WO1991019173A1

WO1991019173A1 - Method for assessing structural integrity of composite structures

Info

Publication number: WO1991019173A1
Application number: PCT/US1991/003857
Authority: WO
Inventors: Gino A. Pinto; Richard E. Hayden; Charles S. Ventres
Original assignee: Technology Integration And Development Group Incorporated
Priority date: 1990-06-01
Filing date: 1991-05-31
Publication date: 1991-12-12
Also published as: AU8508491A

Abstract

Disclosed is a method and system for non-destructively assessing the structural integrity of composite structures (10). The present invention makes continuous or periodic quantitative measurements of changes in structural stiffness and structural damping and relates these changes to the effective degradation of structural integrity (24) of the composite structure (10). The composites monitoring system employs vibration methods (12, 14) to monitor the changes in stiffness and damping properties of the structure (10) being monitored. Frequency tracking (22) of structural resonant modes of vibration along with compensation for environmental effects (16), can readily yield changes in stiffness of the composite structure (10). The flexible system allows for algorithms describing any composite fatigue or damage behavior characteristic to be provided as an input to the system.

Description

METHOD FOR ASSESSING STRUCTURAL INTEGRITY OF COMPOSITE STRUCTURES

Background - Field of Invention

The present invention relates to a method and system fo -nondestructively evaluating the structural integrity of a mechanical component constructed from fiber composites, specifically for assessing the stiffness, strength and damping characteristics of a composite structure.

Background - Description of Prior Art

Nondestructive evaluation techniques provide a viable means for prediction of damage due to cyclic fatigue and environmental exposure of critical structures. Of particular interest is the degradation of strength which occurs throughout the life of the structure due to normal loading cycles. Unexpected damage such as impacts, delaminations, drastic environmental changes and loading may also occur throughout the life of a composite. It is vital to monitor the structural integrity of these critical structures during in-service operation and/or provide routine inspection while out of service. The realized benefits of a composites monitoring system are: improved safety/reliability of structures, reduced cost in developing and maintaining structures, and a means to make critical decisions on the reliability of aging composites structures.

Composite materials are known to fail in a progressive manner through gradual deterioration (damage development) in the material caused by cyclic fatigue loading, impact loading and/or environmental exposure. Unlike metals, composite materials are inhomogeneous and quite often anisotropic. This results in a very complex fatigue process that reduces stiffness and strength while increasing the structural damping of the composite. Also, the stages of the damage development are highly dependent on the construction of the composite and the applied loading. The unique properties of composite structures make the process of predicting safe operating life quite difficult using conventional nondestructive evaluation techniques. These factors present a unique challenge for a versatile composites health monitoring system that effectively determines if any form of structural degradation has occurred, then concludes whether or not the component is suitable for service.

There are several options available for the nondestructive testing of composite structures. These nondestructive testing (NDT) methods often require the involvement of experts in the field of NDT to periodically inspect the composite structures while they are out of service. The more popular composites NDT options are: Radiography, Acoustic Emission, Thermal Methods, Optical Methods, Corona Discharge, and Chemical Spectroscopy. Because of the specialized equipment involved, it is not practical to use either radiography or chemical spectroscopy outside of the laboratory environment for continuous in-service monitoring.

Acoustic emission is a method where sensors are placed on a composite to "listen" to low level sonic or ultrasonic signals generated by damaged or degraded materials and is a candidate method for in-service monitoring. The main problems with Acoustic emission is that the acoustic levels received at the sensor are dependent on the load applied to the structure. Acoustic interference from the operating environment (i.e., noise) may also be a problem for an in-service monitoring system relying on this technique.

Thermal methods rely on thermal imaging of stress generated heat fields within a composite. Cyclic loading induces heat generation in a material due to fatigue cracks or hysteresis within the matrix material. This method cannot be reliably used for continuous monitoring in many cases because of the sensitivity of the thermal sensors to other heat sources.

