GB2545271A

GB2545271A - Determining physical characteristics of a structure

Info

Publication number: GB2545271A
Application number: GB1521898.5A
Authority: GB
Inventors: Saleh Sayeda-Samira; Campagne Benjamin
Original assignee: Airbus Operations Ltd
Current assignee: Airbus Operations Ltd
Priority date: 2015-12-11
Filing date: 2015-12-11
Publication date: 2017-06-14
Also published as: GB201521898D0

Abstract

A method comprising illuminating an area of a structure with a generation laser 104 to produce at least one ultrasonic signal 106 within at least the coating 103 of the structure; illuminating the area with a detection laser 105 to produce at least one reflected signal A having waveform characteristics that are influenced by the ultrasonic signal; and then detecting the reflected signal. The waveform characteristics of the reflected signals are used to determine a first parameter related to a physical characteristic of the coating and a second characteristic related to a physical characteristic of the body 102 of the structure. The first parameter may relate to the thickness of the coating or paint layer. The second parameter may relate to the integrity of the body, material density, presence of a defect, size of defect, surface properties and/or thickness. The coating may be in a fluid state during testing. The coating may comprise aeronautical paint and the body may comprise an aeronautical part. The waveform characteristics of the reflected signals may comprise time-of-flight information.

Description

DETERMINING PHYSICAL CHARACTERISTICS OF A STRUCTURE

TECHNICAL FIELD

[0001] The present invention relates to determining physical characteristics of a structure. In particular, but not exclusively, the invention employs laser ultrasonics.

BACKGROUND

[0002] Laser ultrasonics is a known, optical technique which may be used in nondestructive testing of structures. In very general terms, laser ultrasonics deploys a first laser beam or pulse, which is directed onto a surface of a structure in order to generate ultrasonic waves within the structure, which can be detected by a second laser beam. The second laser beam is reflected from the surface and acquires waveform characteristics determined by the ultrasonic waves, which can be analysed to reveal physical properties of the structure. US 6,092,419, for example, proposes a laser ultrasonics process for measuring a thickness of paint on an automotive body.

SUMMARY

[0003] A first aspect of the present invention provides a method comprising illuminating with a generation laser an area of a structure, comprising a body and a coating thereon, to generate at least one ultrasonic signal in the thermo-elastic regime within at least the coating, illuminating the area with a detection laser to produce at least one reflected signal comprising waveform characteristics that are influenced by the ultrasonic signal, detecting the reflected signal(s), and determining, from waveform characteristics of the reflected signal(s), a first parameter related to a physical characteristic of the coating and a second parameter related to a physical characteristic of the body.

[0004] A second aspect of the present invention provides an apparatus comprising a generation laser to illuminate an area of a structure, comprising a body and a coating thereon, to generate at least one ultrasonic signal in the thermo-elastic regime within at least the coating, a second laser to illuminate the area to produce at least one reflected signal comprising waveform characteristics that are influenced by the ultrasonic signal, a detector to detect the reflected signal(s), and a processor arranged to determine, from waveform characteristics of the reflected signal(s), a first parameter related to a physical characteristic of the coating and a second parameter related to a physical characteristic of the body.

BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: [0006] Figure 1 is a schematic diagram of a structure comprising a body having a coating thereon which is illuminated by a generation and a detection laser for determining physical characteristics of the structure; [0007] Figure 2 is a flow diagram of a process for determining physical characteristics of a structure; [0008] Figure 3 A is a graph which depicts a time-varying reflected signal; [0009] Figure 3B is a graph which depicts a frequency domain representation of a region of the graph of Figure 3 A; [0010] Figure 4A is a schematic diagram of a calibration structure comprising a body having a coating with a varying stepped thickness; [0011] Figure 4B is a graph indicating a relationship between a coating thickness and an associated peak resonance frequency, for example, determined using the calibration structure of Figure 4 A; [0012] Figure 5 is a flow diagram of a process for determining physical characteristics of a structure; [0013] Figures 6A-C are graphs depicting various peak resonances determined using laser ultrasonics; [0014] Figure 7A is a schematic diagram of a robotic arm and scanning head that may be used to perform methods described herein; and [0015] Figure 7B is a schematic diagram of a feedback control arrangement for controlling a coating application based on a thickness measurement.

