Rf and dc planar magnetron sputtering systems were used to deposit high-temperature nickel- based super alloys, INCONEL 617, 625, Haynes 214, and thin films of palladium, as coatings on optical fibers for use in temperatures approaching 1000 degree(s)C. The nickel-based alloy coatings were applied on-line as the optical fiber was drawn, minimizing the exposure of the fiber to the deleterious effects of humidity. The thin film coatings of pure metals were sputtered using a new rf magnetron sputtering system custom designed and built for the Fiber and Electro Optics Research Center. The resulting coatings were analyzed using scanning electron microscopy, Auger electron microscopy, and energy dispersive x-ray spectroscopy. The coated fibers exhibit promise for embedded sensors in high temperature, high load composites used for advanced aerospace and energy applications.

The polarization preserving properties of single crystal sapphire optical fibers are investigated experimentally for different modal power distributions and different lengths of the fibers. Experimental results indicate that linearly polarized light launched along one of the principle axes of the fiber birefringence can be partially maintained. The polarization extinction ratio at the fiber output end was measured to be 6 dB and 3 dB for 7 cm and 32 cm long sapphire fibers, respectively. The temperature coefficient of the different phase delay between the orthogonal polarization modes was measured to be 0.028 rad. m^-1 degree(s)C^-1. A sapphire fiber-based polarimetric temperature sensor is constructed based upon both the polarization maintaining characteristics of the sapphire fiber and the temperature dependence of birefringence-induced phase delay. In this sensor a 1.4 cm long sapphire fiber serves as the sensing element. This sensor has been demonstrated for measurement of high temperature up to 1200 degree(s)C with a resolution of 2 degree(s)C.

A technique that uses embedded or attached high temperature optical fiber sensors is demonstrated for the intelligent processing of multilayer ceramic actuator elements. Presented are the results of the fiber optic strain sensor used to monitor internal displacements associated with the burnout of organic binders from a green actuator sample. Also presented is the method of operation of the low-finesse Fabry-Perot interferometric sensor, and post processing results obtained using the same type of sensor for tracking actuator performance and hysterisis.

Advanced composite materials have evolved to be the class of materials which meet the strict requirements of many ground- and space-based critical structures. Structures that incorporate these novel materials are frequently exposed to harsh environments during service and must be manufactured using high quality standards. Not only is it desirable to monitor and control the fabrication process of these materials using embedded sensors, but it is also necessary to instrument these structures with devices that allow in-service health monitoring. This paper describes an approach which integrates an optical fiber time domain strain and temperature monitoring network with a fiber optic cure monitoring system for composite materials. The proof-of-concept tests consisted of fabricating a thermoset composite laminate that combined a multi-segment optical fiber sensor network, monitored the state of cure during the fabrication of the laminate, and determined strain and temperature during tests performed after fabrication. Software was written to interface a commercially available optical time domain reflectometer system to a personal computer.

An adaptation of an extrinsic Fabry-Perot interferometer (EFPI) strain sensor is described that permits the state of cure of an epoxy matrix to be monitored when the sensor is embedded in a polymeric matrix composite. By using a glass rod with a retroreflecting end for the target fiber in the EFPI sensor, the intensity of the light reflected depends on the refractive index of the host matrix, if a low coherence source is used. As the epoxy cross-links during cure, the refractive index of the epoxy will increase to a value exceeding that of the target fiber. The resulting increased loss in the fiber can be detected at the sensor output and correlated to the state of cure of the epoxy. After cure, the sensor may be operated as a conventional EFPI strain sensor if a coherent source is used.

The commercialization of optical fiber sensors in smart structure applications largely depends on the development of a multi-sensor system, capable of simultaneously monitoring different optical fiber sensors. Efficient multi-sensor systems can be realized through multiplexing arrays of sensors within a system, reducing the number of input/output information channels between the support electronics and optical sensors. We demonstrate a sensor system, comprised of four multiplexed extrinsic Fabry-Perot interferometers, that uses a code division multiplexing technique to address each sensor. Qualitative measurements of strain and temperature, obtained with this multi-measurand multiplexing system, and an evaluation of crosstalk between sensors is presented.

This paper uses 3 X 3 coupler passive demodulation of combined Michelson and polarimetric optical fiber sensors to simultaneously measure axial and transverse strain components in a single optical fiber bonded to a vibrating cantilever beam. The separated Michelson and polarimetric signals are used with isothermal Michelson and polarimetric phase strain-models to provide two independent equations describing the strain in the fiber. These equations provide enough information to calculate axial and transverse strains in surface mounted sensors.