Of the optical methods, fiber optic methods are gaining the most attention for in-service monitoring. Light conducting fibers can be integrated into a composite structure during its manufacturing process. Fractures, cracks or delaminations are detected when these fibers are broken or damaged. The main problems with fiber optic sensors are that this method cannot be implemented into existing composite structures and damage where no optical fibers exist cannot be detected.

The corona discharge method can be used for detecting voids within a material, where an electric field of high intensity is imposed in a dielectric composite material. The void is detected by a minute pulse of current or by the radiation of electromagnetic wavelengths as a result of electron collision within the void walls. The two main problems with this method are: (1) it is not practical to apply this method on a continuous in-service basis because of the risks and electromagnetic interference, and (2) it cannot be applied to conductive materials such as carbon (graphite) and metal fiber or metal matrix composites.

Development of damage within a composite can be empirically related to its stiffness reduction and in most cases so can residual strength reduction and an increase in structural damping. These damage relationships provide the link between the complex micromechanical damage mechanisms and the resulting macromechanical response of the composite (static or dynamic) . For the present invention, it is necessary to establish a mathematical relationship (model) that relates the stiffness degradation of a fatigued composite to the residual strength degradation and/or to the fractional life expended. Analytical or empirical relationships can be developed for specific types of laminates and specific structural components. Monitoring of structural damping may also be used as a damage indicator. The accuracy with which damping is resolved depends greatly on the method of measurement (i.e. contact points between part and exciter/receiver, changing boundary conditions of part, etc.). A major disadvantage of using damping measurements as an indication of damage is that the measurements are not always reliable. For this reason an appropriate decision weighting function must be placed on damping measurements for each particular composite structure.

Summary of the Invention

Since most composites exhibit a measurable reduction in stiffness and increase in damping over their useful life, monitoring changes in stiffness and damping can provide an effective nondestructive means for predicting the composite's remaining life. The present invention determines the composite's stiffness by tracking the frequency of a mechanical resonant mode(s) of a composite element and relating frequency changes directly to a change in stiffness. The present invention also provides a supplemental nondestructive test by monitoring structural damping of these modes. The composite integrity monitoring system of the present invention contains a database having predetermined (analytically or empirically) stiffness vs. strength relationships to infer a reduction in strength. Other environmental sensor information can also be used to compensate for any environmental frequency-influencing effects.

An observable feature resulting from degradation of composite structures is the reduction in the resonant frequencies of structural modes of vibration, -be they longitudinal, flexural, torsional, or shear modes. This resonant frequency reduction is directly related to the reduction in the composite's stiffness (or increase in compliance). The present invention obtains a quantitative change in stiffness of a structure by measuring a change in resonant frequency of the structural mode(s). Relationships can be empirically established between a composite structure's measurable stiffness reduction and a corresponding reduction in strength or increase in damage. An increase in damping is also an indication of an increase in damage. The structural integrity monitoring system for composites tracks shifts in stiffness and damping, compensates for environmental effects, and assesses the structural integrity of the composite by virtue of empirically or analytically defined relationships and criteria for a particular composite structure.

The method in which the present invention tracks the dynamic stiffness (modulus) degradation of a composite element is by monitoring changes in the frequency of a structural resonant mode of vibration. As an example, the fractional reduction in frequency, of a mode of vibration (bending or longitudinal). caused by a given modulus reduction is given by the following equation (valid for both beam and plate elements) :

where f₀ and E₀ represent the initial resonant frequency of the mode being tracked and initial modulus of the element respectively.

The significance of this relationship means that the measured reduction in normalized resonant frequency is a constant for all modes of vibration. Tracking frequency shifts of any given resonant mode will therefore readily yield the composite's residual modulus. This is true provided that the damage is distributed uniformly throughout the entire element. If the damage is localized (such as an impact or a delamination), the mode shapes will change, some modes will be damped more than others, and their associated resonant frequencies may not be simply related to the overall stiffness reduction. Tracking more than one resonant mode may therefore be necessary to identify local damage.

The present invention gives two options for tracking the resonant modes of a composite element. The first option is a passive sensing method where data is collected from a sensor or sensors and the time domain vibration history is transformed into the frequency domain spectrum via a process such as a Fast Fourier Transform (FFT). The modal parameters of frequency and damping can then be determined through a spectrum analysis procedure.