DETAILED DESCRIPTION

[0016] In arriving at embodiments herein, the inventors have determined that it is possible to deploy a laser ultrasonics process for multiple purposes in tandem, for example, for non-destructive testing of structures in addition to coating thickness measurement. Such testing may traditionally be time-consuming and costly (in terms of time and equipment) and an ability to conduct multiple kinds of tests in tandem is advantageous. In the aerospace industry in particular it is important to monitor the painting quality of parts, since a paint layer that is too thick will add additional weight to an aircraft and could damage fuel efficiency. For example, if a thirty micron additional thickness of paint is applied over the entire surface area of an Airbus A350-900 aircraft, it is estimated that this could add around 980kg of additional weight to the aircraft. If a similar amount of additional paint were added to each coat, for example a primer, a base coat and one or more top coats, the increased weight could amount to several tonnes.

[0017] For qualification and safety reasons, non-destructive testing of parts in the aerospace industry may be performed multiple times at different stages of aircraft and aircraft part production. For example, a part may be tested after its manufacture and it may be tested again after it has been painted and installed into (or onto) an aircraft.

[0018] As will be described by way of example, advantageously, a single set of measurements can be generated and used for non-destructive testing and paint coating thickness measurements. In this way, a set of non-destructive testing results may be attained in addition to or instead of other sets of non-destructive testing results. This may lead to increased production efficiency and reduced costs without increased risk. Such testing would be suitable for coatings other than paint.

[0019] Methods of the present invention may be used in the aerospace industry to measure using laser ultrasonics a wet film thickness of aeronautical paint applied to the surface of an aircraft or aircraft parts. Aircraft or aircraft parts may comprise composite or non-composite materials. The technique can be applied on various coatings such as a primer, base coat or top coat, for example, to prevent excess-weight due to a greater-than-optimal coating thickness. The methods described herein may provide a thickness accuracy up to or in excess of ±1.5pm. Other perceived advantages include prevention of damage caused by inaccurate paint film thickness, reducing the amount of wasted resources such as paint, time and cost required to strip dried paint on the discovery of a thickness error or inconsistency, reducing in-service issues of the structure such as crack growth, fatigue life failure, adhesive bond failure among others, or improving surface texture of applied coatings for customer satisfaction.

[0020] A composite material may be of a known kind and comprise reinforcing fibres in a thermoplastic or thermosetting organic matrix, for example such as an epoxy resin.

[0021] Figure 1 shows a schematic view of a structure 101 comprising a body 102 having a coating 103 thereon and an exemplary arrangement for performing a non-contact (non-destructive) analysis of the structure using a laser ultrasonics technique. The structure 101 is typically illuminated by a pulsed generation laser 104 and a continuous or pulsed detection laser 105. The generation laser 104 illuminates an area of the surface of the coating 103 with laser shots (or short pulses), which each generate and cause an ultrasonic wave 106a or waves to propagate within at least the coating 103. An ultrasonic wave 106b or waves may also propagate through the body of the structure 101. The ultrasonic wave or waves may be caused by one or more effects. For example, in an area of the coating that is illuminated by the generation laser, the laser light may be absorbed to provide a rapid local temperature increase, causing an instantaneous change in a respective density gradient of the coating, which generates the ultrasonic wave or waves. In addition, or alternatively, ablation due to rapid heating of an area of the surface of the coating may be accompanied by the movement of a solvent mass from the surface of the coating, thereby to produce a force in the opposite direction due to evaporation of the solvent, to generate an ultrasonic wave or waves. An ultrasonic wave in this context may be thought of as a ‘signal’ insofar as it carries information which is passed via interaction with incident detection illumination 105a to form a respective reflected light signal that is received by a detector 107, which incorporates an interferometer. The reflected optical signal thereby acquires and comprises waveform characteristics that are influenced or determined by the ultrasonic signal. The waveform characteristics may be processed by a processor 108 to determine physical characteristics of the structure, which may be presented at an output 109 for downstream analysis and/or inspection by an operator of the system.

[0022] The generation laser 104 may comprise a pulsed, CO2 laser. In one example, for a structure comprising a composite material, the generation laser may generate laser pulses having a wavelength of 10.6pm, with a pulse duration of 100ns and a power of around 180mJ. The generation laser pulses are sufficient to generate ultrasonic waves or signals in the thermo-elastic regime of the coating 103. A spot diameter of the generation laser 104 may be in the region of 14mm. The detection laser may be a continuous wave, Nd:YAG laser. The detection laser may comprise a wavelength of 1,064pm, a pulse duration of 100ps and a power of around 60mJ. The spot diameter of the detection laser may be in the region of 5mm. Desirably the detection laser may be directed to the same area of the structure as the generation laser, in order to be able to maximise pickup of the ultrasonic waves. The interferometer of the detector may be a differential confocal Fabry-Perot interferometer, with a bandwidth sensitivity of 0.5-15.0 MHz, for detection of the ultrasonic wave characteristics. The interferometer may collect the detection laser light reflected off the surface of the structure to record the ultrasonic modulations in a known way.

