PIEZOELECTRIC SENSOR, SYSTEM AND METHOD FOR MONITORING THE INTEGRITY OF STRUCTURES

Abstract

The present invention relates to a piezoelectric sensor (1,100) comprising a piezoelectric material (10) interposed between a first (11) and a second (12) electric contact element. The first electric contact element (11) comprises at least two sensing areas (110, 111) spatially separated along a sensing direction. It also describes a sensor node that includes the piezoelectric sensor, a system and a method for monitoring the integrity of a structure using said piezoelectric sensor.

Claims

1. Piezoelectric sensor comprising a piezoelectric material interposed between a first and a second electric contact element, wherein the first electric contact element comprises at least two sensing areas spatially separated along a sensing direction.

2. Sensor according to claim 1, wherein the sensing areas of the first electric contact element are spatially separated and have a shape such that their distance of said separation varies according to the sensing direction considered.

3. Sensor according to claim 2, in which the sensing areas are electrically connected.

4. Sensor according to claim 2, in which the sensing areas are electrically insulated and the first electric contact element comprises two distinct electrodes, each of said distinct electrodes comprising a respective output terminal each one connected to a respective sensing area of the first electric contact element.

5. Sensor according to claim 1, wherein a first of the said at least two sensing areas has the shape of a segment of an Archimedean spiral or of a logarithmic spiral.

6. Sensor node comprising a piezoelectric sensor according to claim 1, and a processing unit designed to digitize an electrical signal generated by the piezoelectric sensor in response to the detection of an elastic wave.

7. Sensor node according to claim 6, wherein the processing unit is adapted to transmit the digitized signal to a remote device.

8. Sensor node according to claim 6, wherein the processing unit is adapted to calculate the direction of origin and/or the energy of the elastic wave and to transmit data on the direction of origin and/or the energy of the elastic wave to a remote device.

9. Sensor node according to claim 7, further comprising means for receiving and/pr transmitting data over optical fibre.

10. Sensor node according to claim 7, further comprising a memory for temporarily storing data received on a data input.

11. System for monitoring the integrity of a structure, comprising plurality of sensor nodes for detecting elastic waves in the monitored structure, a central processing system operatively connected to said sensor nodes and designed to process data transmitted by said sensor nodes, wherein each of said sensor nodes comprises a piezoelectric sensor according to claim 1, and a processing unit designed to digitize an electrical signal generated by the piezoelectric sensor in response to the detection of an elastic wave which is propagated in the structure, said processing unit being also designed to transmit the digitized signal to a remote device.

12. System according to claim 11, wherein said remote device is one between a sensor node of said plurality and the central processing system, wherein the processing unit comprises an input for receiving data transmitted from another sensor node and is designed to retransmit from an output the data received on said input.

13. System according to claim 11, wherein the processing unit is designed to transmit the identifier of the sensor node to which it belongs and temporal information on when the electric signal generated by the piezoelectric sensor was detected, and in which the central system is designed to receive the digitized signal, the temporal information and the identifier transmitted from each sensor node, for each sensor node to calculate the direction of origin and/or the energy of said wave.

14. System according to claim 13, wherein the central system comprises a memory area containing information on the position and shape of said plurality of sensor nodes and is designed to calculate a point of impact and/or the energy of an object's impact with the structure.

15. System according to claim 11, wherein the sensor nodes are designed to receive and transmit data over optical fibre.

16. (canceled)

17. Method for monitoring the integrity of a structure, in which a wave guided within the structure is detected by means of a piezoelectric sensor, and a direction of origin of the guided wave is detected, wherein the piezoelectric sensor is a sensor according to claim 1, and wherein the direction of of origin of the guided wave is detected by a single piezoelectric sensor.

18. Method according to claim 17, wherein the two sensing areas of the first electric contact element are electrically connected to one another, and wherein the method provides for: calculating the autocorrelation envelope of the signal generated by the piezoelectric sensor or by a signal derived from the one generated by the piezoelectric sensor, extracting the second peak of the autocorrelation signal; determining the direction of origin of the guided wave on the basis of the position of the second peak of the autocorrelation signal.