The second option for the present invention's method is an active sensing/excitation method. The actively excited composite structures are designed to use a sensor and actuator feedback system to cause a self-excited resonant mode of vibration; where a sensor (or sensors) and an actuator (or actuators) are used to simultaneously sense and excite a selected resonant mode. The sensor senses a response (i.e. displacement, acceleration, strain) and provides a signal (proportional to the response) that is amplified and fed back to the actuator to produce an excitation force or bending moment in the same structure, thus causing a self-excited mechanical resonance. A frequency counting device is used to track the frequency of the resonant mode while a voltage sensing device is used to infer structural damping within the composite. For either case, the measurements can be made by using integral sensors or actuators attached to or embedded in the composite structure at pre-selected locations during or after the manufacturing stage. This method is therefore used to track the mechanical resonant frequencies of the composite's structural modes and also provide a measurement for inherent damping within the structure.

Both passive and active experimental modal techniques can be selected to measure stiffness degradation and inherent structural damping. Once damage relationships have been established, the measured stiffness properties can be related to degradation in structural integrity (i.e. strength loss or reduced life expectancy). These methods will then provide the means for a system to nondestructively predict the degradation of the composite's strength or its life expectancy for in- service monitoring or periodic inspection.

The passive modal methods offer the opportunity to use existing broadband excitation, provided by the structure's operating environment, to excite all resonant modes. The actively excited structures offer the benefits of "self-excitation" of selected resonant modes, real-time tracking of resonant modes, simplified signal processing, and simplified system architecture.

The premise for the system of the present invention utilizing these modal methods is that the stages of damage development are well characterized for the composite structure under known loading conditions, and suitable damage relationships can be found empirically. This system can then utilize these known statistical strength degradation/stiffness reduction and damage/damping relationships in order to determine structural integrity. Stiffness changes are detected through measurement of changes in resonant behavior of the structure and strength loss is inferred through statistical relationships established after a particular material is developed and tested. These vibration measurements can be easily performed in an on-board in-service system or used as a periodic inspection technique for any type of structure or component. The system is capable of returning a health status report, giving probability of composite failure due to the measured stiffness reduction and increase in damping.

Objects and Advantages

It is a principal object of the present invention to provide a system and method for nondestructively evaluating the structural integrity of a composite structure.

It is further an object of the present invention to provide a system and method for automatic in- service monitoring or periodic inspection of the structural integrity of a composite structure. Still another object of the present invention is to provide a system and method for providing continuous monitoring of changes in modal frequencies and other parameters (such as temperature and strain), and making computations and corrections for the effects of any frequency influencing parameters.

Yet another objective of the present invention is to provide a system and method for sensing and excitation of several resonant modes for added assurance and reliability of predictions made on structural integrity.

Other and further objects of the present invention will be apparent from the following description and claims, and are illustrated by the accompanying drawings, which by way of illustration, show the principles of operation, and what are now considered to be the best modes in which to apply these principles. Other and different embodiments of the invention incorporating the same or equivalent principles may be used, and various structures and materials may be considered as desired by those skilled in the art without departing from the invention.

Ramifications and Scope

Furthermore, the method and system for assessing structural integrity of composite structures have the additional advantages that a quantitative measurement of the material stiffness and damping properties coupled with redundant sensing and statistical techniques can be used to ensure a high degree of reliability and a low false-alarm rate; it can be used to monitor any stiffness critical component; it can be used to monitor mass changes (or density changes) in structural components; it can be used for in-service condition monitoring of aircraft structural components; it can be implemented into an on-board system for long-term monitoring of manned and unmanned spacecraft and orbiting structures; it can be used to monitor the structural integrity of helicopter rotor blades and advanced turboprop blades; it can be used for engine structure monitoring and mechanical fault isolation; it can be used for on-board monitoring of flutter in aircraft components; and it can be used for any ground based inspection of composite structures.