[0023] The generation laser and detection laser properties may be different for performing laser ultrasonics on different structures, for example, comprising a noncomposite or metallic material. For example, a generation laser wavelength may be 10.6pm with a pulse duration of 10ns with a power around 180mJ, and a detection laser wavelength may be 1.570pm with a pulse duration of 10ns and a power higher than 180mJ. The selection of appropriate generation laser and detection laser parameters for different structures may be different and may be determined through knowledge of the structural characteristics of various materials and/or through known testing and calibration techniques.

[0024] In some embodiments, the illumination on the structure from a detection laser may be collinear and superimposed with the illumination from a generation laser. The generation laser and the detection laser may be applied in the same or substantially the same location in pulse-echo mode. The beams of each laser may be steered using an arrangement of scanning mirrors (not shown) to perform a fast scanning measurement across a 2-dimensional inspection area of a structure. The mirrors may for example be galvanised. Given an appropriate scanning arrangement, an inspection area of around 1.5m x 1.5m may be scanned in less than one second, for example, by using a scanning step of at least 75mm and a sampling rate of between 10-l,000Hz. The number of point measurements taken for the scan of the area controls the resolution of the scan. Moreover, if the scanning of the inspection area is performed using steps of 1 mm, where the dimension of the inspected area is 280mm x 70mm, this relates to a scanning time of less than one minute for the area.

[0025] The coating, which may be a paint, may be in a fluid state or in a solid state, or somewhere in between (for example, at a point in time during the curing/drying process).

[0026] Figure 2 shows a flow chart identifying at a high level a method for determining physical characteristics of a structure. At block 201 an area of the structure is illuminated with a generation laser. At block 202 the area is illuminated with a detection laser. At block 203 a signal that is reflected from the structure is detected at a detector. At block 204 a first parameter relating to a physical characteristic of the coating and a second parameter relating to a physical characteristic of the body of the structure are determined. The first parameter may relate to the thickness of the coating. The second parameter may relate to the integrity of the body of the structure.

[0027] Figure 3A shows a graphical time-domain representation of an exemplary detected signal 301, which has been reflected by a structure. The reflected signal may be processed to produce a frequency-domain processed signal 302 as shown by Figure 3B, as will be described.

[0028] As illustrated in Figure 3A, the reflected signal is typically time-varying and contains waveform characteristics that are related to physical properties of the structure. For example, the reflected signal may comprise information related to a thickness of the coating and/or the body of the structure. The reflected signal may comprise features (A) of a ‘surface bang’, which is for example caused by light from the generation laser ablating the surface of the coating. The surface bang creates the ultrasonic waves. The reflected signal may also comprise features (B), a short time later, related to so-called ‘back-wall echoes’, which comprise ultrasonic signals reflecting at interfaces within the coating and/or body of the structure. The features (B) related to back-wall echoes have waveform characteristics that provide information that can be used to determine characteristics of the structure.

[0029] For example, a time Td between features (A) and features (B) may be indicative of an acoustic signal travelling from the surface of the coating, reflecting off an interface between the coating and body (or from an opposing side of the body), and being detected again at the surface. This equates to time-of-flight information of an acoustic signal travelling through twice the thickness of the coating (and/or the body). If, for example, characteristics of the coating are known and/or are pre-calibrated - such as times-of-flight of acoustic signals through the coating material at different degrees or times of drying - then a coating thickness can be deduced from the time between features (A) and features (B), using a graph of the kind that is illustrated in Figure 3A. With reduced coating and/or body thicknesses, where overall thickness may be relatively thin and thicknesses of the various layers may be comparable, for example, for painted bodywork panels of an aeroplane, features (A) and features (B) may be very close together and hard to differentiate between using a time-domain analysis. In such instances, a frequency-domain analysis may be useful. In many scenarios, both time-domain and frequency-domain analyses may be performed to extract information that is more beneficially determined via one analysis method or the other.