Arrays of fiber Bragg gratings (FBGs) have been produced on-line during fiber draw using a KrF excimer laser and a computer-controlled interferometer. A table of the desired Bragg wavelengths (lambda) _B of the gratings is used by the computer to appropriately vary the intersection angle of the two interfering beams, maintain focus and beam intersection at the proper position relative to the fiber, and trigger the laser. A large number of narrow-line Type I gratings have been produced, and in some instances small Type II gratings have been found to overlap the Type I FBGs. The reflectance of the gratings in one series was found to be polarization-dependent, lending further support to our previous model whereby the index change induced by the excimer laser occurs at the core-clad interface asymmetrically on one side of the core.

A number of techniques to address fiber Bragg grating (FBG) sensors have been described in recent years. This paper discusses the use of wavelength division multiplexing (WDM) couplers to perform the detection of wavelength shifts in the optical return signals from FBG sensor elements. These components allow all-fiber versions of the wavelength filtering ratiometric demodulation scheme to be implemented. Systems based on the use of a fused WDM coupler and a planar waveguide device are described.

Fiber Bragg gratings are of particular interest for distributed sensor applications such as embedded sensors for structural strain monitoring. This paper discusses the use of a wavelength division multiplexed fiber Bragg grating array for structural shape sensing and vibrational mode analysis. The gratings are surface attached to a cantilever beam and demodulated by a scanning fiber Fabry Perot filter to obtain strain information at different locations along the structure. A PC is used to read the strain information and perform static shape modeling of the beam.

Intracore Bragg grating sensors have been bonded on to the CFRP tendons of a prestressed concrete girder. The sensors survived both the installation procedure and casting of the concrete beam. Sensor performance is characterized in terms of maximum tensile strength and fatigue behavior. The fiber sensors survived strains of greater than 8000 (mu) (epsilon) and showed no change in either center wavelength or spectral content for 2000 (mu) (epsilon) over 320,000 cycles. The intracore grating sensor was used in a static loading test of the girder to failure and showed excellent stability and durability in comparison with the conventional technology.

We are developing low cost, rugged, and reliable fiberoptic sensors to meet current and future needs in civil engineering, including those of smart civil structures. Our work has concentrated on load, pressure, and displacement sensors, including pore water pressure sensors. We have built and demonstrated sensors in the laboratory with loads up to 50 lb., water pressures of 100 psi, and displacements up to 1 mm. Repeatability of sensor measurements are within 5% and are being improved with continued development. The range and sensitivity of the sensors can be easily changed without changing the basic sensor design. We also have multiplexed two water pressure sensors on a single fiber. We describe the sensor construction and experimental performance.

A Sagnac interferometric based strain sensor has the potential to measure strain of less than 100 microns over distances of 10 km or more. By incorporating these strain sensors into telecommunication grade fiber otic cable it would be possible to monitor earth movement to high accuracy over very long links at low cost on a continual basis. This technique would be extremely complementary to systems based on the Global Positioning Satellite and Satellite based radar imaging. The potential exists for incorporating the system directly into the local, regional, and national fiber optic telecommunication infrastructure, which would for the first time allow widespread data on earth movement to be obtained. This information would be useful in studying precursory deformation related to earthquake and volcanic activity, landslides, movement of land due to river outflows, and other earth movement features that directly impact the environment. This system could be used to reduce risks in hazardous waste site areas, to monitor strain o power and telecommunication lines, to monitor potential damage due to earth movement of utilities and buildings, and the movement of oil platforms at sea due to river outflows reducing oil spillage risks.

This paper describes an optical fiber interferometer that uses a short segment of silica hollow- core fiber spliced between two sections of single-mode fiber to form a mechanically robust in- line optical cavity. The hollow-core fiber is specifically manufactured to have an outer diameter that is equal to the outer diameter of the single mode lead fibers, thereby combining the best qualities of existing intrinsic and extrinsic Fabry-Perot sensors. Uniaxial tension and pure bending strength tests are used to show that the new configuration does not diminish the axial strength of bare fiber and reduces the bending strength by 17% at most. Similar tests confirm that the fiber sensor has 1.96% strain to failure. Axisymmetric finite element analysis is used to investigate the reliability of the in-line etalon during typical thermoset composite cure conditions, and parametric studies are performed to determine the mechanically optimal cavity length. The sensor strain response tests demonstrate a dynamic strain resolution of 21 n(epsilon) /(root)Hz at frequencies > 5 Hz with a sensor gauge length of 137 micrometers .