19. Method according to claim 17, in which the two sensing areas of the first electric contact element are electrically connected to one another, and wherein the method involves: compensating the signal generated by the piezoelectric sensor upon detection of a guided elastic wave within the structure for any dispersion suffered by the elastic wave, calculating the autocorrelation envelope of the compensated signal, extracting the second peak of the autocorrelation signal, determining the direction of origin of the guided wave on the basis of the position of the second peak of the autocorrelation signal.

20. Method according to claim 19, wherein in the event of a multimodal propagation of the elastic wave within the structure the compensation of the signal is effected using the dispersion curves of the most energetic method.

21. Method according to claim 17, wherein the two sensing areas of the first electric contact element are electrically connected to one another, and wherein in the event the structure consists of an anisothropic and dispersive material the method involves: characterizing the structure by detecting, for N different preselected directions within the structure, different dispersion curves for different propagation modes of an elastic wave in the N different preselected directions, applying N times an algorithm to compensate the dispersion of the signal generated by the piezoeletric sensor in such a way as to generate N compensated signals, each time considering the dispersion curves for the most energetic of the propagation modes of the wave along the preselected direction taken into consideration calculating the autocorrelation signal of the N compensated signals for each of the N calculated autocorrelation signals, extracting the second peak of the envelope of said autocorrelation signals, among the extracted second peaks, determining the one of maximum amplitude, determining the direction of origin of the guided wave on the basis of the position of the second peak of maximum amplitude.

22. Method according to claim 17, wherein the two sensing areas are electrically isolated from one another and constitute separate electrodes, and wherein the method involves: calculating the cross-correlation envelope of the two signals generated by the piezoelectric sensor when a guided wave within the structure arrives on the two sensing areas, extracting the maximum peak of the cross-correlation envelope, determining the direction of origin of the guided wave on the basis of the position of the maximum peak of the cross-correlation envelope.

23. Method according to claim 17, wherein the two sensing areas are electrically isolated from one another and constitute separate electrodes, and wherein the method involves: compensating the two signals generated by the piezoelectric sensor upon detection of a guided elastic wave within the structure for any dispersion suffered by the elastic wave, calculating the autocorrelation envelope of the two compensated signals extracting the maximum peak of the autocorrelation signal determining the direction of origin of the guided wave on the basis of the position of the maximum peak of the autocorrelation signal.

24. Method according to claim 23, wherein in the event of a multimodal propagation of the elastic wave within the structure the compensation of the two signals is effected using the dispersion curves of the most energetic method.

25. Method according to claim 17, wherein the two sensing areas of the first electric contact element are electrically isolated from one another, wherein the piezoelectric sensor generates a pair of electric signals upon detection of an elastic wave, wherein in the event the structure consists of an anisothropic and dispersive material, the method involves: characterizing the structure by determining, for N different preselected directions within the structure, different dispersion curves for different propagation modes of an elastic wave in the N different preselected directions, applying N times an algorithm to compensate the dispersion of the pair of signals generated by the piezoeletric sensor in such a way as to generate N pairs of compensated signals, each time considering the dispersion curves for the most energetic of the propagation modes of the wave along the preselected direction taken into consideration, calculating N cross-correlation signals of the N pairs of compensated signals, for each of the N calculated cross-correlation signals, extracting the maximum peak of the relevant envelope, among the N maximum peaks extracted, determining the one of maximum amplitude, determining the direction of origin of the elastic wave on the basis of the position of the peak of maximum amplitude.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The invention will be described hereinafter with reference to non-limiting examples, provided by way of explanatory and non-limiting examples in the appended drawings. These drawings illustrate different aspects and embodiments of the present invention and, where appropriate, reference numerals illustrating structures, components, materials and/or similar elements in different figures are indicated by similar reference numerals.

[0032] FIG. 1 shows a top view of a piezoelectric sensor according to one embodiment of the present invention.

[0033] FIG. 2 shows a sectional view of the sensor in FIG. 1.

[0034] FIG. 3 schematically shows the detection of two wavefronts being detected by the sensor in FIG. 1.

[0035] FIG. 4 shows the autocorrelation signals of the signals generated by the sensor in FIG. 1 in response to the waves shown in FIG. 3.

[0036] FIG. 5 shows a sensor node that contains the sensor in FIG. 1.