Although the description above contains many specificities, these should not be construed as limiting the scope of the invention but as merely providing illustrations of some of the presently preferred embodiments of this invention. Thus the scope of the invention should be determined by the appended claims and their legal equivalents, rather than by the examples given.

Brief Description of the Drawings

Figure 1 illustrates an example of a generic composite structural integrity monitoring system's functions.

Figure 2 gives an example of a system utilizing the passive sensing method for monitoring of structural integrity.

Figure 3 gives an example of distributed sensing and actuation applied to the active feedback method for modal excitation of a composite structure. Figure 4 gives an example of single point excitation/sensing applied to the active feedback method for modal excitation of a composite structure.

Detailed Description of the Drawings

Figure 1 illustrates an example of a generic composites-health monitoring system that can incorporate the passive modal tracking method or the active modal excitation/tracking method. The composite structure 10 is outfitted with sensors 12, for a passive approach, or sensors 12 and actuators 14 for an active approach. For either method, passive or active, the sensors are used to sense strain or motion and may be non-contacting types (e.g., laser vibrometer or optical sensors), surface mounted or embedded types. Environmental sensors 16 can be placed near or on the structure 10 if it is necessary for compensation of environmental effects. Signal conditioning 18 such as pre-amplification, powering, signal summation, signal differencing, amplification, filtering or phase control is performed on sensor and actuator signals if necessary. The sensor signal(s) 12, conditioned or otherwise, are then passed on to the signal analyzer 22. The function of the signal analyzer 22 is to resolve the frequency (or frequencies) of a distinct mode (or modes) of resonance and/or resolve the structural damping of the composite structure 10. The structural integrity model 24 is an algorithm used to infer the structural degradation of a composite. The structural integrity model 24 takes, as an input, either the changes in resonant frequency of a structural mode, the structural damping, or both. It can also compensate for changes in frequency and damping given by the environmental model 20, if necessary. Since the system of Fig. 1 can reliably track the stiffness reduction of the composite structure 10 during its life duration, then the strength reduction may be inferred through relationships that correctly reflect stiffness vs. strength behavior for that particular composite.

Environmental influences can cause shifts in natural frequencies of a resonant structure. If necessary, the system of Figure 1 can make provisions to compensate for these effects so that their frequency changing influence is not misinterpreted as changes in stiffness caused by the structural degradation that is being monitored. An example of such environmental concerns are: changes in temperature and in-plane loads. In-plane loads change the resonant frequencies of bending modes for structural elements (i.e. plates, shells, beams) much in the same way that tension does in a guitar string. Some causes of in-plane loading are: cabin pressurization, aerodynamic loading, centrifugal forces, or variations in payload, fuel load, etc. Therefore, the loads themselves must be measured in order to correct for their effect on the shifting of the structure's resonant frequencies. Also, when a structure is heated it expands, or if constrained, tries to expand. The effect of unconstrained expansion is easily predicted (for a symmetric laminate) , and therefore easily compensated for with a simple temperature measurement (i.e., the effective modulus as a function of temperature must be accounted for). If the structure is prevented from expanding, compressive stresses are generated within the structure which cause a reduction in its natural frequencies of bending (flexural) modes. The composite structures integrity monitoring system of Figure 1 is designed to accept environmental sensor 16 data (e.g., from strain sensors and temperature sensors), provide signal conditioning 18 if necessary, and correct for the frequency- influencing and damping-influencing effects of the environment using the environmental model 20. The environmental model 20 is an algorithm used to infer the effect of these environmental influences on the structural dynamic behavior of the composite structure 10 (i.e., stiffness and damping).

An example of the environmental model 20 that is generally applicable to any cfomplex complex structure, is to assume that the natural frequency is linearly related to changes in any of several environmental sensors, say 8j_, then the resonant frequency f (due to environmental effects) can be found using:

f = f_n +- of Δ . 3 i ""

By frequently (or continuously) measuring f, along with the environmental sensor data Δotj_, it would be possible to use a Kalman Filter (or other system identification algorithm), to estimate best (least- square error) vales for f₀ (which is the current natural frequency in the "nominal environment") as well as for 3f 8αi, i=l,...N, (which are the effects of changes in the environment). As more and more data are processed the confidence (precision) with which (f₀, 3f/5i) can be estimated gradually improves. Therefore, changes over time in the nominal natural frequency f₀ are changes which cannot be attributed to environmental effects, and so must be attributed to changes in stiffness. The results of structural integrity model 24 are transferred to the storage/display control module 26 containing information on stiffness, strength, damping, and damage within the composite structure 10.