[0030] As illustrated in Figure 3B, a reflected signal is detected and may be processed, for example, using a Fast Fourier Transform (FFT), to extract frequency component characteristics of the time-varying waveform. In particular, if it is assumed that features (B) repeat periodically multiple times (not shown in Figure 3A), due to repeated reflections (or echoes) of the acoustic wave, then time-of-flight information may be deduced from a resonance peak that represents the periodicity of the reflected signal. As has been indicated, such a peak may be more discernible from other peaks in the frequency domain than in the time domain. In this way peak resonance frequencies associated with each layer of the structure, and in particular the respective thicknesses thereof, may be identified. Accordingly, one would expect there to be at least one peak resonance frequency associated with the thickness of the coating and at least one peak resonant frequency associated with the thickness of the body or substrate of the structure. Other frequency peaks associated with the body of the structure may provide information related to structural characteristics of the body, such as flaws, cracks, or defects in the body, amongst other physical properties including material density, for example. By way of example, for characteristics of the reflected signal comprising time-of-flight information, a defect present in the body or coating of the structure may be detected via an echo, arriving at a time before (that is, with a higher frequency) than that which would be expected for a peak associated only with a thickness of the body of the structure in the absence of a defect.

[0031] In the following description, a reference to a ‘peak’ may relate to a time-domain peak or to a frequency domain peak, as the context dictates, depending on whether time-varying or frequency characteristics are being evaluated. In addition, it will be appreciated that information that is discernible from one kind of peak or peaks in one domain (time or frequency) may in addition be discernible from a peak or peaks in the other domain.

[0032] In order to determine a coating thickness from a laser ultrasonics measurement technique, it is beneficial to know some properties of the coating and/or body of the structure. For example, these properties may include the acoustic velocity of ultrasound within the respective material, the speed of the ultrasonic waves in the coating and/or body, the acoustic impedance of the coating and/or body, or density of the coating and/or body. If the coating when subjected to laser ultrasonics has a fluid state, such as for freshly applied paint, the way in which these properties change over time during the respective drying or curing process may need to be understood in order to determine a coating thickness (at any particular time and degree of drying). A preliminary calibration may be used to determine a relationship between coating thickness and such other properties. The calibration may for example consist of monitoring an actual thickness (Δχ in mm) of the fluid coating and measuring a time of flight (At in ps) of an ultrasonic wave, to determine how the coating properties change with time. The time-of-flight may then be associated with a peak resonance frequency f of ultrasonic waves in the associated layer (where f=l/At in MHz). When the time-varying thickness and time-of-flight information have been determined, it is possible to determine the behaviour and velocity (ν=Δχ/Δί) of the ultrasonic signal over time.

[0033] A calibration may be performed to provide information about the behaviour of the ultrasonic signal that is generated in the structure. As has been indicated, the behaviour of the ultrasonic signal in a thin layer or coating may be different to the behaviour of the ultrasonic signal in a thick layer or coating. This is due to the ultrasonic waves propagating in the coating being reflected differently at the surface and back-wall of the coating. The thickness of the coating affects the repetition rate of the ultrasonic waves being reflected between the surface and back-wall of the coating, causing a distinct peak resonance for each thickness of coating. As such, once the peak resonance associated with a coating has been determined, the calibration information may be referenced to provide a conversion between the peak resonance (from processing of the detected reflected signal) and layer thickness of the coating at any given time after its delivery, such as by spray painting. In general, the thinner the coating or layer is, the higher the frequency of the respective peak resonance is (and the shorter the associated time-of-flight is).

[0034] Figure 4A shows a calibration structure 401 comprising a body 402 having a coating 403 with a varying stepped thickness. In this example, the body is a 2.5cm thick aluminium plate having paint thereon, where the stepped thickness of the paint varies between 32pm to 330pm in a plurality of steps. The paint may be applied, for example, using a tool that scrapes the paint over the surface to obtain a homogeneous layer at each desired thickness. Other appropriate ways of applying the paint may be used. The aluminium plate is preferably thick enough to allow easy discrimination between peak resonance frequencies or time-peaks associated with the coating 403 and the body 402 of the structure.

[0035] The calibration structure in Figure 4A may be used for determining the thickness of a coating (of the same material, in this instance paint) on any structure, as will now be described.

[0036] An optical technique may be used to provide a reference measurement of the thickness of each step of the calibration structure, relative to a border region or edge of the body that has no coating. A known optical coherence tomography (OCT) technique may be used to provide a mean layer thickness measurement and details of the surface shape of the structure. OCT is a non-contact interferometric system using a low coherent laser source, which may be adapted to map a shape of a transparent and low diffusive media, which provides multiple scattering paths. The border region having no coating is used as a reference point to determine a difference in thickness between each step of the coating compared to the body surface. Thickness measurements using OCT may be accurate to 7pm or more, which is of sufficient accuracy to calibrate paint thickness measurements.

[0037] Other optical methods may be used to obtain a thickness measurement of the coating, such as laser-profile measurements based on triangulation. A direct measurement may be performed by analysing the surface shape of the structure.

