A basic review of the traditional method of damage detection and some issues will be listed in the introduction below. Moreover, an idea is going to be presented and aimed to use several types of embedded optical strain sensor to set an advanced technique called in-situ strain-based fibre optic damage detection assessment system (FODDAS) which can address and improve detection of damage in the composites. Classic fibre optic sensors and fully distributed sensor network system will be discussed in the literature review. In additional to, fracture of the optical fibre is supposed to be identified as an approach related sensor network to detect damage, and an example for detecting matrix crack by the harm of optical fibre will be pointed at the last part as well as advantages of the method.
Fibre reinforced composites are consisted of and composited of reinforced fibre materials, such as glass fibre, carbon fibre, aramid fibre, and the matrix materials through winding, composites moulding or extrusion moulding process. However, there are some unexpected damages in the fibre reinforced composites. For instance, micro-crack is a kind of those damages in the fibre reinforced composites, which are created greatly under impact conditions and fatigue, should be healed before fast crack propagation and catastrophic failure happen. Therefore, it is essential that the methods and technologies should be created or improved to monitor the damage in the composites. Due to the barely visible impact damage usually existing in the composites, a competent technology of non-defect test for damage determination and continuous material structural integrity monitoring has been considered as the professional method to detect damage. At present, most conventional NDT technologies, such as visual inspection, thermography, ultrasonic C-Scan and X-Ray radiography are restricted as they have to test structural components of complex geometry which are taken out of service for the considerable length of time because of post-damage check and evaluation. For continuous and in-situ monitoring of integrity structures, a method of using strain gauges shows great further potential. However, strain gauges are susceptible to electrical and electromagnetic effect in addition to the damage. As for the issues, how to find a feasible method to improve damage detection techniques have to start from following representative modes of damage in the composites.
According to previous research (Reifsnider K L,1991), there are two typical modes of damage in the fibre reinforced composites. Figure 1 shows a series of crack situations, which are happening when a unidirectional fibre reinforced composite (Figure 1 a) is suffered from tensile loading. In this case, the failure strain of matrix is presumed to be importantly higher than that of the fibres. Therefore, with increasing tensile loading on the composites (Figure 1 b), a reinforced fibre will break at its weakest position anywhere along its length. Obviously, a redistribution of stress will take place with additional stress being brought to bear on nearby fibres. This failure mode is a characteristic process in most fibre reinforced composites. Other processes (Figure 1 c) which will take place in the vicinity of broken fibre are interfacial debonding or matrix yielding. The last process (Figure 1 d) represents the position where the fibre debonding shown in the Figure 1 c has been transformed to a longitudinal crack along
the direction of fibres and become a plane thickness crack as well.
Figure 1: schematic illustrations of damage mode 1 (Reifsnider K L,1991)
Above mode is the same complexity as when the failure strain of fibre is higher than that of the matrix, which shows in Figure 2.
Figure 2: schematic illustrations of damage mode 2 (Reifsnider K L,1991)
Basically, the consequences of all types of damage in smart structures are part or whole changes in strengths and stiffness. Fibre optic strain sensors can measure these changes through one of the optical properties, such as intensity, wavelength, phase or state of polarization. Additionally, fibre optic strain sensors can be integrated into a present composite structure and form the kind of superb smart structure so as to access the interior of material where other sensors or devices cannot probe. The research was directed by Bhatia V (1995) and gathered some results, which have indicated that embedded fibre optic sensor carried through the same way as either non-embedded sensors in terms of failure stresses in tension and compression, notwithstanding the stress and strain concentrations around the embedded fibre optic sensor. As for detecting damage, embedded sensors should have appropriate mechanical bonding with the host composites structures so as to existing the same strain gradient situation as host composite structures.
The present essay intends to focus on using embedded fibre optic strain sensors to establish a technical system, which can develop in-situ damage detection and assessment systems. The major issues needed to be solved are an in-situ detection of complex structural damage and damage state, which will be addressed via in-situ strain-based fibre optic damage detection assessment system (FODDAS).