A four element, time-division multiplexed, fiber grating sensor array operating between the wavelengths of 1280 and 1310 nm was constructed and tested. The array consists of a 1300 nm edge-emitting LED which illuminates four gratings spaced five meters apart. Each grating in the array reflects a spectrally narrow-band (0.2 - 0.5 nm) wavelength region. Measurand- induced changes in the grating period change the wavelength of the light reflected by each grating. The wavelength shifts are converted to phase changes by routing the reflected signal through a nearly path balanced fiber Mach Zehnder interferometer. The minimum detectable strain was measured to be as low as 2 nanostrain/(root)Hz for the frequencies between 10 and 100 Hz.

This paper describes an ongoing effort to develop techniques capable of locating the position of space debris impacts and to quantify the strain energy absorbed by the space structure as a result of these impacts. The techniques under development use optical fiber sensors and neural networks as the primary sensor and decision making components. To date, this project has resulted in the development of (1) a mathematical model of plate impact dynamics for use in sensor and neural network paradigm development, (2) a sensor demodulation system specifically designed for moderate impact energies, (3) several neural network paradigms with the potential to locate impacts, and (4) a test configuration to experimentally confirm the proposed paradigms.

To optimize military fleet readiness, life cycle tracking of aircraft is required under the Air Force Aircraft Structural Integrity Program. Improvements in aircraft inspection and maintenance procedures are essential in today's climate of increasingly complex aircraft and decreasing defense outlays, forcing life extension programs for current aircraft. Automated structural health monitoring incorporating remote damage detection, load/environment tracking, and structural integrity assessment could provide significant cost savings over the life of an airframe. This paper presents the design requirements for development of an aircraft structural health monitoring system (SHMS). Design of a SHMS requires careful analysis of structural geometry, operational environment, expected damage modes, etc., to determine sensor categories and locations. Additionally, the task of integrating data collection, processing, and storage hardware into the airframe must be addressed. Sensors and sensing technologies are discussed along with specific requirements for monitoring system hardware and software. Anticipated life cycle savings versus implementation costs are also presented for the ideal SHMS.

This paper focuses on the signal processing aspect of a smart structure computational support environment for health monitoring, investigating the use of neural networks to identify and locate structural damage in a steel truss structure instrumented with accelerometers and strain gauges. Cracking damage is simulated by introducing sawcuts into the main members of the structure. Results using accelerometer data alone indicate that Quickprop backpropagation neural networks constitute a promising tool for these purposes, although network performance in locating damage should be improved by use of strain data as well.

The effective use of neural networks for fault detection, location, and classification requires training data. Distinguishing features in the data can be enhanced by pre-processing. Feature vectors that often prove advantageous in training networks for fault detection are modeshapes and curvatures; however, the procedures and sensors used to determine these can introduce problems. This paper uses numerical and practical experiments to investigate the use of acceleration, displacement, and strain response signals to extract modeshape and curvature functions from a cantilever plate and beam with localized damage. The importance of spatial accuracy, noise, and fault severity for fault detection is studied. It is shown that limiting spatial conditions occur with direct dynamic displacement measurements that have to be differentiated to obtain curvatures that can be overcome by using strain gauges that directly return quantities proportional to the curvature.

NRL's Mechanics of Materials Branch has developed the Embedded Sensors for Smart Structures Simulator (ES) design tool which relates the output of a finite number of sensors to strain induced structural damage. This tool is based on the use of the dissipative part of the bulk nonlinear material behavior. The methodology used to identify this behavior has evolved at NRL over the past 20 years. This paper describes the role of strain measurements and their relation to sensor type and location, the conceptual framework of dissipated energy density as the metric employed for assessing material/structure performance, the facilities provided by the simulator and their use, as well as implementation details. Through this we hope not only to make designing and verifying embedded sensor layouts on composite material structures a tractable task, but also to promote the use of dissipated energy density as a foundation upon which to build an effective means of measuring material and structural health. 13

The Smart Structures Concept Requirements Definition Contract objectives were to: (1) establish requirements for an on-board smart structural health monitoring system (SHMS); (2) assess relevant smart structures technologies; (3) develop conceptual near-term and ideal SHMS designs; (4) identify gaps in current technologies; (5) perform selective testing; (6) investigate system benefits; and (7) outline future R & D needs for development of an operational SHMS. The salient results and key lessons learned from the contract are summarized. Under the contract, monitoring requirements for metallic and composite structures have been established. A crucial requirement that cannot reliably be met with current sensing technologies is the need to monitor damage remotely, when the location is not known a priori. Other required components of a SHMS are more mature and are or will be available for near-term applications. The development of a fully automated SHMS will require coordinated multidisciplinary efforts, the paper discusses these and related issues and presents an overall assessment of the SHMS concept.