[0038] Since the thickness of the coating may change over time between application and curing, the calibration may be repeated at multiple times to determine the time-varying nature of the coating thickness after application. The thickness of a freshly applied coating layer will become thinner as the liquid components evaporate to leave only solid constituents when cured. For example, when dry, it may be expected that the thickness of a paint may be reduced by up to a half due to solvent evaporation and the cured thickness may be further affected by use of a hardener, for example. The curing time may depend upon the volume of paint applied (for example, the thickness of a single applied coat) to the structure in one go or on environmental factors such as temperature. The curing time for aeronautical paint may typically be three to four hours. The use of generation and detection lasers may help to evaporate solvent in solvent-based paints, to reduce the paint drying time and save time in the curing process.

[0039] A laser ultrasonics calibration may then provide a peak resonance frequency for a given coating thickness, which has been determined via the OCT measurements (or by other methods). The laser ultrasonics calibration may be performed multiple times, for example, at substantially the same times and over the same stepped structure as the OCT measurements. Processing of the reflected signals using FFT provides a peak resonance frequency associated with each thickness step of the coating for each test time. The peak resonance frequencies may then be correlated with the respectively timed OCT measurements. In this way, each OCT thickness measurement can be correlated with a peak resonance frequency, and the relationship may be used to determine the respective acoustic velocity of the ultrasonic waves in the coating, as will be described.

[0040] In addition, processing of the reflected signal using Inverse Fast Fourier Transform (IFFT) of the back-wall echoes provides the time-peaks associated with each thickness step of the coating for each test time, which can be correlated with the respectively timed OCT results. Hence, a peak resonance frequency or time-peaks analysis of the calibration structure may be generated, with different values for each thickness step of the coating. It is found that the peak resonance frequency or repetition rate of the ultrasonic signal for a 2.5cm aluminium plate is around 3MHz, and this remains constant through the calibration testing and so can be treated as a reference peak resonance frequency.

[0041] The calibration structure may be covered with a protection layer, for example of around 20pm, to assist in the generation of the ultrasonic signal within the structure.

[0042] An example of a calibration curve is shown in Figure 4B. The curve was generated using the foregoing calibration process for a coating comprising cured paint according to the structure in Figure 4A. However, as indicated, the calibration may typically be performed on any material coating in any state of dryness. A painting process may in addition, or alternatively, be calibrated based on variables such as paint type, pressure or shape of the painting jet, projection time, working distance or environmental factors.

[0043] The calibration curve in Figure 4B can be seen to provide a conversion between layer or coating thickness (in microns) and an associated peak resonance frequency (in Megahertz) at a particular time. One or more further calibrations may be performed at different times according to the state of the coating (from wet to dry). The calibration thereby provides multiple curves (only one of which is shown) each related to a conversion between layer or coating thickness and the associated peak resonance frequency at a different time after the coating has been applied. The calibration shows that, in general, the lower the thickness of the coating or layer is, the higher the peak resonance frequency is.

[0044] The results of the calibration process may be applied to determine the coating thickness on any structure, for example a painted aeronautical part for use in the aerospace industry, given the same coating material and drying conditions.

[0045] As indicated, the curve in Figure 4B is fitted to data measured for each step thickness of Figure 4A and provides a velocity of the ultrasonic waves through the coating. The relationship between the peak resonance frequency and coating thickness is described by equation (1) below, where/is peak resonance frequency, v is the velocity of ultrasonic waves in the coating, and x is the coating thickness.

Equation (1) [0046] In this example, the acoustic velocity of ultrasonic waves through dry paint may be determined to be about 600m/s.

[0047] The velocity of ultrasonic waves through paint, for example as obtained using the calibration described above and performed on a calibration structure comprising an aluminium plate, is required to determine physical characteristics of a structure comprising a body having a coating of paint thereon. The above calibration (performed on a structure comprising a metal) provides the velocity of ultrasonic waves through a coating such as aeronautical paint. Physical properties of the structure, such as determination of the thickness of the coating and determination of the integrity of the body may then be determined using laser ultrasonics (without further use of other techniques such as OCT).

[0048] The calibration may of course be performed on a structure comprising a body other than a 2.5cm thick aluminium plate. The reflected signal 301 may be analysed to determine characteristics of the reflected signal, of which the characteristics determined depend upon the thickness of the body or coating of the structure. In any event, the calibration may comprise determining a mean value of the velocity of ultrasonic waves in the body based on analysis of a series of time-peaks associated with each layer of the structure.