[0037] FIG. 6 shows an alternative sensor node to that in FIG. 5.

[0038] FIG. 7 shows an alternative sensor node to those in FIGS. 5 and 6.

[0039] FIG. 8 shows a flow diagram of a method for calculating the direction of origin of an elastic wave detected by a piezoelectric sensor according to the invention.

[0040] FIG. 9 shows a system for monitoring the integrity of a structure.

[0041] FIG. 10 shows a flow diagram of a method for calculating the position and energy of an elastic wave generated by the impact of an object on the structure.

[0042] FIG. 11 shows a flow diagram of a method for calculating the position and energy of an elastic wave generated by the impact of an object on the structure in the case of a structure in anisotropic material.

[0043] FIG. 12 shows a top view of an alternative piezoelectric sensor to that in FIG. 1, with electrically isolated sensing areas.

[0044] FIG. 13 shows the cross-correlation signals of the signals generated by the sensor in FIG. 12 in response to the waves shown in FIG. 3.

DETAILED DESCRIPTION OF THE INVENTION

[0045] While the invention is susceptible to various modifications and alternative constructions, some preferred related embodiments are shown in the drawings and will be described below in detail. It should be understood, however, that there is no intention to limit the invention to the specific illustrated embodiments, but, on the contrary, the invention intends to cover all the modifications, alternative constructions, and equivalents as fall within the scope of the invention as defined in the claims.

[0046] The use of “for example”, “etc.”, “or” indicates non-exclusive alternatives without limitation unless otherwise indicated. The use of “includes” means “includes, but is not limited to” unless otherwise indicated.

[0047] Use of the term “gain” for a transfer function is meant to indicate the ratio between the amplitudes of the output and input signals; a gain can therefore be greater than 1 (amplification) or lower than one (attenuation).

[0048] Use of the term “sensing direction” is meant to indicate any direction along which the piezoelectric sensor is able to detect the arrival of an elastic wave. The sensing direction may coincide with or be different from the direction of origin of the wavefront, given that a wavefront may impact the piezoelectric sensor even if it comes from a direction of origin that is external to the sensing lobe of the piezoelectric sensor.

[0049] FIG. 1 shows a top view of a piezoelectric sensor 1 that is particularly suitable for use in networks for monitoring the integrity of structures.

[0050] The sensor 1, whose section along the axis AA is shown in FIG. 2, comprises a substrate 10 of piezoelectric material and a pair of contact elements 11 and 12, in conductive material, e.g. copper, placed on two opposite surfaces of the substrate 10. In the example in FIG. 1, each contact element is an electrode, however, according to other embodiments described below, each contact element may comprise a number of electrodes.

[0051] The electrodes 11 and 12 can be formed in a manner known per se, for example by depositing a metal layer on the piezoelectric substrate 10, and may be of a variable shape and thickness according to the principles better described hereunder.

[0052] In the example of FIGS. 1 and 2, the sensor 1 is configured to detect the arrival of guided waves within an angle of 90° identified in FIG. 1 by the axes x and y. The electrode 12, placed along the inferior surface of the substrate 10 is a single piece of constant thickness such as to include the area of the electrode 11, while the electrode 11 includes two sensing areas 110 and 111 spatially separated along a sensing direction AA.

[0053] In this embodiment, the two sensing areas 110 and 111 are kept in electrical contact by means of an electrical connection, for example a copper wire, 112, preferably external to the sensing angle of the sensor. The presence of the electrical connection is not essential to the operation of the sensor, but its presence or absence requires a different treatment of the output signal from the sensor.

[0054] The sensing areas 110 and 111 have a shape such that their distance (preferably measured between the outermost points, i.e. those that may first encounter an elastic wave propagating through the monitored structure, such as the points P110 and P111 in FIG. 1 along the direction AA) varies when the sensing direction taken into consideration changes. In the example of FIG. 1, this is achieved by means of a first sensing area 110 of circular shape and a second sensing area 111 whose shape corresponds to a portion of a logarithmic spiral. Clearly in order to obtain the angular variation of the distance between the sensing areas of the electrode 11 other forms, such as Archimedean spirals, can be used, and it is possible to vary not only the shape, but also the thickness or the number of conductive areas which constitute the first electrode.