An in-house laboratory independent research program was initiated at the U.S. Army Missile Command (MICOM) to develop a Fiber Specklegram Sensor (FSS) for structural monitoring of fiber optic bobbins during long-term storage. The purpose of this effort is to develop a nondestructive measurement technique to monitor the structural fatigue of the Fiber Optic Guided Missile (FOG-M) fiber optic payout bobbin. Issues associated with the fiber optic bobbin or dispenser are inter-layer stress buildup due to multi-layer winding, structural fatigue of fiber packs during long-term storage, and mechanical defects in the bobbin due to temperature fluctuations. A bulk optics FSS has been assembled, and test results quantify the FSS as a highly sensitive strain sensor. Several fibers were characterized by using compression and temperature tests to select a fiber that would provide the most sensitivity. The Corning graded index multimode 100/140 micron fiber was selected as the sensing fiber for the FSS. The sensing fiber was successfully embedded in a miniature composite bobbin.

Delaminations between composite laminate are of particular interest because they may cause catastrophic failure of the composite structure. The ability to detect the occurrence and location of a delamination is therefore important in predicting and preventing catastrophic failures within composite structures. This paper describes a delamination detection system for a composite panel. The detection is based on the changes in the acoustic properties of the materials before and after delaminations. Extrinsic Fabry-Perot interferometric (EFPI) optical fiber sensors are embedded and attached in/on a composite panel for detection of the acoustic property changes. Experimental results are presented and discussed.

A three-step damage detection procedure using modal data is developed for space trusses. The procedure is based on determining parameters associated with member stiffness matrices using a system identification technique. The first step is the refinement of the matrix using modal data from the undamaged structure. A modal residual vector is then used to determine whether or not damage has occurred. If it has, a minimization scheme is used to determine the parameters for only those members found to have been damaged. An eight-bay, hybrid-scaled truss designed and tested at NASA's Langley Research Center is used to verify the procedure. One damage case is used to illustrate the success of the procedure.

In this paper we investigate the performance of a SCS-6/Ti-15-3 composite system, a carbon/SiC woven composite system, and an AS4/3501-6 composite system, subjected to long term mechanical fatigue, with an extrinsic Fabry-Perot interferometric (EFPI) fiber optic strain sensor. Both stiffness reduction and the degradation of thermal expansion coefficient (TEC) are monitored. The obtained results show that the EFPI sensor provides reliable data during long term fatigue loading up to 1 million cycles. The results suggest that the EFPI sensor is a viable means to monitor current and proposed characteristic damage metrics for various composite systems. We also monitor the TEC degradation during the thermal fatigue of celion G30-500/PMR-15 woven cross-ply composite system, and present a simple micromechanical model, that employs a shear lag approach, utilized to predict the effect of matrix cracking on the TEC of the composite. Results show that good agreement between experimental data and theoretical results is obtained, and that this parameter changed by as much as 80% over the fatigue life of the composite.

A smart structural health monitoring system (SHMS) requires various sensing technologies to detect and locate flaws, and assess their criticality to the structural integrity of the aircraft. To realize its full potential, a SHMS must be capable of remotely sensing flaw growth and location. Acoustic emission (AE) is one of the few sensing technologies that is capable of direct and remote sensing of flaw growth. Currently, there are two AE sensing techniques used for monitoring, detecting and locating flaw growth in structural components. In one technique, specific AE event parameters are captured by narrowband transducers and are studied to identify their source and location. The other technique studies the whole AE waves captured by wideband transducers and then detects and locates flaw growth based on waveform analysis and the wave propagation characteristics of the structure being monitored. This paper investigates both AE techniques, establishes their limitations, and defines the goals that need to be achieved in AE technology before it can successfully be implemented into a SHMS.

An experimental study is conducted to develop a technique for detecting and assessing damage in laminated composite plates using piezoceramic (PZT) and acoustic emission (AE) sensors. Test specimens of glass/epoxy and graphite/epoxy composite plates are fabricated using a hot press cure technique. PZT patches and AE sensors are surface mounted on the composite plates to serve as sensors. Low velocity impact tests of plates with all four edges clamped are conducted using a drop weight testing frame with the composites in an undamaged state and again after the composites are damaged. Two test methods are used to assess the damage of the composite plates. The piezoceramic sensor output and the acoustic emission sensor signals are measured during the impact tests. Modal testing is performed to determine the frequencies and dampings of the structures from the frequency response function. The plate is then damaged by a high velocity impact and the results are correlated with the undamaged plate data. It is found that the piezoceramic sensor output is very sensitive to the composite damage and change in modal frequencies and dampings is observed.

