Structural Health Monitoring

Systems influencing structural design

“On-condition maintenance” is a relatively new concept that aircraft manufacturers are aiming to implement one day. Structural health monitoring (SHM) is an important step along the way towards this goal. SHM enables structural components of an aircraft to “sense” signs of damage, thus reducing downtime for maintenance. It is likely to have a major impact on future aircraft designs.

One group deals with invisible flows of electrons while the other builds solid structural components. Systems engineers think in terms of abstract functions, design engineers in terms of physical structures – two working methods and design philosophies that could hardly be more opposite. Two different worlds – so far, at least. In the not too distant future, however, the difference between structures and systems will not merely start to fade but virtually disappear. Structures will rely on systems to indicate their state of health, and systems will influence structures in return. The magic formula can be expressed in three letters: SHM, or structural health monitoring.

SHM is defined as the “continuous and autonomous monitoring of defects, stress/strain, environmental and flight parameters by means of permanently attached or embedded sensor systems in order to ensure structural integrity”. SHM is the nervous system that furnishes inert material with rudimentary senses, enabling it to signal “injuries” to a “brain” that not only stores this information but also interprets it and triggers the appropriate response. SHM systems thus have both active and passive functions. Passive systems merely listen to signals from the structure, whereas active systems also send signals to the structure and are capable of interpreting its response.

A technique that has already proved its effectiveness on firm ground, for instance as a way of monitoring the retaining cables on suspension bridges, is now about to conquer the air. Airbus engineers have been heavily involved in the development of SHM since the 1990s – seeing it as an enabling technology with the potential to improve the structural integrity of present and future aircraft designs, and as a means of saving weight and reducing inspection times.

Certain SHM technologies are already being employed by the major aircraft manufacturers as part of their materials testing programme. One of these is comparative vacuum monitoring (CVM), a technology that played an important role in the acceptance testing of GLARE (GLAss-fibre REinforced aluminium) composites. GLARE is a laminate material consisting of three layers of aluminium held together by intermediate layers of glass-fibre-reinforced epoxy resin. Among other applications, the sandwich structure developed by Airbus has been used in the construction of the A380 – the upper part of its outer skin is made of it. Like any other new material, GLARE had to undergo extensive testing to obtain the approval of the air safety authorities before it could be incorporated in an aircraft.

And as with any other composite material, structural integrity was one of the major concerns during the development of GLARE. How stable is the bonding between the individual layers? Does the outer layer flake off, and do the layers show signs of coming apart (debonding, delamination)? Are there inclusions? And last but not least, how much stress can the rivet holes withstand? Are they the starting point for crack growth due to fatigue? Answers to certain of these questions could be found using CVM in a fraction of the time required by conventional testing methods, partly because the CVM sensors were permanently attached to the material. Each measurement could be reproduced under exactly the same conditions. The onerous task of installing and removing sensors only had to be carried out once, which saved a great deal of time, especially in the less easily accessible parts of the airframe.

In order to carry out crack tests on the riveting points, the CVM sensors were positioned inside the lap joint before riveting began. Here they were able to detect cracks of a magnitude of one or two millimetres – cracks that most other test methods would be incapable of detecting. CVM sensors work on the basis of differential pressure. Each sensor, which is just 125 millimetres thick, is perforated with fine galleries alternately containing air and a vacuum. The presence of a crack or other defect in the monitored material creates a connection between the two types of gallery, altering the distribution of pressure inside the sensor at this point.

Several aircraft manufacturers are meanwhile routinely using CVM for materials testing on the ground. But before SHM sensors can be allowed to perform their intended function on an aircraft in flight, they themselves have to undergo an extensive series of tests in order to determine their resistance, their reliability at detecting failures (POD, Probability of Detection), the dependability of their contact with the material, and in order to test practical procedures such as those to be employed by the service station for maintenance of the sensors. Airbus engineers carry out most of these tests using their A320 MSN001 as an airborne technology platform.

Something of a special opportunity presented itself in connection with full-scale fatigue testing of the A380 in Dresden. In a series of tests that began in September 2005, the aircraft is being subjected to simulated mechanical stresses that will correspond to 47,500 flights by the time testing ends. While conventional strain gauges monitor the load on the airframe, the SHM developers tested the reliability and functionality of well over 100 SHM sensors. They also verified their adherence to the test article. The sensors are usually secured with an adhesive, as this creates a positive bond with the airframe.

But mechanical stresses such as those tested in Dresden are not the only factor influencing the performance of the adhesive bonds and the sensors themselves. They also have to tolerate damp conditions, withstand temperature variations of up to 160 degrees Celsius, and all in all be slightly more robust than the airframe they are monitoring – and that for an entire aircraft service life of 30 years.

To shield the sensors against harmful effects acting on their outer surface, they are often given a protective coating. It is possible to dispense with the protective coating and the adhesive bond by embedding the SHM sensor in the airframe. In that case it is the airframe that protects the sensor, and there is no longer any risk that the sensor might become detached from its monitoring site. Embedded sensor systems are above all a solution that works well with composite materials. Piezoelectric fibres or fibre-optic sensors can be merged almost invisibly with carbon-fibre or glass-fibre reinforced plastics (CFRP / GFRP), or fibre-ceramic composites.

However, two major practical problems have to be overcome when sensors are embedded in the airframe. Firstly, if a structural component needs to be replaced because of wear or damage, the sensor it contains is removed too; and secondly, when sensors are embedded in the airframe, it is difficult to maintain them, virtually impossible to repair them, and in no way at all can they be replaced. Development engineers are currently working all-out to find satisfactory solutions to these problems.