[0055] In an embodiment not shown in the Figures, in addition to the electrode 11 the electrode 12 also can have a contoured shape and/or comprise one or more spatially separated sensing areas. The combination of different shapes and thicknesses for the two electrodes 11 and 12 can be used differently to obtain as a result a pair of electrodes formed by sensing areas differently spaced along different sensing directions.

[0056] The shape of the sensor 1 permits it to distinguish the angle of arrival of a wavefront, for example due to the impact of an object with the structure monitored by the sensor. FIG. 3 shows two wavefronts, F1 and F2, which impact from different directions of origin (identified respectively by the angles θ1 and θ2) on two different contact points (P1 and P2) on the sensing area 111 of the electrode 11 (events Ea1 and Ea2).

[0057] The wavefronts then impact the second sensing area (events Eb1 and Eb2). The time interval (deltaT) between events Ea1 and Ea2 and events Eb1 and Eb2 depends on propagation speed of the elastic wave and on the distances D1 and D2.

[0058] In response to each of the two wavefronts, the piezoelectric sensor 1 outputs (i.e. to the electrodes 11 and 12) a different potential difference, according to the direction of origin of the wavefront due to the different deltaT that are determined.

[0059] FIG. 4 shows the trend of the autocorrelation envelope (Ia1 and Ia2) of the two signals generated by the wavefronts F1 and F2. As can be seen from FIG. 4, Ia1 and Ia2 both have a main origin peak and a secondary peak (hereinafter also referred to as “2nd peak”) whose distance from the main peak depends on the point of impact of the wavefront with the sensing area 111.

[0060] In order to distinguish the direction of origin of the wavefront, the output of the sensor 1 is connected to the input of a processing unit 2 which determines the direction of origin of the wavefront.

[0061] In a preferred embodiment, the piezoelectric sensor 1 is connected to the processing circuit 2 through an impedance matching circuit 3 as shown in FIG. 5.

[0062] In the example of FIG. 5, the piezoelectric sensor 1 is connected to a processing unit 2 through an impedance matching circuit, which in this embodiment is constituted by a field-effect transistor, in particular a JFET whose gate and source are connected to the electrodes 11 and 12 of the sensor.

[0063] Again with reference to FIG. 5, to supply the processing unit 2 a power supply 4 is provided.

[0064] The power supply 4, the processing unit 2, the impedance matching circuit 3 and the piezoelectric sensor 1 are preferably mounted on a same substrate, such as a same PCB (Printed Circuit Board) or more preferably the same substrate of flexible plastic (such as polyamide, PEEK film or transparent conductive polyester). Thus mounted and assembled, these elements constitute a sensor node that can be used as an element of a network for monitoring the integrity of a structure. Alternatively, the various elements of the sensor node can be separated and maintained in electrical connection by means of conventional wiring, for example of twisted copper pairs. The solution of integrating the above elements as much as possible into a single sensor node, however, has the advantage of reducing the weight of the overall structure.

[0065] In one embodiment, shown in FIG. 6, the impedance matching circuit 30 is a filter circuit. The piezoelectric sensor can be represented as a voltage source in series with a capacitor whose capacity depends on the characteristics of the sensor. This capacity of the piezoelectric sensor is connected to the node N1, where the capacitor C1 and the resistor R1 are connected. The entire circuit 30, and the capacity of the piezoelectric sensor, constitute a Bessel filter of the second order, in which the capacitor C1 is connected between the node N1 and the non-inverting input of an operational amplifier 31, the output of which is shorted on the inverting input. One resistor R1 is connected between the node N1 and the inverting input of the operational amplifier 31, while another resistor R2 is connected between the non-inverting input of the operational amplifier 31 and an earth wire which is also connected to the piezoelectric sensor 1.

[0066] Knowing the characteristics of the piezoelectric sensor 1 and its equivalent circuit, it is possible to size the resistors R1 and R2 and the capacitor C1 in order to define the bandwidth of the filter. This is particularly advantageous in the case of monitoring the integrity of structures that, during their operation, may be subject to vibrations, for example in the case of monitoring the fuselage of an aeroplane that is subjected to vibrations induced by the engines. Such vibrations are characterized by specific frequencies that corrupt the signal and which must therefore be filtered out. Additionally, the filter itself can be used for an anti-aliasing function before the sampling and the digitizing take place.