Future advanced fixed- and rotary-wing aircraft, launch vehicles, and spacecraft will incorporate fiber-optic smart sensor networks to monitor structural integrity and manage overall structural health. This paper describes the design of an acoustic emission (AE)-based micromachined sensor for new and aging aircraft applications to assess preflight readiness, in- flight structural integrity, and post-flight time-based maintenance. This unique sensor approach combines silicon micromachining, free-space optical waveguides, high-speed optical interconnects, and supervisory sensor management to measure stress waves related to AE events. A unique second-generation polysilicon resonant microbeam sensor design is described. It incorporates a micron-level vacuum-encapsulated microbeam to optically sense structural-integrity acoustic parameters and to optically excite the sensor pickoff. Its principal of operation, significant payoffs and benefits, and wafer-level laboratory test data results are summarized.

This paper describes a fiber optic system for detection of acoustic emission location in which extrinsic Fabry-Perot interferometric fiber sensors are incorporated in two different configurations: (1) attached to a thin aluminum panel, and (2) embedded in a graphite/epoxy composite laminate. The system computes the coordinate of the acoustic emission source using sensor recorded differential arrival times of the acoustic signals generated by the source. The impact location can be determined with a 0.5 millimeter resolution and an accuracy typically less than five millimeters.

Many standard non destructive testing systems utilize ultrasonic pulse echo techniques to localize defects within structures. These techniques rely upon the scattering of acoustic energy by the defect and two or more acoustic detectors for localization. In this paper, an alternate method of structural damage detection and location is proposed. The technique involves the use of a leaky acoustic waveguide and a unidirectional acoustic waveguide attached to the structure of interest. Only one acoustic detector is required for defect location. An analysis of the concept is provided and experimental results are presented.

A study is conducted to determine the possibility of detecting internal delamination in a tapered composite laminate using embedded ultrasonic transmitter and receivers. Sound propagation was modeled in the tapered laminate. Ply interfaces, resin pocket interfaces, absorption losses, delaminations and resin pocket cracks are included in the model. Because of the anisotropic behavior of composite materials, the speed of sound is a function of the orientation of the ray with respect to the fiber direction. This parametric study results in the optimal locations for acoustic receivers and the capability of those receivers to determine the extent of delamination via characterization of the energy versus time plot, also termed as the ultrasonic signature. The results indicate the possibility of a successful non-destructive technique for delamination detection in tapered beams.

An absolute displacement and strain measurement system has been used to characterize plastic deformation in metal specimens. Here, an absolute extrinsic Fabry-Perot interferometric (AEFPI) sensor system uses a modified EFPI sensor element head with support optical and electronic signal processing to give an output signal proportional to the instantaneous air gap length within the sensor element. This information may be used to determine the local axial strain if the gage length of the sensor is known. Because the system does not require a reference phase, the support electronics may be turned off and back on with no loss of optical fringe information or ambiguity in reference point. Applications in the long term monitoring of residual strain and creep are suggested.

We model and experimentally demonstrate a Hilbert Transform-based signal processing scheme that maps optical phase shifts in a fiber optic interferometer into electrical phase shifts in the harmonics of an imposed modulation frequency. The interferometer is a two-loop Mach- Zehnder with only one loop exposed to the measurand. A phase modulation at a convenient carrier frequency is applied to the first loop, and its second harmonic is applied to the second loop. A 30 Hz measurand test signal is applied, upshifted, and demodulated using a simple digital signal processing algorithm. In contrast to the nonlinear dependence of transmittance on phase shift of conventional interferometric sensors, this digital signal processing scheme yields a demodulated output that is linear with phase shift, and is capable of extracting measurands with dynamic ranges much greater than 2 (pi) radians.

A real-time mixed signal (analog and digital) demodulator for interferometric fiber optic sensors operating at high signal levels was designed, implemented, and tested. The inputs to this fiber optic sensor demodulating instrument are a phase modulated optical signal and an electrical carrier reference signal. This high performance demodulator has an intrinsic 84 dB linear dynamic range that is switch selectable over a 193 dB range, a demodulated signal output amplitude-frequency product of 500,000 rad (DOT) Hz (-3 dB), and an absolute accuracy to better than 0.4%. The group delay from the sensor input is 4.8 microsecond(s) to the demodulated analog output, and 2.3 microsecond(s) to the digital demodulated output. One of the unique simplifying design attributes of this arctangent type demodulator eliminates the need for jump detector/counter circuitry. The details of this design, the tests performed, and the test results are examined.