A full coverage composites structural life integrity system may need to make various measurements (such as -frequency tracking of modes, amplitude monitoring, damping monitoring, temperature and strain monitoring) that are incorporated into algorithms to track the life consumption of critical parts. Also, redundant sensing must be considered, and statistical system identification techniques developed to ensure a high degree of reliability and a low false-alarm rate. Selective weighing of these different methods and factors will result in the ability to prognosis failure and to thus schedule on-condition maintenance for parts replacement near the end of their useful life.

This flexible system allows for algorithms describing any composites structural integrity degradation or damage behavior characteristics to be provided as an input. Alert thresholds and variances are also provided as an input to this system. These inputs may be up-datable to allow for future changes/ refinements of the composites degradation behavior algorithms, thresholds, as derived from the global experience with the system or R&D results. Once acceptance criteria are established for making reliable decisions on the composite's health, statistical methods are used to determine if the variance in the composite data and variance in the data for measurement of other environmental frequency influencing parameters are acceptable for making a reliable decision on the health status of the composite. Figure 2 illustrates the passive method used by the system of Figure 1 for the tracking of resonant frequency of structural modes and/or structural damping. The composite structure 10 is outfitted with a sensor 12 or group of sensors. The distributed sensor elements can be bonded, embedded or noncontact that give an output signal in response to a strain or motion of the composite structure. Examples of such sensors 12 are strain sensors, accelerometers, velometers, or any type of displacement sensors. Filtering of the sensor signal(s) can be performed, if necessary by the signal conditioner 18. In the signal analyzer 22 the time history signal is transformed from the time domain to the frequency domain via a processing method such as a Fourier Transform. The frequency of any selected resonant mode of vibration can be identified by a peak response in the Fourier Transform (i.e., peak amplitude response). Near real¬ time performance for in-service composites monitoring may be accomplished by utilizing fast analog-to- digital (A/D) converters to capture the time history output of the sensors, followed by digital spectrum analysis. Provided that the structure is being acted on by some broad-band in-service excitation and the damping is linear, it is possible to determine the structural damping by using the half-power bandwidth technique. The damping ratio may be resolved by using a standard technique called the half-power bandwidth method. The results of frequency shift and/or damping can then be passed to the structural integrity model 24 and environmentally compensated, displayed or stored as illustrated in Figure 1.

Figure 3 illustrates an example of a composites monitoring system that utilizes the active frequency tracking technique. The composite structural element 10, is outfitted with a sensor or sensors 12, that can be a strain sensor, accelerometer, velometer, displacement sensor, non-contact optical sensor or any other vibration sensor. The signal from the sensor is passed to the signal conditioning circuit 18. The signal conditioning circuit 18 provides rate feedback (phase lag between the sensor inpu (s) 12 and the actuator output(s) 14) to ensure self sustained oscillation of the structure. The signal conditioning circuit 18 can contain filtering capabilities 28 such as a low-pass, high-pass or band-pass filter and/or phase control of the signal such as an inverting amplifier. The filtered signal is then passed to an amplification circuit 30, that amplifies the signal and may have the additional feature of providing additional phase control. The signal is then passed on to the actuator 14, that can be a piezoelectric pad or shaker, electrodynamic shaker, acoustic speaker or any other strain or vibration inducing device. This simple feedback circuit is sufficient to cause the element to which the actuator and sensor are bonded to 10, to achieve a state of self induced mechanical resonance of a selected mode. Mode control can be accomplished either through spatial distribution of the sensor(s) 12, actuator(s) 14, or both, selective band-pass filtering, or any combination thereof. A frequency counting device 22, is used to track the resonant frequency of the composite element 10. The results of frequency shift and/or damping can then be passed to the structural integrity model 24 and environmentally compensated, displayed or stored as illustrated in Figure 1.

