4. Aerospace Requirements

Power consumption is part of the physical requirements of the SHM system. Traditionally power is supplied to the system through wires; this however does have some significant disadvantages such as the cost associated with installing the wiring. This wiring also adds significant weight to an aircraft, reducing efficiency and increasing the cost of operation. That is the reason why there is a great interest in implementing wireless SHM system (Noel et al. 2017; Lee et al. 2016; Dutta et al. 2005; Rusci et al. 2017; Sutton et al. 2017; Li et al. 2010; Champaigne and Sumners 2007); in fact, it is estimated that in the lifetime of an aircraft, cost savings of 14–70 million dollars could be achieved (Gao et al. 2018). The removal of wires although have weight saving benefits, creates a clear problem: how to power the device in a manageable way? This has led to developments in terms of energy efficiency of the electronic units (DAQs and sensors) (Fu et al. 2018, 2019). In terms of passive SHM system monitoring (e.g. impact events), advancements have been caried out to develop energy efficient event-triggered systems (Fu et al. 2018 Yuan et al. 2016; Qiu et al. 2013; Yuan et al. 2014). Similarly, for active sensing (guided wave excitation and sensing), several wireless monitoring systems have been developed, each with advantages and shortcomings in terms of size and weight of the units, power consumption and the response time (Fu et al. 2019; Aranguren et al. 2013; Zhang 2013; Zhong et al. 2015). Most of the proposed wireless DAQ units are battery operated, with some recent research in terms of energy harvesting capabilities. A unique development which overcomes the wiring and energy challenges of a passive unit was proposed through inductively coupled transducer system (Zhong et al. 2014) where its ideal for ground maintenance but will require access to the embedded sensors. Therefore, not ideal for areas with limited accessibility.

Batteries are common within wireless sensors; however, concerns over their safety and their limited battery life mean that application within aerospace structures is restricted. To remove the need for batteries, or to extend their lives, energy harvesting is being utilized. Available sources of power vary depending on the application and location, typically when available solar power has been used, as it can produce the highest power for its size. Within aviation, however, this is not a reliable single source of power and its integration will interfere with the aerodynamics of the structure, so alternative sources must be considered; for aircraft applications, the most applicable are vibration and thermal (Le et al. 2015).

Vibration harvesters are very effective in areas where consistent vibration exists, such as near engines, or helicopter gearboxes, as available energy is abundant at a known frequency. Elsewhere, vibration may also be high because of airflow; however, this is highly dependent on operating conditions, making it random and difficult to harvest. Although values are very dependent on the aircraft type and operating conditions, studies have shown that the typical frequency is between 100 and 500 Hz, up to 5 g (Thomas 1962). Within helicopters, rotor speeds will also lead to lower frequency vibration (Le et al. 2015). Although advancements had been made into less resonant sensors (Townsend et al. 2019), most vibrational energy harvesters only operate efficiently at resonance, making it possible near consistent sources. Stated values for vibration based-energy harvesters in research vary, as the acceleration that they are exposed to is not uniform. The size of the harvesters is also variable, some being micro-electromechanical systems and others not; this makes direct comparison difficult. It however can be concluded that a vibrational energy harvester suitable for a wireless system is likely to produce at most milliwatts of power (Wei and Jing 2017).

Another option that has the potential to produce higher power values is the thermoelectric generator (TEG), which harvests energy between two faces where a temperature gradient exists. This includes areas near the engine outlets or the outer shell of the aircraft during take-off and landing, where over 50 °C of temperature difference can be temporarily produced. Most TEG utilize the phenomenon called the Seebeck effect, and when operating in the conditions found on-board an aircraft can generate hundreds of milliwatts (Hamid Elsheikh et al. 2014), be it for a limited time period.

Power availability on-board an aircraft is limited, meaning the power consumption of any wireless device must be kept to a minimum, whether it is powered by battery or energy harvesting.

A key feature for the certification of the SHM is its reliability compared to existing NDE techniques. It should follow MIL-HDBK-1823 which is the handbook for Nondestructive Evaluation System Reliability Assessment. The reliability requirements should cover the minimum performance constraints. The reliability of the SHM system includes both on-board and on-ground equipment. This means that the SHM installation, wiring, connectors and sensors must survive the harsh environmental and operational conditions of flight and their service life should be comparable to the life of the structure. If their service life does not comply with the service life of the host structure, there should be reliable ways of detecting, repairing or replacing faulty parts (Yue et al. 2018).

Moreover, fault-tolerant design of equipment and installations should be considered in reliability analysis for the chosen intended function, e.g. tolerance in sensor placement, baseline measurements. These requirements can be linked with the design, development and optimization of the SHM system to increase its reliability and redundancy.

The APR 4671 guidelines can be followed for the safety requirements of an SHM system which is acceptable by industry. These include the following:

The SHM should remain effective for the expected life duration of the host structure and survive its surrounding environmental and operational conditions of flight.

The SHM system should function reliably under all foreseeable operation conditions of flight. The operational requirements for an SHM enabled aircraft will specify interfaces between the maintenance crew and on-aircraft and ground SHM components

The on-board requirement of the SHM system should follow RTCA/DO-160C certification guidelines for environmental conditions and test procedures of airborne equipment to ensure the reliability, durability and long life of the installation of on-board equipment as well as their operation.

The on-ground requirements are primarily directed towards condition-based monitoring (CBM).