It is well documented that high void content and delaminations are observed during the fabrication of thick carbon/phenolic composites due to the high percentage of volatile release that occurs during polymerization. In order to eliminate these imperfections and enhance product quality for thick carbon/phenolic composites, smart sensors are used to monitor the property changes during the processing and control of the component to minimize this effect. This paper documents an effort to develop a technique to monitor and collect sensor data during the curing process of a general material system. Data obtained from sensors are used to generate an expert processing knowledge base which automatically controls the composite cure state based on direct sensor response, in lieu of classical time/temperature techniques (i.e. recipes). Microdielectric, ultrasonic, thermopile, thermal couple, and Extrinsic Fabry-Perot Interferometer (EFPI) sensors are investigated as potential candidates to monitor and subsequently control the manufacturing process of a composite material. In addition to classical optical sensors, a modified EFPI thermal sensor is employed to monitor temperature variations.

The key to reducing processing costs, improving product yield, and optimizing material properties in composite materials fabrication is the use of intelligent process control (IPC). IPC completes the processing feedback loop by obtaining in situ sensor information, analyzing this information to make processing decisions, and then applying control parameters to the process in real-time. Information for IPC sensors must go beyond the traditional measurements of temperature and pressure, and include data such as resin viscocity, resin position within the mold, resin gelation-point, degree-of-cure of the composite, types of polymerization reactions taking place, presence of moisture, and the like. Two light-based sensors are described which have been utilized to obtain this information: (1) a novel infrared (IR) fiber-optic sensor, developed at BIRL, and (2) a commercially available index of refraction sensor. The information provided by these sensors relates directly to material performance. This paper also describes how neural networks were used to interpret the data collected by the IR fiber-optic sensor.

The field of smart structures has been largely driven by the development of new high performance designed materials. Use of these materials has been generally limited due to the fact that they have not been in use long enough for statistical data bases to be developed on their failure modes. Real time health monitoring is therefore required for the benefits of structures using these materials to be realized. In this paper a non-contact method of powering and interrogating embedded electronic and opto-electronic systems is described. The technique utilizes inductive coupling between external and embedded coils etched on thin electronic circuit cards. The technique can be utilized to interrogate embedded sensors and to provide > 250 mW for embedded electronics. The system has been successfully demonstrated with a number of composite and plastic materials through material thicknesses up to 1 cm. An analytical description of the system is provided along with experimental results.

Due to their small size, light weight, geometric flexibility, and their possible uses for monitoring various different physical parameters, optical fiber sensor technology offers numerous opportunities and advantages for instrumenting structures and materials for their analysis and control. While optical fiber sensors have advantages over conventional electric sensors, connectorization of optical fiber sensors to the supporting laser, detector, and electronics has proven to be difficult. In this paper we demonstrate the possibility of having an optical fiber sensor system that can be completely embedded within composite materials. The fiber sensor as well as the supporting opto-electronic components required to launch and receive light into and from the fiber could be embedded in a panel, avoiding connector problems. Interaction with the electronics to obtain strain and vibration measurements from the fiber sensor was made via external antenna coils.

A new strain sensing methodology for the measurement of peak strain in engineering materials has been developed. The approach involves the correlation of change in the magnetic susceptibility attendant with the strain-dependent, solid-state phase transformation in a sensing element. The sensing materials are metastable steel alloys that irreversibly transform from nonferromagnetic to ferromagnetic behavior as a function of the peak, applied normal strain. The technology makes available a reliable method for passive, semi-active, or active monitoring of strains or deflections and is applicable as either embedded sensors or attached strain gages, i.e., monitors. Strain assessment devices of various types have been conceptualized to meet a variety of needs. The history, principles, and test data on the development are presented. Discussion and results of a prototype system installed on the I-95 Savannah River bridge are presented. Examples of current projects and future applications of the technology are presented and discussed. A range of applications is discussed that illustrates the versatility of the approach and the value of such systems in materials and structural safety assessment.

A hollow core optical fiber waveguide has been designed, fabricated, and demonstrated for reinforcing, sensing strain and temperature and for healing cracks in a concrete beam. This paper discusses how hollow core and solid core optical fibers can be used to reinforce a concrete specimen, monitor parameters affecting the structure and be used to heal cracks occurring in the structure as well. Measurements of strain and temperature can be made using the liquid core optical fibers or can be inferred indirectly by collocated solid core fiber temperature or strain sensors. As soon as fiber sensors detect a crack or break in the concrete structure, a commercially available concrete restorer liquid is pumped through the hollow core fibers embedded in the structure and `heals the structure.' The healing of the concrete structure and the measurement of strain and temperature using liquid core optical fiber sensors is discussed in this paper.