Whether permanently attached or embedded, many SHM technologies are based on principles that have long since won their spurs as methods of inspecting or servicing aircraft – such as ultrasonic or eddy current testing. The biggest difference compared with established methods of non-destructive testing is that, in future, the sensors will remain permanently in place. This applies to both online and offline detection modes. Online means that stresses acting on the airframe are measured continuously while the aircraft is in flight, whereas in the offline mode the data is not read out until the next inspection or maintenance session, either by connecting an analysis device or reading from memory units integrated into the sensors.

Once they have been installed, the sensors provide a simple means of monitoring even those areas of an aircraft that are difficult or dangerous for human inspectors to access. These could be fuel tanks, the engines – or wing flaps. Wing flaps are normally inspected visually, using a borescope: the maintenance technician introduces a small, flexible optical probe into the wing flap and examines the structure for cracks. But this type of inspection is of limited scope. It does not allow the technician to vary the illumination and detect anomalies by the shadow they cast, nor to remove dirt from the surface.

It is difficult to detect microscopic cracks or patches of corrosion in this way. An in-situ SHM sensor would be capable of spotting such defects much faster, leading to considerably shorter inspection times. Nevertheless, structural health monitoring is not intended to entirely replace existing non-destructive maintenance and inspection routines, not least because conventional checks are rarely limited to the precise area specified in the maintenance manual. It is customary for maintenance technicians to cast a critical eye over the surrounding area outside the actual inspection range.

SHM will supply valuable supplementary information and speed up many inspection routines. And it will make it easier to assess damage such as dents or scratches on doors and doorframes caused by loading platforms or airport ground vehicles – a common occurrence however carefully they are manoeuvred. The time taken to estimate the damage to doors alone could be cut by up to half through the use of SHM.

It is less straightforward to calculate the total potential savings for a whole passenger airliner, given that the prescribed checks in this case extend beyond the airframe to include the cabin and the electronic systems as well. The situation is not the same for military aircraft, where passenger comfort is treated as a factor of secondary importance. Here the focus is on systems and structures that are expected to withstand far more extreme stresses than those encountered by commercial aircraft. One study came to the conclusion that it would be possible to reduce the maintenance time on a modern fighter aircraft by more than 40 percent through the systematic use of SHM.

Commercial airlines, too, may soon be able to benefit from significantly shorter maintenance times and lower maintenance costs thanks to SHM. In the longer term, SHM will provide the momentum that sets the ball rolling in terms of conservative approaches to aircraft design, for it provides entirely new insights into the stresses acting on various parts of the airframe during individual phases of operation – taxiing, takeoff, cruise flight and landing – or in extreme situations such as a hard landing. The newly won knowledge about the behaviour of specific structural components will help to optimise the structural design and reduce the weight of future models of aircraft.

Irrespective of safety issues, and however often numerous structural engineers may worry about questions like “How can I possibly design a structure that depends on sensors for its integrity?”, systems engineers are likely to retort “Logically, anyone who thinks this way should never board an aircraft! From the moment an aircraft ceases to glide under its own momentum but is actively piloted, then its ability to fly is at least as equally dependent on systems as it is on the structure.” The list is headed by the sensors that determine flight altitude or rate of descent, and is by no means complete when it reaches the instruments used to gauge velocity or monitor the air in the cabin.

Doubts similar to those expressed by certain manufacturers and their design engineers are shared by many officials working for the airworthiness authorities. SHM is a relatively recent technological development that has yet to gain universal acceptance. On the systems front, it has to meet demanding standards in terms of durability and functionality, and on the structures side it must comply with the severe regulatory requirements for non-destructive testing. These include a 90/95 probability of detection (POD) (in other words, the ability to detect all faults with a reliability of 90 percent and a confidence level of 95 percent. This is a mandatory requirement for all non-destructive testing methods, even today.

Even though its proponents are convinced that SHM will meet all requirements without exception, it will be some time before the fully networked, “sensory” aircraft comes into being. This is mainly due to the extensive qualification testing required for SHM systems. Airbus therefore plans to introduce SHM systems in four gradual stages, or four generations, “generation zero” – the use of SHM in structural testing on the ground – already being common practice. The next stage will be to integrate the use of offline sensors in maintenance routines. Once their reliability has been demonstrated in this application, it will be possible to progress to online sensors and complete onboard systems. Full integration of SHM systems in existing onboard systems will not happen until “generation three”.

Before that stage is reached, there are still a number of obstacles to be overcome in addition to the challenge of testing the sensor systems. For instance, it needs to be ensured that the time and effort required to maintain the SHM systems themselves remains within reasonable limits. Another issue is how the sensor data is to be evaluated. Ultimately, after all, the SHM systems will be expected to not only collect data but also interpret it autonomously.

There is a precursor to the SHM system installed in the two long-haul super-airliners, the A380 and A340-600: the Tail Strike Indicator (TSI). Due to the relatively high angle of attack when taking off and landing, exceptionally long aircraft have a tendency for the back end of the fuselage to touch the runway. The TSI is installed at this particularly vulnerable point. The box-like device sits slightly proud of the fuselage and contains two separate copper loops in addition to other components. If these loops are damaged by abrasion, the circuit is broken and the pilot receives an appropriate notification as soon as a safe phase of the flight is reached.

European Aeronautic Defence and Space Company EADS N.V.
Le Carré · Beechavenue 130-132 · 1119 PR Schiphol Rijk · The Netherlands

EADS Deutschland GmbH · 81663 Munich · Germany
EADS France S.A.S. · 37, boulevard de Montmorency · 75781 Paris Cedex 16 · France
EADS CASA · Ava. de Aragón, 404, 28022 Madrid · Spain