[0049] Figure 5 is a flow chart showing in greater detail steps for processing a detected reflected signal (generated using laser ultrasonics) to determine a first and a second parameter of the structure. The detected signal in the time domain, as shown in Figure 3, undergoes signal processing. At block 501 a region of the reflected signal comprising the back-wall echoes (shown as B in Figure 3) is isolated for processing. The spectral content of the back-wall echoes is selected in this way since this region of the spectrum contains the information of the multiple reflections that occur inside the coating and body of the structure. Isolating the region in this manner improves the signal-to-noise ratio for the following peak resonance frequency analysis. At block 502 an FFT analysis is performed on the isolated waveform. The FFT analysis is performed when using a frequency-domain analysis and may not be required if using a time-domain analysis. The FFT analysis identifies the separate frequency components of the reflected signal to provide a series of peak resonance frequencies in the frequency domain. At block 503 a peak resonance frequency associated with each layer is determined. The intensity and time-of-flight information (which is related to respective frequencies) of the peaks assists in the determination of the peak resonance frequency associated with each layer. Once the peak resonance frequency of each layer has been identified, the calibration data may be referenced, at block 504, to convert the peak resonance frequency to a thickness of the associated layer. At block 505 the thickness x of a layer is determined by equation (1), from the peak resonance frequency/and the velocity v. The calibration data provides the velocity of ultrasonic waves through the coating and/or body to allow the conversion from peak resonance frequency / to thickness x of the associated layer. In addition, at block 505, according to embodiments herein, structural integrity information associated with the body of the structure may also be determined. For example, material density, the presence of defects, defect size, surface properties, and/or a thickness of the body are among properties that may be determined. This information is also obtained from the measured properties of the reflected signal, such as velocity of the ultrasonic waves in a particular medium or material such as that for the body, amplitude of the peaks, time-of-flight, frequency, attenuation and/or loss among other properties. For example, resonances of the ultrasonic wave within the body may be represented by a “family” of frequency peaks (or time-peaks in the time domain), where the frequency and intensity of each peak, and/or the rate of decay of the intensity for each subsequent reflection, contributes to an overall impression of the properties of the structure.

[0050] While referencing the calibration data is typically not needed, for example, for conducting a structural analysis of the body of a structure, it serves to assist in identifying frequency peaks that are related to body and coating thicknesses and, thereby, to identify other peaks that are related, for example, to structural integrity.

[0051] The signal processing of Figure 5 may be performed using non-destructive testing software. A “smoothing” window or FFT gate (or IFFT gate) may be applied to the reflected signal and/or filtering may be used to improve the signal-to-noise ratio for the peak resonance frequency analysis.

[0052] A calibration performed on aeronautical paint applied to the aluminium plate, for example as described with reference to Figures 4A and 4B, allows determination of the thickness of an aeronautical paint coating on an aluminium body of the structure. However, other coatings and body material combinations may be calibrated in substantially the same manner. Information on the integrity of the body of the structure may be determined whether a coating is present or absent. Information on the physical properties of the coating and body may be determined whether the coating is wet or dry (or somewhere in between these two extremes). When, for example, a metal body is used in the calibration structure for the calibration process, a large difference in acoustic impedance between the coating and body layers provides resonance peak frequencies or time-peaks of high intensity, making it easier to identify and differentiate between the peaks. However, when a composite body is used in the calibration structure for the calibration process there may be an additional requirement to provide a calibration of a representative composite structure without the coating in order to characterise and identify associated peak characteristics. This is because the lower acoustic impedance between the coating and composite layers produces peak resonance frequencies or time-peaks having much lower intensities, making it relatively more difficult to differentiate between the peaks associated with each layer. The calibration obtained from the composite without coating provides “signature” peaks for the composite layer to assist identification of peaks associated with a coating when a coating is present. A relatively larger proportion of the ultrasonic signal generated is coupled into and propagates through composite materials, in contrast to a relatively smaller proportion of the ultrasonic signal that is coupled into and propagates through metallic materials, due to a larger reflectance between the coating and metal layer, resulting in turn from a large difference in acoustic impedance. For this reason, as discussed above, if laser ultrasonics is performed to determine a coating thickness on a structure comprising a metal, the generation and laser properties will typically be selected to be different to those properties required for a structure comprising a composite material. As such, when a calibration structure comprising a composite material is used for performing the calibration, as described above with reference to Figures 4A and 4B, the ultrasonic signal propagating through both the coating and composite layer must be separated for each layer for signal processing. That is, the velocity of ultrasonic waves through the composite body may be determined using a calibration of a structure comprising a composite without a coating.