Electrophysical properties of highly filled fiber polymer composites, containing disperse conducting or ferromagnetic fillers, are examined. The methods for controlling their electrical, magnetic, and other properties by changing the ingredients and compounding methods are reported. Some examples of using polymer composites in the form of functional elements or structures, prospective for different fields of science and technology, are presented.

Over the past 25 years in the United States there have been more than 85 collapses of structures under construction that have been directly attributable to formwork failure. Sensing systems and techniques applicable to the monitoring of construction site shoring and scaffolding are designed and implemented with preliminary systems being used in the field and in the laboratory. Such a sensor network can provide significant information about the load distribution on shoring systems. This information can allow dangerous situations to be quickly identified so that corrective action can be taken. Furthermore, the load data acquired with this system can be used to formulate improved construction codes that enhance construction work safety. Laboratory proof-of-concept experiments as well as actual field site measurements presenting the in-service use and capabilities of an intelligent shoring system are described in this paper.

This paper describes initial results of a fiber optic-based sensor during on-site testing performed by FEORC and Fiber and Sensor Technologies at Ingersol-Rand. Advantages of the fiber optic sensor are a demonstrated enhanced survivability, higher sensitivity, smaller size, electromagnetic interference immunity, and reduced risk of explosion. The conventional wire strain gages typically survive only a few minutes attached to the drill steel and drive chain, while the fiber sensors described here have survived over 400 hours and are currently still functioning properly. The tests described include the demonstration of strain energy measurements on the drive chain and drill steel, and displacement measurements of the piston within the drifter. All of the sensors tested can be used as both a laboratory evaluation and testing tools, as well as being an integral part of a proposed control and health monitoring system.

A method for measuring lateral displacements in structures is proposed based on the motion of a Gaussian optical beam. A sensor is designed and constructed using the free space output of a single mode optical fiber, which well approximates the Gaussian intensity distribution, and is tested in both quasi-static and dynamic displacement modes. Good agreement between the experimental data and the predicted output is obtained for small displacements of the size expected in real buildings. It is envisioned that this sensor could serve as the input to an active control system used to stabilize smart buildings experiencing earthquake and wind loads.

Intracore fiber optic Bragg gratings have many attractive features that make them well suited for smart structures. The Bragg grating laser sensor makes optimum use of the optical energy and a 4-channel system has recently been demonstrated in connection with a fiber optic sensing system installed in a new highway bridge. In this paper we introduce the new concept of Bragg `intra-grating' sensing whereby additional sensing information from within the grating can be extracted. We show theoretically that the presence of a strain gradient across a Bragg grating distorts its reflection spectrum and have verified this with a preliminary experiment. We also show that debonding of the Bragg grating from its host structure can be detected by changes in the reflection spectrum providing a warning of sensor malfunction.

Fiber optic sensors were embedded into a 7.5 MW hydroelectric dam during its construction phase. Power generation began at the facility during the spring of 1993 as did initial verification of the on-line use of certain key embedded sensors. Reliability information, as well as structural vibration signatures for the dam, has been obtained under various operational regimes. This information as well as the initial perspectives from builders, owners, users, and regulatory officials regarding embedded sensors is presented.

The installation of a fiber optic Bragg grating strain sensor network in a new road bridge is described. These sensors are attached to prestressing tendons embedded in prefabricated concrete girders. Three types of prestressing tendons are being monitored: conventional steel strand and two types of carbon fibers reinforced plastic tendons. Sensor durability issues are reviewed and the installation is described. Initial measurements indicate that the sensors are operational and provide some early comparison of tendon performance.

The development of a phenomenologically based phase-strain model for structurally embedded polarimetric sensors is presented and then applied to Bow-Tie and e-core high-birefringence optical fibers and a standard circular core low-birefringence optical fiber. This phase-strain model is developed by postulating the existence of a fictitious residual strain state in the optical fiber core, and then following standard model development techniques. Calibration procedures are used to determine all parameters required by this model.

