication including joining with a high degree of controllability and repeatability. It is particularly the aspects of sensitivity and reliability of manual inspection that stimulate the study of more sophisticated automated and integrated damage sensing systems as a smart substitute for the current practice relying heavily on human intelligence. The paper identifies the study and development of smart damage detection systems as an interdisciplinary activity involving specialists in different technical areas. The risks involved are of a technical nature but also relate to approval and acceptance of automated systems in air transportation systems. It describes the current situation in Europe that is international in character by definition. Different supranational bodies have recognized the development of smart systems as substitutes for human operations as a key technology and they provide funding that enables the formation of development teams with a good basis for continuity.

The present paper offers structural damage detection methods based on the relative changes in localized flexibility properties. The localized flexibility matrices are obtained either by applying a decomposition procedure to an experimentally determined global flexibility matrix or by processing the output signals of a vibration test in a substructure-by-substructure manner. Three methods are evaluated in the present paper: a free-free substructural flexibility method, a deformation-based flexibility method, and a strain-basis flexibility method. These methods are applied to a model ten-story building problem, an experimental bridge damage data, and a hybrid experimental-simulated engine structure. The present methods are found to correctly identify damage locations.

This paper summarizes the theory used in a system that identifies foreign body impact on a composite plate using built-in strain sensors. The identification system consists of a model of the composite plate and an identification algorithm. The algorithm compares the measured response of the plate to the model response and estimates the impact location and force time history. The solution, which uses a smoother/filter optimization, includes a new computational algorithm that saves considerable computation time for systems with many degrees of freedom.

This paper summarizes the implementation of an impact identification system on a composite plate. The impact, normal to the surface of the plate, generates a traveling wave that is measured by surface mounted sensors. The identification system compares the measured response of the plate to the model response and estimates the impact location and force time history. The accuracy and reliability of the system was tested using many impacts distributed across the plate and the robustness of the system was tested by adding simulated noise to the data. The system was tested on the composite plate with free boundaries and clamped boundaries. The system proved to be accurate with an average location error ranging from .39 inches for the free boundary case to .49 inches for the clamped plate with noise. The area monitored by the 13 sensor array was 20 inches by 20 inches.

Optical accelerometers developed in both fiber and integrated optics technologies and several traditional sensors are integrated in a quasi-smart structure for real-time monitoring and predictive maintenance of large Electric Power Generators. The structures were properly tested in field trials.

The feasibility of detecting stress-wave acoustic phenomena with Honeywell's resonant microbeam-based MEMS sensor technology is presented. The baseline technical approach described here incorporates an integrated silicon microstructure and a resonant microbeam with micron-level feature size, and features frequency sensitivity up to 500 kHz. The stress-wave acoustic MEMS concept has been demonstrated successfully in the laboratory test environment to sense simulated acoustic emission (AE) events for structural fatigue crack monitoring applications. The technical design approach and laboratory test results are presented.

In this paper we investigate an eddy current microsensor which uses a low input impedance trans-impedance amplifier to minimise the influence of stray capacitances on the sensor operation when being used for micromachine applications. A technique using a two transistor feedback pair for realising very low input impedance over a wide bandwidth in these amplifiers is studied. An improved circuit configuration, which further reduces the amplifier input impedance at higher frequencies, is investigated by computer simulation. At the same time as minimising the influence of the stray capacitances, the trans-impedance amplifier improves the signal conditioner sensitivity and output drift performance. The operation of the sensor signal conditioner has been investigated experimentally using target materials suitable for micromachine applications. Results of these experiments are presented which show that acceptable sensitivities are achieved.