[0053] By way of example, when the calibration, as described above with reference to Figures 4A and 4B, is instead performed on a calibration structure comprising a composite material, a curve fitted to the data measured for each step thickness of a coating may provide a velocity of the ultrasonic waves through the coating and body. That is, the peak resonance frequency is a function of the velocity of ultrasonic waves in the coating and body, and thickness of the coating and/or body, where/ =/(vc, v/>, x) and where / is the peak resonance frequency, vc is the velocity of ultrasonic waves in the coating, v* is the velocity of ultrasonic waves in the body, and x is the coating thickness and/or body thickness.

[0054] Figures 6A-6C show graphical representations of illustrative exemplary peak resonance frequency analyses.

[0055] Figure 6A illustrates a processed reflected signal from a structure comprising a body without a coating. A peak resonance frequency associated with the body may be determined in the absence of a peak resonance frequency associated with a coating. The absence of any significant higher-frequency peaks is indicative of a lack of structural imperfections.

[0056] Figure 6B illustrates a processed reflected signal from a structure comprising a body having a coating thereon. Peak resonance frequencies may be determined for the coating and body, and the peak associated with the coating may indicate the thickness of the coating. The peak resonance frequency of the body will generally be lower than the peak resonance frequency of the coating, as the body will tend to be thicker than the coating.

[0057] Figure 6C illustrates a processed reflected signal from a structure comprising a body having a coating thereon, wherein the body comprises a flaw or defect. As in Figure 6B, the peak resonance frequency of the body will generally be lower than the peak resonance frequency of the coating. A defect present in the body may be characterised by a peak (or peaks) having a higher frequency than the peak resonance frequency of the body and may be higher or lower than a peak resonance frequency associated with the coating, depending on whether the depth of the defect within the body is, respectively, smaller or larger than the thickness of the coating. The peak resonance frequency associated with the defect may for example be determined and analysed to assess an extent of any associated integral damage to the body of the structure. For example, the depth and extent of the defect in the body, and/or size of the defect may be determined by looking to the peak resonance frequency across one or more readings in one or more different areas. The presence of a defect or crack in the body modifies the propagation of the ultrasonic signal through the body since the distance that the ultrasonic waves propagate may depend on the shape and size of the defect.

[0058] A particular region of the reflected signal may be isolated or selected for analysis by applying a time or frequency window or gate (in software) over just the features of interest, for example to provide an optimised signal-to-noise ratio for analysis of the coating or body of the structure. For example, a window may be applied across region B of Figure 3 A to analyse the coating, or applied across a similar region C of Figure 3 A to analyse the integrity of the body. Depending on the nature of the body or substrate material (composite, metal or other non-composite) the settings for the signal processing may be different. As such, laser ultrasonics provides a method whereby a coating thickness measurement and non-destructive integrity testing may be achieved via signal processing of the same reflected signal measurements. Signal processing may be simultaneously performed using a plurality of different gates. For example, a gate may be localised around a first back-wall echo using an FFT analysis in the frequency-domain, and a gate may be localised around the first back-wall echo using an IFFT analysis in the time-domain. The signal processing may be performed on the first back-wall echo or a plurality of subsequent back-wall echoes (not shown in Figure 3 A).

[0059] Figure 7A shows a schematic diagram of an articulated robotic arm 701 and scanning head 702 that may be used to perform a laser ultrasonics measurement on an inspection area 703 of a structure. The inspection area may be around 1.5m x 1.5m (but could be larger or smaller). The technique may be performed using a robotic arm 701 with at least one head 702 on the end of the arm to provide a multi-purpose tool. The head may perform the laser ultrasonics on the inspection area as described above using mirrors (not shown) to scan the area and collect measurement data in steps. The robotic arm 701 may operate at a distance of up to 2m or more from the surface of the structure.

[0060] In some embodiments, the robotic arm may comprise a further head 704 carrying spraying equipment or the like for applying a coating to the body. As illustrated in Figure 7B, a feedback loop 706 fed by a coating thickness measurement, which is determined by a programmed microprocessor arrangement 708, may then be used to control the further head 704 and ensure that an appropriate volume and thickness of coating is applied to the structure. The microprocessor arrangement 708 may be programmed to control the movement of the robotic arm 701, the recording and processing of laser ultrasonic data to determine coating thickness and the operation of the further head 704.

[0061] Known methods of spray painting of aircraft or aircraft parts may be performed using a high volume low pressure spray gun in which a current of compressed air is blown over a hollow needle immersed in a paint mixture. The paint is forced through a small orifice and the air current causes it to be atomised into a fan of fine droplets which are propelled on the substrate. The droplets coalesce into a paint film where the thickness of the film may be changed by adjusting parameters such as paint to air ratios, pressure, orifice diameter among others.