Attenuation spectra of single mode fibers, subjected to periodic spatial deformations, have been investigated. The transmission spectra of the deformed single mode fibers show attenuation bands. These bands are a function of the period of the microbends, fiber radii, and cutoff wavelengths of the fibers. Mathematical relations have been developed to predict the location of the attenuation bands in the wavelength domain for specific microbending periods. Several standard single mode fibers were characterized for an increasing series of spatial deformation periods. Characteristic curves are plotted indicating the variation of parameters of the transmission spectra with change in the number and amplitude of the spatial deformation periods. Sensing applications utilizing the attenuation bands at specific spatial periods are presented and a multiplexed system is demonstrated.

Standard Michelson interferometric optical fiber sensors are used to make strain measurements on a carbon/carbon composite beam at temperatures as high as 800 degree(s)C. Copper, titanium, aluminum, and combinations of the three are investigated as sputtered end mirrors for the Michelson sensor, and it is found that the copper mirrors maintained the highest reflectivity in the 100 degree(s)C to 1000 degree(s)C temperature range. All measurements indicate that the optical fibers become more brittle with prolonged exposure to high temperatures.

This paper demonstrates acoustic speed measurements in anisotropic composite materials using a fiber-optic extrinsic Fabry-Perot interferometric sensor (EFPI). Acousto-ultrasound technique is used to generate unidirectional surface acoustic waves in a multilayered composite specimen. The principle of operation and the fabrication of the EFPI sensor are explained. The composite specimen is interrogated by a piezoelectric transducer driven by 1.2 MHz signal pulses from an rf generator. The acoustic speed is calculated by noting the difference in the arrival times of the acoustic signal detected by the sensor for different locations of the piezoelectric source separated by a known distance. The possible variation of the acoustic signal speed with respect to the direction of the fibers is studied. This study could be used in determining the dispersion curves of materials and impact locating detection in composite materials.

We discuss fatigue test results using extrinsic Fabry-Perot interferometric (EFPI) strain gages on a F-15 fighter within a full-scale test frame at the Structures Test Facility, Wright Patterson Air Force Base, Ohio. A linear array of EFPI sensors were surface-mounted to the trailing edge of the F-15 wing to monitor strain concentration along the inboard flap hinge. Two types of EFPI strain gages and strain gage support systems were employed; linearized output from the standard, differential EFPI sensor system determined dynamic strain on the flap hinge during the application of load, whereas, absolute EFPI strain sensors were implemented to measure residual strain or cracks in the aluminum hinge.

This paper is concerned with the control of responses from linear and nonlinear systems. In the first part, the recently proposed method of pseudo-fault induction is reviewed. It is shown that a linear system can be made to respond to harmonic excitation as if a nonlinearity of given type and location were present. The method makes use of additional or auxiliary excitation. In the second part of the study, the method is extended to cover the introduction of auxiliary inputs into nonlinear systems. A particular application is described, i.e., the suppression of super-harmonic components from sinusoidally excited nonlinear systems. In both cases, the theory is illustrated using results from numerical simulation.

This paper describes a feasibility study into developing an intensity-based corrosive environment sensor. The sensor concept takes advantage of the reduced reflectivity that accompanies the development of aluminum oxide on thin film aluminum mirrors deposited on to cleaved optical fiber ends. The test results indicate that this technique merits further investigation.

A fiber-optic Michelson interferometer is employed for sensing the vibration of a cantilevered beam. A small section of the sensing fiber arm is attached to the beam to sense the vibration of the beam. The active homodyne technique is used to obtain an electrical output which is proportional to the vibrational signal of the beam. A closed-loop control system comprises a pair of sensors and actuators, which are mounted nearly at the same point of the vibrating body, and an inverting power amplifier. The fiber sensor and a piezoelectric actuator are co- located on the root of the cantilevered beam. The fiber sensed signal is amplified and inverted, then fed into a piezoelectric actuator for exerting a dynamic control force on the body. Experimental results show that vibration of the beam is substantially reduced by applying a single control system with the fiber-optic Michelson interferometric vibration sensor.

Optical fiber Fourier transform infrared evanescent wave spectroscopy has been used to remotely detect solid aluminum hydroxide Al(OH)₃. This compound is a principal corrosion product of aluminum. The strongest mid-infrared spectral features of Al(OH)₃ are found between 3350 cm^-1 and 3650 cm^-1. Identification of five aluminum hydroxide absorption features in this spectral region has been made. The relative positions, transition strengths, and widths of the recorded bands are in agreement with reference data. The experimental arrangement included a 4 cm^-1 resolution Michelson interferometer equipped with an InSb detector and an optical accessory to match the optical fibers to the interferometer spectrometer. Detection of aluminum hydroxide using optical fiber FTIR evanescent wave spectroscopy is an important step in the development of a technique for the remote detection of corrosion products of aluminum.