2. Literature Review
As for improving damage detection, the FODDAS should be enhanced via various major types of fibre optic sensors and related optical properties which are helpful to measure mechanical strain. The general optical fibre consists of a centre silica core covered by an annular silica cladding with an outermost protective coating, as shown in figure 3. Main fibre optic strain sensors are categorised into intensity, interferometric and polarimetric sensors in terms of which optical properties are inflected by the external loading. Due to their fibrous characterisation, they can defiantly measure axial strains. For measuring mechanical strains, there are two aspects of requirements to fibre optic strain sensors. One is that fibre strain sensors should be too adjacent the damage to gain the reliable data and responses. Another point is that they have to guarantee an adequate strain resolution, in case that the positions of the damage are far away from them. In many applications (Tay A K,1990), one-loop serpentine fibre optic sensors were used to increase the degree of sensitivity. Furthermore, optical fibres have a critical value of failure strain as the same level as that of the reinforced fibres of the host composites. Obviously, if the failure of the optical fibres is used to indicate the damage, indeed, it is unnecessary for the strain measurement.
Figure 3 Schematic of the general optical fibre (Sim L,2002)
2.1 Intensity sensors
As everyone knows, the simplest fibre optic sensor is based on a kind of light-intensity modulation in a multi-mode optical fibre. In its working process, the light wave with greater intensity is going to spread, Meanwhile, when the optical fibre is under strained state, it will reflect losses of light intensity. An advantage is that these sensors are simply to establish and do not present complex instrumentation and signal processing. However, they can only gather limited data and information about the location of damage. For this drawback, if a large number of intensity sensors can be used in a complete system or a network (Kulman R,1987. McBride R,1998) for detecting significant damage, they could be useful and practical because of the low cost in the co-operation system or network.
2.2 Fibre Bragg Grating (FBG) sensor
The fibre Bragg grating sensor is based on a single mode optical fibre, and has a suit of periodic reflective Bragg gratings along with the length of the individual fibre (Udd E,1996), as illustrated in figure 4. When an embedded FBG sensor is impacted by the external load, the changes in the wavelength of the gratings could be defiantly associated with mechanical strain. During the producing of FBG sensors, the pitch spacing of every grating is able to control individually so that the wavelength shift of a set of individual FBG sensors can be tracked simultaneously for multiplexing. According to an existing research (Liu T,1998), the positive aspects of FBG sensors are easy to use wildly, unaffected by discrepancy calibrations, and having a high value of failure strain, up to 2%. Additionally, the sensors are also easy for multiplexing. On the other hand, the potential drawbacks that the measured strain could be three-dimensional features so as to that a detailed analysis of the output is necessary to correctly work out axial strains, and that they are very sensitive to the influence of temperature variations.
Figure 4 Diagram of Bragg Grating Sensor (Sim L,2002)
2.3 Polarimetric sensor
The polarimetric sensor is a special optical fibre based on an elliptical core duel mode with birefringent polarisation-maintaining. Basically, the birefringence will be generated by the remanent strain field across the core in addition to that could be induced by an asymmetry character of the core geometry. To specifically, the localised sensing region of a polarimetric sensor usually is made by either two in-line splices (Figure 5 a), or one in-line splice with a reflecting mirror in the end (Figure 5 b). It is worth noting that this type of sensor costs too much and has a complex sensing system. Moreover, low axial strain sensitivity and three-dimensional feature of measured strains will sometimes get the process of detection into trouble. whereas the excellent character of high transverse sensitivity will bring the sensors more help
to confirm the location of impact.
Figure 5 Schematic of (a) in-line splice and (b) single-ended polarimetric sensor (Sim L,2002)
2.4 Fibre optic sensor system or network
To roundly inspect the damage in the smart composites, it is necessary that the multiple strain value is measured and determined. As for how to achieve the complex and multivariable measurements, mentioned fibre optic detection damage sensor system (FODDAS) is able to provide such technique. According to a research (Culshaw B,1985), A direct way to set a sensor network system is easily using enough number of orthogonal arrays of intensity optical fibre at different ply interfaces to form a composite structure. Consequently, it is possible that the impact location, damage severity and distribution could be estimated as optical fibre across the depth of the smart structures. An existing research (Rogers A,1999) results showed a type of multiplexing technique called quasi-distributed fibre optic sensor network system. As the Figure 6 shows, the individual sensor in the network system can be posed into either series topology (see Figure 6 (a)), parallel topology as fleshed out figure 6 (b) or a combination of both as presented in figure 6 (c). Another idea which the sensing location can be at any points along with the direction of the optical fibre called fully distributed sensor network system. When the external load causes the optical fibres change such as pressure or tensile in the extent of the reflected signal, the mechanical strain will be detected by mentioned multiplexing techniques. In general, this type of sensor network system is considerable to provide guarantees in the further damage assessment and detection.