Figure 4 illustrates another example of a composites monitoring system that utilizes the active frequency tracking technique. The composite structural element 10, is outfitted with a device that can function as an actuator and sensor 32, such as an impedance-head shaker. This device can produce an output sensor signal 12, that is proportional to acceleration, velocity, displacement, strain, or force. The signal from the sensor is passed to a signal conditioning circuit 18. The signal conditioning circuit 18 provides rate feedback (phase lag between the sensor input 12, and the actuator output 14) to ensure self sustained oscillation of the structure. The signal conditioning circuit 18 can contain filtering capabilities 28 such as a low-pass, high-pass or band-pass filter and/or phase control of the signal such as an inverting amplifier. The filtered signal is then passed to an amplification circuit 30, that amplifies the signal and may have the additional feature of providing additional phase control. The signal is then passed on to the actuator 14, that can be piezoelectric pad or shaker, electrodynamic shaker or any other strain or vibration inducing device. This simple feedback circuit is sufficient to cause the element to which the sensor/ actuator device 32, bonded to 10, to achieve a state of self induced mechanical resonance of a selected mode. Mode control can be accomplished through selective band-pass filtering. A frequency counting device 22, is used to track the resonant frequency of the composite element 10. The results of frequency shift and/or damping can then be passed to the structural integrity model 24 and environmentally compensated, displayed or stored as illustrated in Figure 1.

Tracking the resonant frequency of only one mode will suffice to get an assessment of the reduced distributed stiffness. Tracking the frequencies of more than one mode may give added redundancy and the ability to detect localized damage. This can be accomplished by having a sequence of signal conditioning steps that cause the structure to shift from one self-excited mode to another. A frequency counter can be used to track frequency shifting of a single self-excited mode thus eliminating the need to perform an indirect transform of a captured time history signal into the frequency domain (i.e. through FFT processing). This allows for changes in resonant frequencies to be tracked in real-time (a desirable attribute for in-service monitoring).

The active excitation methods of Figures 3 and 4 provide the opportunity to measure the change in structural damping by measuring the voltages produced by the sensor pad and applied to the driver pad. The ratio of the measured sensor voltage to the applied driver voltage is inversely proportional to the structural damping within the composite.

Claims

What is claimed is:

1. A system which monitors a change in frequency of a structural resonant mode of vibration of a composite element and relates changes in this resonant frequency to a corresponding change in stiffness.

2. The system of claim 1 where the measured stiffness change is used for the purpose of quality assurance testing during the manufacturing of composite materials.

3. The system of claim 1 which uses the measured stiffness change to infer the corresponding change in strength or the corresponding increase in damage for said composite element/structure.

4. The system of claim 3 which gives notice for caution, concern, or warning once it has sensed a significant decrease in strength or increase in damage in a composite structure.

5. The system of claim 1 that employs an active method for the excitation and frequency tracking of resonant structural modes.

6. A system that employs claim 5 utilizing distributed sensors and distributed actuators on a composite structure to actively excite and track the frequency of resonant modes for the purpose of nondestructive evaluation, using a change in frequency to infer a change in structural integrity.

7. The system of claim 5 that assesses structural damping of a material by tracking changes in voltage amplitude or voltage ratios of an active resonant element.

8. A system that employs claim 7 utilizing distributed sensors and distributed actuators on a composite structure to actively excite and track the inherent damping in resonant modes for the purpose of nondestructive evaluation. Using a change in damping to infer a change in structural integrity.

9. The system of claim 1 that employs a passive sensing method for tracking the frequency of a resonant structural mode of vibration.

10. A method for inferring strength degradation of a composite material by relating a measured stiffness degradation to pre-established relationships for strength degradation vs. stiffness degradation for a particular composite material.

11. A system that compensates for environmental effects on the measurement of stiffness degradation. Environmental frequency influencing parameters such as pressure, strain, temperature, moisture, etc.