[0067] The circuit in FIG. 6 foresees connecting to the same earth the electrode 12 of the sensor 1, the impedance matching circuit and the control unit 2. This type of circuit is preferably applied if the sensor is applied on conductive materials, such as aluminium.

[0068] In the event of the sensor being applied on materials that are not good electrical conductors, for example for application on a laminate of carbon fibre (where the resin that aggregates the carbon fibres is an insulator), it is preferable to the use an impedance matching circuit of the type shown in FIG. 7.

[0069] In FIG. 7, the impedance matching circuit 300 is a circuit for active adaptation, wherein to each of the two electrodes (11 and 12) of the sensor 1 is connected an impedance matching circuit of the type in FIG. 6, namely a circuit comprising an operational amplifier with an inverting input that is short-circuited at output, a first resistance (R1, R3) connected to the inverting input of the operational amplifier, a capacitor (C1, C2) connected to the first resistor (R1, R3) and to the non-inverting input of the operational amplifier, a second resistor (R2, R4) connected to the non-inverting input of the operational amplifier. The resistors R2 and R4 are connected to each other. As per the example in FIG. 6, in practice there are two second-order Bessel filters which determine the bandwidth of the signal provided at the input to the processing unit 2.

[0070] Regardless of the impedance matching circuit used, when a wavefront affects the piezoelectric sensor 1, this generates a potential difference which, as stated above, depends on the direction of origin of the wavefront.

[0071] In order to distinguish the direction of origin of the wavefront, the control unit 2, for example a microprocessor or a microcontroller, executes a method described hereinafter with reference to the flow diagram of FIG. 8.

[0072] Firstly, the signal generated by the piezoelectric sensor 1, and possibly filtered by an impedance matching circuit, e.g. of the type illustrated in FIG. 6, is digitized (step 801) by the processing unit 2.

[0073] Then the calculation of the autocorrelation envelope of the signal thus processed is made, and the extraction (step 802) of its second peak. This operation can be performed by detecting the position of the maximum of the autocorrelation envelope in the time interval that corresponds to the extreme values of deltaT. These extremes are determined by taking into consideration the minimum and maximum distances between the sensing areas of the sensor and multiplying these distances by the speed of propagation.

[0074] On the basis of the position of the second peak (PSP), the processing unit 2 calculates (step 803) the angle of arrival of the wavefront (for example the angles θ1 and θ2 in FIG. 3) detected by the piezoelectric sensor 1. The calculation is made by multiplying PSP by the wave propagation speed in order to obtain the distances that correspond to D1 and D2 in the examples taken into consideration, from which the angular directions can be derived with geometrical considerations related to the shape of the spiral. In particular, in the case of the logarithmic spiral, the fact that the arms of the spiral widen with a geometric progression and that the angle “α” between the radius of the spiral at a point and the tangent to the spiral at that same point is constant, are exploited. In a preferred embodiment, the processing unit is configured to transmit to a remote device a scalar value representative of the direction of impact detected by the sensor 1. Preferably, in addition to this scalar value, the processing unit 2 is configured to transmit to the remote device an identification datum of the sensor node (or sensor 1 if the latter is not integrated into a sensor node as described above) and a temporal indication relating to the moment of receipt of the elastic waves. Even more preferably, the control unit is also configured to transmit to the remote device self-diagnosis data of the control unit or the sensor node in general. While in the above described example each processing unit 2 is configured to calculate the direction of the wavefront, it is, however, possible to provide a sensor node with a simpler configuration that does not calculate the direction of the wavefront, leaving the calculation of the direction of origin of the wavefront to a remote device. In this embodiment, which reduces the computational complexity and therefore the cost and power consumption of the sensor nodes used for monitoring the integrity of a structure, the control unit does no more than digitize the signal generated by the piezoelectric sensor and transmit this signal, for example via a fibre optic connection, to a remote device where the calculation of the direction of origin of the wavefront takes place.

[0075] FIG. 9 shows an example of a network of sensors for monitoring the integrity of a structure. One or more of the network elements, and preferably all the sensor nodes, are integrated into the structure to be monitored, in particular they are inserted in the surface layers or in the layers immediately underlying the surface layers, of the structure to be monitored.

