Skip to main content

Global in flight health monitoring platform for composite aerostructures based on advanced VIBRATION based methods

Final Report Summary - VIBRATION (Global in flight health monitoring platform for composite aerostructures based on advanced VIBRATION based methods)

Executive Summary:
The VIBRATION project developed a new in-situ SHM platform that can estimate the possibility of damage in a composite structure during flight from changes in the vibration characteristics of the structure. The development started by manufacturing 40 small scale composite booms. The vibration response of these boom was studied in order to form a baseline for the healthy state of the structure. All variability from materials, manufacturing process and post process machining was revealed in these tests. Some of the small scale booms were damaged afterwards a specific locations and under specific conditions. The damaged boom were also analysed in order to get their vibration response and compare it to the healthy state. The high variability of the responses led to the development of dedicated mathematical analysis tools in order to identify the damaged parts from the healthy one with very high probability (very close to 1). Therefore, a methodology for damage detection has been formed, making sure that the probability of a false positive detection is almost zero. At the same time, numerical modelling of the structure was performed in order to extend the number of "visual experiments" on the boom and enhance the accuracy of damage prediction. Both analytical/experimental and numerical results were fed to the final prototype equipment. The Graphical User Interface of the software that was developed for the prototype was simplified according to the End User requirements. The final output from the prototype equipment is a series of bars measuring the probability that a composite boom structure is damaged. A horizontal line in the same graph denotes the threshold between a healthy and a damage part.
Project Context and Objectives:
The economic and life-safety implications of early damage diagnosis, including damage detection, localization and quantification (magnitude estimation), in the aerospace industry have motivated a significant amount of research in Structural Health Monitoring (SHM). Reducing the structural weight and enhancing the customer’s satisfaction by decreasing the maintenance cost, while ensuring the aircraft’s efficient operation and structural integrity, friendliness to the environment and high level of safety are some of the key drivers for the aerospace industry to enable them to become more competitive. Maintenance aspects are increasingly significant in reducing the Direct Maintenance Costs (DMC) of the total operational costs of the airline companies.
At the same time, over the last decade, composites are increasingly used in civil aircraft. For example, their total weight percentage for the new Airbus airplanes has more than doubled. Thereby, the development of appropriate SHM technologies capable of monitoring the integrity of composite aerostructures is very important, whilst the biggest challenge is their implementation under in-flight conditions. This can significantly reduce life cycle costs by minimizing inspection time and effort, by extending the useful life of new and aging aerospace structural components and by increasing aircraft availability.

VIBRATION main objectives and enabling technologies are:
Enabling technologies: In the VIBRATION project the actual (real world) “training” experiments will be replaced by simulated experiments, using an analytic model, which will be fine-tuned (updated) using a limited number of actual experiments. The analytic modelling of the structure under its healthy and damaged states will be based on Finite Element (FE) methods, while the fine tuning (model update) of the analytic model will be based on inverse modelling techniques
Technological objective: SHM platform training based on a limited number of actual experiments

Enabling technology: The developed SHM methodology incorporates vector output-only system identification and non-stationary signal modelling techniques.
Technological objective: SHM platform operation under unobservable and non-stationary excitation.

Enabling technology: The formulation of the SHM methodology enables the use of continuous structural topologies through the proper incorporation of the Functional Pooled (FP) modelling techniques.
Technological objective: Precise damage localization and magnitude estimation on continuous structural topologies.

Enabling technology: Advanced statistical signal processing technologies employed for the extraction of damage-sensitive non-modal characteristic quantities. Statistical sensitivity analysis tools are used for the characteristic quantity selection.
Technological objective: Damage-sensitive and robust Vibration-based SHM methodology.

Enabling technology: The proposed SHM methodology employs effect-removal or effect-inclusive advanced signal processing technologies in order to account for the varying environmental conditions. Random coefficient pooled modelling technology is used in order to account for the various uncertainties.
Technological objective: SHM platform operation under varying environmental conditions and/or uncertainties.
Project Results:
The demonstrator component was a real size composite boom used in the manufacturing of UAVs by beneficiary IAI. (Figure 1).
A representative small scale component was used for the development of the SHM methodology was a composite beam of rectangular cross-section. 45 units of the small scale component were manufactured for by beneficiary ATR. (Figures 2, 3 and 4).

