Non-Destructive Testing and Structural Health Monitoring Systems for Damage Assessment in Aerospace Structures
: (Alternative Format Thesis)

  • Christos Andreades

Student thesis: Doctoral ThesisPhD


Advanced engineering structures, particularly in the aerospace field, are expected to operate in harsh environments for extended periods while maintaining the highest possible efficiency. This is achieved with continuous optimisation of the structural design combined with the use of materials that are both lightweight and strong, such as metal alloys and fibre reinforced plastic (FRP) composites. Since the integrity and durability of structures is dependent on the material condition, the existence of structural damage must be identified immediately using various non-destructive evaluation (NDE) methods and structural health monitoring (SHM) systems. These methodologies often involve the transmission and reception of ultrasonic waves through the material using transducers, thus enabling effective, practical and reliable ultrasonic inspection. The damaged interfaces inside a material can interact with the propagating waves, and this can give rise to recognisable linear and nonlinear ultrasonic effects. In general, inspection methods have been found to offer earlier detection of flaws when assessing the nonlinear than the linear features in the acoustic/ultrasonic response of the material.
The research study outlined in this thesis aimed at proposing innovative designs of SHM systems and alternative NDE procedures based on nonlinear ultrasound, for the enhancement of sensitivity and accuracy in the detection of defects in metallic and FRP materials. This work focused on three main SHM/NDE research topics associated with specific challenges currently existing in aerospace applications.
Firstly, a new design of “smart” carbon fibre reinforced plastic (CFRP) laminate containing internal piezoelectric lead zirconate titanate (PZT) transducers was proposed, which could be utilised in the development of on-board SHM systems without exposing the sensors to extreme operating conditions. The novelty is on the methodology used for the electrical insulation of the embedded sensors from the conductive carbon fibres. Specifically, sensors were covered with a thin glass fibre patch to enable enhanced adhesion with the epoxy matrix of the composite plies. This prevents the formation of internal delamination which is usually the case for conventional insulation techniques involving the use of polymeric films and coatings. Experimental mechanical tests proved that this layout of internal sensors had no impact on the tensile, compressive, fatigue, flexural and interlaminar shear strength of CFRP composite samples. In addition, the results from several ultrasonic experiments confirmed that such embedded transducers could detect defects of different type and size in CFRP plates, and monitor the growth of impact damage in composite samples subject to repeated tensile loading. Both delamination and impact damage were assessed using two nonlinear ultrasonic techniques.
In the second research topic, an SHM method involving the transmission and reception of ultrasonic waves within a surface-attached array of PZT transducers was developed for the detection and localisation of barely visible impact damage (BVID) in composite panels. In contrast to most of the available ultrasonic SHM methods, the accuracy of the developed algorithm did not rely on the recording of signals at the original state of the material (baseline data) or a priori knowledge of the wave velocity. In addition, the reliability of the method was enhanced by adding some initial steps to the algorithm for the identification of malfunctioning transducers. Moreover, the method included a simple process for the selection of a suitable frequency for signal transmission through the material. The signals recorded between all pairs of PZTs in the array were used to evaluate the level of material nonlinearity in each sensor-to-sensor path based on the received amplitude at the fundamental and second harmonic frequencies. Thus, a surface plot was generated showing the variation of material nonlinearity in the area enclosed by the transducers, with the peak amplitude being the position of damage. The proposed method was experimentally tested on three CFRP panels with different dimensions and shape, and the results verified the correct localisation of BVID, as well as the identification of a partially damaged transducer.
Regarding the last proposed SHM/NDE topic, an ultrasonic phased array method was optimised for the improvement of signal-to-noise ratio (SNR) in the acoustic response recorded in pulse-echo testing experiments. The aim was to achieve higher effectiveness in the detection of contact-type defects (e.g. closed crack and delamination) at several depths in metallic and laminated composite structures. The presented technique involved the processing of single-frequency and dual-frequency signals that were acquired with four different firing orders of the piezoelectric elements in the phased array probe. This allowed the filtering of unwanted linear and nonlinear acoustic features resulting in the extraction of the nonlinear response corresponding only to the response of damaged surfaces under excitation. Experiments were performed on piles of aluminium disks and CFRP laminates representing samples with several horizontal interfaces of contact defects. The obtained nonlinear ultrasonic response of the material was compared with the linear response acquired under standard phased array inspection. In all cases, the results proved that the defect-related peaks exhibited in the nonlinear response were characterised by higher SNR and their positions indicated the locations of contact interfaces with a smaller error, relative to the linear peaks.
Therefore, the individual designs and methods proposed by this research study could potentially be utilised in the relevant SHM and NDE applications for improved detection and localisation of defects in metallic and composite materials.
Date of Award17 Feb 2021
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorMichele Meo (Supervisor), Francesco Ciampa (Supervisor) & Fulvio Pinto (Supervisor)

Cite this