Figure 6 Schematic of multiplexed sensor system in (a) serial topology, (b) parallel topology and (c) serial and parallel combined arrangement (Sim L,2002)
2.5 Method of detecting damage
Damage usually exists in the form of three states, such as matrix cracks, delamination and fracture of reinforced fibres. One of the most straightforward methods of detecting damage is based on whether the embedded sensing fibres are to fracture or not. In essence, fracture of optical fibre approach is a reliable technique to inspect damage at present. For example, matrix crack or fracture of reinforcing fibre could lead a part or complete fracture, especially, of the nearby optical fibre. This feature is very useful in some applications where it is possible to know whether the composite is under sustained impacts above the thresholds (Martin A,1997). Usually, the fibre does not have to fracture completely, whereas it suffers some level of breaking after the impacts. In a sense, the broken fibre will emit light because of the fracture, then the light can be detected by monitoring devices. It is no denying to say that this approach can only fix with one-time damage led by a transverse loading. However, the nature of this method depends on the magnitudes of local normal stress, so there is no need to concern about the quality of interfacial bonding between the fibre and host composite. The distinct advantage is that it can be applied to the smart composite with complex geometry and do not to be required to be in the all-time standby mode so as to reduce the cost of monitoring devices.
So far, many efforts to improve the methods of damage detection have been proposed and discussed. Although traditional approaches to detect damage, such as visual inspection and thermography, are basic techniques and pioneer in the area of damage detection, they still have the obvious weak aspect which cannot be applied to detect such damage with complex structural geometry. FODDAS is built in the structures of laminated composites and is based on the embedded fibre optic strain sensors, achieving multiple strain measurements. These sensors have been enhanced to the position where they have realised capabilities to be the enable technique in the development of such network systems. The core part of this technology has concentrated on detection of damage caused by loading impacts embodied in a typical damage mode of matrix crack. Moreover, FODDAS must be custom-built and have the specific structure not only with many different types sensors or networks but also with assessment models, further formulated for different application field. All in all, to make a progress on the FODDAS and bring it to be a viable and practical technology, more conclusive research and experiments must be carried in the near future.
Bhatia, V., Murphy, K., Claus, R., Jones, M., Grace, J., Tran, T. and Greene, J. (1995). Multiple strain state measurements using conventional and absolute optical fibre-based extrinsic Fabry-Perot interferometric strain sensors. Smart Materials and Structures, 4(4), pp.115-26.
Tay A K, Wilson D A and Wood R L. (1990). Microdamage and optical signal analysis of impact induced fracture in smart structures. Pp.328–43.
Kulman R, Duncan B and Claus R O. (1987). Fiber optic composite impact monitor. Proc. IEEE South East Conf. 2, pp.414–17.
Jones J D C and McBride R. (1998). Multiplexing optical ?bre sensors Optical Fibre Sensor Technology. Ed K T V Grattan and B T Meggitt (London: Chapman and Hall),4, pp 117–65.
Udd, E. (1996). Fibre optic smart structures. Proceedings of the IEEE, 84(1), pp.409-44.
Liu T and Fernando G F. (1998). The application of optical ?bre sensors in advanced ?bre reinforced composites Optical Fibre Sensor Technology. Ed K T V Grattan and B T Meggitt (London: Kluwer),3, pp.87–129.
Culshaw, B. (1985). Optical fibres in NDT: a brief review of applications. NDT International, 8(5), pp.265-8.
Rogers, A. (1999). Distributed optical-fibre sensing. Measurement Science and Technology, 10(8), pp.75-99.
Martin, A., Fernando, G. and Hale, K. (1997). Impact damage detection in filament wound tubes using embedded optical fibre sensors. Smart Materials and Structures, 6(4), pp.470-6.
Reifsnider, K.L. (1991). Damage and damage mechanics. Fatigue of Composite Materials, pp. 11–77.
Zhou, G. and Sim, L. (2002). Damage detection and assessment in fibre-reinforced composite structures with embedded fibre optic sensors-review. Smart Materials and Structures, 11(6), pp.925-939.