The specifications for the SHM platform were defined. The selected damage type was damage after impact. Three distinct impact energies were studied.

A total of 1800 vibration experiments in the small scale composite parts have been defined (Tables 1 and 2). A total of 3953 signal have been retrieved from these experiments.

The impact location was defined. Ten points for impact damage have been selected (Figure 5).

The NDT of the healthy parts revealed a number of defects that were due mainly to variabilities in the manufacturing conditions, machining after processing and transportation conditions. Typical defects are shown in Figures 6. These defects resulted in wide variabilities in the vibration response of the healthy parts and presented a challenge to the consortium. A high number of characteristic quantities have been analysed and investigated in order to analyse the vibration response (Figure 7).
The experimental setup is shown in Figure 8. An analysis based on non-stationary time dependent excitations for the extraction of the characteristic quantities has been performed using non-stationary excitation and the TARMA model, which enhanced the probability of detecting a true defect while the probability of detecting a false event is greatly reduced. Representative vibration response is shown in Figure 9. The variability in the dynamics of the healthy composite parts is shown in Figure 10.

A model based on Random Coefficient (RC) models were used for the analysis of the vibration responses of the small composite parts and for the actual demonstrator, RC models are global representations based on multiple vibration signals obtained under different conditions. They can capture significant information of uncertainty. They use a single model with parameters (coefficients) being Gaussian distributed random variables. This is referred to as an RC–G model. The compact representation of the structural dynamics under uncertainty is an important advantage that may lead to an online Structural Healthy Monitoring (SHM) scheme of reduced complexity.
A full description of the mathematical analysis is attached to this report.

The prototype system developed in the project can be seen in Figure 11. The final Graphical User interface (GUI) was simplified in order to include just bars. The height of the bar indicated how damaged the composite part is (in effect how far the vibration response of the part is from what is considered to be a healthy part). Some screenshots are shown in Figure 12 (healthy part) and Figure 13 (damaged part). Different colours are used in order to simplify the interpretation.

Potential Impact:
The prototype system developed in the project is a step towards a fully automated in-situ NDT assessment of composite structures in-flight. The overall TRL of the technology developed is 4 with the development of the methodology for the analysis of the signals and the hardware that is needed to implement the system reaching much higher maturity. Further development by the SMEs and End Users involved will focus on the demonstration of the technology during flight conditions (ruggedisation of the prototype equipment in order to be airborne, communications with the ground base, and handling of the vast amount of data that will be generated during a typical flight).
A number of dissemination activities promoted the project work to a wider audience. The main dissemination event for the industry was the AERODAYS 2015 event, held in London in October 2015. The academic community was reached through the European Workshop on Structural Health Monitoring (EWSHM 2016) and the Conference on Noise and Vibration Engineering (ISMA 2016).
A project flyer has been developed (attached as part of the present report).

The main actors for the commercial exploitation of the project results at this stage are ADVISE and IAI. ADVISE developed the prototype DAQ system. The company is currently investing in finalising the product offering, inlcuding certification (CE marking) in order to launch it to the European market. At the same time, a close collaboration with IAI is on-going in order to develop a version of the system that can become airborne and used initially in UAVs. IAI will start capitalising from the added value provided by the inclusion of an in-situ SHM tool in their range of civil UAVs once all the airborne version of the system is ready.
Exploitation of know-how is undertaken by TWI. The project results are disseminated to its members. Interest for the technology has been registered by a number of aerospace companies. Their feedback (commercial pull) focuses priority for commercial projects led by TWI in the areas of accurate location of the damage (for given geometries), the identification of the damage (for specific composite structures) and the reliability of the technology. Commercial opportunities in the area of power (wind turbines mainly) have also been registered for future application of the VIBRATION technology
List of Websites:

Contact details: Dr Mihalis Kazilas, TWI ltd, Adhesives, Composites & Sealants Section, Cambridge, Granta Park, CB21 6AL, UK