The ground operational requirements should report on the following:

As part of the reliability assessment of the SHM system, the accuracy of the measured parameter (e.g. strain, guided wave) under operational conditions must be compared with the real value that the structure experiences and the error of the measurement evaluated. The Table 4.1 describes the different test methods and procedures that should be followed for on-board electronics (sensors, connectors, wires) (Salmanpour et al. 2017).

For example for a regional aircraft operating within Europe the operational and environmental test following MIL-STD 180G guidelines requires the on-board SHM equipment to be tested in ‘Basic cold C1’ category, with daily air temperature ranging between −32 and −21 °C, (−33 to −25 °C induced) in the low range with the extremely low temperature in Europe set as −55 °C in Russia.

Depending on the location of the equipment on board of an aircraft, the RTCA DO-160 specifies a specific vibration profile. The functionality and longevity of the SHM components as well as the reliability of the diagnosis should be tested in the presence of the vibration load.

For aeronautical applications, the decision to have a permanently installed SHM system for structural prognosis will be driven by its reliability, cost and the added weight of the system. The additional weight of the on-board SHM system might be outweighed by the reduction of structure thickness and hence weight which be one of the benefits of the application of SHM. Therefore, the restriction of size and weight of the SHM components is closely related to sensor optimization which is presented in the next section. Some recent innovations have resulted in reducing the weight of the wires and connectors significantly which is specifically beneficial to guided wave-based SHM systems. Considering that each sensor has to be connected from both positive and negative terminals with a wire to the DAQ unit, for a sparse network of sensors, this results in a substantial additional weight. In addition, handling of the wires is an added challenge to have them fixed to the structure. During the operation, they could become projectile and they can break the sensor terminals as well as cause safety concerns. This challenge is not so significant for fibre optic sensors or embedded sensors. However, for piezoceramic sensors, the proposed solutions are in the form of a ready-made layer that can be installed onto the structure either during manufacture or assembly and will include sensors, wires, connectors and the protective layer. Examples of these solutions are the smart layer by Stanford (Qing et al. 2009), Diagnostic film by Imperial College shown in Fig. 4.2 (Bekas et al. 2018) and the integrated SHM layer (Schmidt et al. 2016), just to name a few.

As it was emphasized in the previous sections, the performance and reliability of the SHM system is a critical factor in its uptake as a maintenance strategy. The performance of the diagnosis based on the sensor data is directly related to the number and position of the sensors. There is a balance between the cost and the benefit of how many sensors should be added to the host structure without affecting its functionality but still diagnose the structure with high reliability. Therefore, specialized optimization algorithms should be developed which takes into account the specific outcome of the SHM system for composite structures as well as the limitations in terms of placement of sensors and their numbers.

Depending on the application of the SHM, different SHM techniques have been developed which include: vibration-based methods, strain monitoring and ultrasonic guided wave-based techniques. Each class of techniques will have different inputs and outputs, different types of sensors and different operational parameters and hence a different objective function to optimize. There have been many approaches proposed for the optimal sensor placement (OSP) for SHM systems mostly based on an objective function which is the POD of the damage detection algorithm. These techniques include data-driven methods such as an artificial neural network (ANN) in (Staszewski and Worden 2001), genetic algorithm (GA) (Guo et al. 2004) and Bayesian approach (Flynn and Todd 2010a, 2010b), which the objective function for all of them is based on the POD of the damage detection algorithm. The disadvantage of such an approach is that first, it requires a lot of data to be able to build the POD function. Secondly, the POD function for damage detection in composite structures is not well defined because of the complex nature of the damage. Last but not least, if any changes will be made to the damage detection algorithm, the optimum sensor number and location will then no longer be valid. In a different work, Fendzi et al. (2014) proposed an optimization strategy based on GA but one which does not depend on the damage detection algorithm, but it depends on a POD function constructed from scattered energy of guided wave interacting with a damage. This scattering is a theoretical model and does not represent the scatter profile in a complex composite structure. Most of the reported techniques are application-specific and consequently cannot be evaluated in a comparative study. Moreover, their performance depends on tuning factors which makes it impossible to compare for two different techniques. A different approach was proposed in (Thiene et al. 2016) where the optimization is based on maximum area coverage (MAC) within a sensor network and applies to any structure of complex geometry and any damage detection algorithm which is based on guided waves in the pitch-catch configuration. In addition, complex geometries such as plates with openings can also be considered. The objective function (which is to maximize the guided wave coverage) is then optimized using a genetic algorithm (GA) optimization framework. An example of optimal locations for 4 sensors on a rectangular plate with a hole is shown in Fig. 4.3, where the objective was set to maximize the area inside the sensor network (to avoid edge reflected waves).

A comprehensive review is presented in (Ostachowicz et al. 2019) where different OPS algorithms for different SHM techniques are thoroughly investigated. This area of research still requires a lot of attention to propose an optimum method with could be adapted to any structure and any SHM method.

In summary, this chapter presented an overview of the requirements that need to be addressed, for any SHM system to be considered for CBM of a real aeronautical structure under real operational and environmental conditions. The most important factor for an SHM to be used in diagnosis and prognosis is for the designer and the maintenance engineer to show that the SHM system can survive the life cycle of the structure, and if not, each component can be identified when malfunctioning, repaired or replaced. The second important factor is to demonstrate that the SHM methodology is capable of diagnosis with the same level of accuracy and reliability (often called 90/95%) of non-destructive techniques that are currently being used by manufacturers and aircraft operators. To this end, the SHM system needs to be considered from the design stage to manufacturing, assembly and finally operational life of the structure.