[0062] For example, a paint application process may be automated using methods described herein based on a real-time feedback system that dynamically monitors the thickness of a coating applied to an aircraft or aircraft part. For example, depending on the laser ultrasonics determination of coating thickness, if a nominal thickness has been reached, the painting process may be stopped or continued with modified or adapted settings. Controlling the painting thickness at early stages of the painting process, when the paint is still wet, for example, may save on costs associated with the painting process since the quantity and thickness of paint is optimised.

[0063] The height of the robotic arm may be adjusted to allow measurement of a larger or smaller inspection area. This allows for separate structures or parts to be placed within the inspection area and to be measured in the same data collection scan. Use of the robotic arm allows for a very large area to be inspected and is suitable for assessing large components used in the aerospace industry.

[0064] The invention has been described herein by way of example only with reference to a structure comprising a body having a coating thereon. However, the invention may be used to determine physical characteristics of a structure comprising a plurality of bodies having a coating thereon, wherein, for example, each body is formed of a different material.

Claims

CLAIMS:

1. A method compri si ng: illuminating with a generation laser an area of a structure, comprising a body and a coating thereon, to generate at least one ultrasonic signal in the thermo-elastic regime within at least the coating; illuminating the area with a detection laser to produce at least one reflected signal comprising waveform characteristics that are influenced by the ultrasonic signal; detecting the reflected signal(s); and determining, from waveform characteristics of the reflected signal(s), a first parameter related to a physical characteristic of the coating and a second parameter related to a physical characteristic of the body.

2. A method according to claim 1, wherein the first parameter relates to the thickness of the coating.

3. A method according to either preceding claim, wherein the second parameter relates to an integrity of the body.

4. A method according to any preceding claim, wherein the characteristics of the reflected signal(s) comprise time-of-flight information.

5. A method according to any preceding claim, wherein the detecting of the signal is performed when the coating is in a fluid state.

6. A method according to any preceding claim, wherein the coating comprises a paint.

7. A method according to any preceding claim, wherein the body comprises a composite part.

8. A method according to any preceding claim, wherein the body comprises a non-composite part.

9. A method according to any preceding claim, wherein the coating comprises aeronautical paint and the body comprises an aeronautical part for use in the aerospace industry.

10. A method according to any preceding claim, comprising using calibration data to determine the first parameter from the characteristics of the reflected signal(s).

11. A method according to claim 10, wherein the calibration data relates to coating thickness.

12. A method according to claim 10 or claim 11, wherein the calibration data provides a relationship between an acoustic velocity of the ultrasonic signal within the coating and the thickness of the coating.

13. A method according to claim 12, wherein the relationship is time-dependent.

14. A method according to any preceding claim, wherein determining the second parameter using characteristics of the reflected signal(s) comprises determining physical characteristics of the structure such as material density, presence of a defect, size of defect, surface properties and/or thickness.

15. A method according to any preceding claim, wherein the method is performed using an apparatus comprising: at least one head carrying the generation and/or the detection laser(s), the head or heads being mounted on an articulated robotic arm, which is movable to move the head relative to the structure.

16. A method according to claim 15, wherein a first head is used for applying the coating to the body and a second head is used for carrying the generation and/or the detection laser.

17. A method according to claim 16, further comprising feeding back from the second head to a processor acquired data relating to a thickness of an applied coating, the processor being arranged to control the operation of the first head in order to attain a desired thickness of coating by controlling a volume of coating that is applied to the body.

18. An apparatus comprising: a generation laser to illuminate an area of a structure, comprising a body and a coating thereon, to generate at least one ultrasonic signal in the thermo-elastic regime within at least the coating; a second laser to illuminate the area to produce at least one reflected signal comprising waveform characteristics that are influenced by the ultrasonic signal; a detector to detect the reflected signal(s); and a processor arranged to determine, from waveform characteristics of the reflected signal(s), a first parameter related to a physical characteristic of the coating and a second parameter related to a physical characteristic of the body.

19. An apparatus according to claim 18, further comprising: an articulated robotic arm; and at least one head on the end of the robotic arm carrying the generation and/or the detection laser.

20. An apparatus according to claim 19, wherein a first head is used for applying the coating to the body and a second head is used carrying the generation and/or the detection laser.

21. An apparatus according to claim 20, comprising a feedback loop to feed back from the second head to a processor acquired data relating to a thickness of an applied coating, the processor being arranged to control the operation of the first head in order to attain a desired thickness of coating by controlling a volume of coating that is applied to the body.