[0076] In the example in FIG. 9, the network 900 includes a central processing system 901, and a plurality of access points 902 that collect the data transmitted by the sensor nodes 903 and transmit them to the central processing system 901. The system in FIG. 9 thus allows the implementation of a method for monitoring the integrity of a structure as described below and illustrated in FIG. 10.

[0077] In the example of FIG. 9, each sensor node comprises an input and an output for respectively receiving and transmitting data. Preferably, the transmission is effected by optical fibre, thus each sensor node 903 is provided with suitable devices, such as photodiodes, for receiving and transmitting fibre optic signals.

[0078] In the example in FIG. 9, each sensor node 903 reacts to the presence of elastic waves in the monitored structure, for example due to an impact of an object with the monitored structure, by transmitting a digital optical signal (step 1001).

[0079] In the preferred embodiment, each sensor node 903 transmits both the digital signal that carries the information on the wave detected, and an identification code of the sensor node.

[0080] In the example in FIG. 9, each sensor node 903 transmits these data to the adjacent sensor node 903, which stores them temporarily and transmits them to the adjacent sensor node until they reach the access node 902 which transmits them to the central processing system 901. Other types of connection for the sensor nodes 903 and other systems and protocols for data transmission in the network 900 are however possible.

[0081] The central processing system then receives the digitized signals and the identification codes of the sensor nodes 903 of the network 900.

[0082] Since in the propagation of elastic waves in structures such as plates and shells there can be dispersion of the elastic wave, within the network 900 suitable forms of compensation of said dispersion can be provided based on known per se algorithms, such as the algorithm described in “A passive monitoring technique based on dispersion compensation to locate impacts in plate-like structures”, L De Marchi, A Marzani, N Speciale, E Viola, Smart Materials and Structures 20 (3), 035021. In this paper it is shown how compensation of the dispersion has the effect of generating waveforms for which it is possible to associate propagation distances at time intervals deltaT as in the case of non-dispersive propagation.

[0083] Such compensation modules (hardware or software) are preferably included in or connected to the central processing system 901, which proceeds therefore to compensate the dispersion of the acquired signal (step 1002) to generate a corresponding compensated signal, however, it is possible to provide dispersion compensation modules inside the sensor nodes also.

[0084] The central processing system 901 includes an area of memory that collects information on the various sensor nodes, such as their position in the structure, their shape and the mathematical relation that binds the signal generated by the piezoelectric sensor to the direction of origin of the wavefront. In this way, the central processing system 901 is capable of calculating (step 1003), for every sensor node 903, the autocorrelation of the received digital signal and consequently determine (step 1004) the direction and energy of the wavefront detected by each sensor node 903 of the network 900.

[0085] With this information (knowledge of the position of the sensor nodes, the direction and energy of the wavefront detected), the central processing system 901 determines (step 1005) the point of impact and the impact energy from the measurements of the various sensor nodes 903.

[0086] There is also the possibility of providing that the sensor node independently determines the direction and energy of the wavefront that activates the piezoelectric sensor 1, as illustrated previously with reference to FIG. 8. In this case the sensor node outputs information regarding its identifier, the direction and the energy of the wavefront detected. The central system then determines the position and energy of the impact by using the information supplied by the various sensor nodes.

[0087] In the case of anisotropic materials, the wave generated by an impact with the monitored structure propagates differently in different directions. In order to calculate the point and the energy of an impact, the method of monitoring a structure described above must be partially modified and a preliminary phase in which the structure is characterized is necessary. In this characterization phase, N (with integer N greater than 1) different pre-selected angular directions within the structure are identified and different dispersion curves of different modes of propagation of an elastic wave in the N pre-selected angular directions are calculated.

[0088] FIG. 11 shows a flow diagram of a method for monitoring the integrity of a structure made of anisotropic materials.

[0089] The method involves use of a plurality of sensor nodes connected in a network, for example as shown in FIG. 9.

[0090] Each sensor node is configured to digitize (step 1101) the signal generated by the respective on board piezoelectric sensor 1 and transmit, preferably by optical fibre, the signal to an adjacent sensor node.

[0091] In the example in FIG. 11, the central processing system receives digital signals with information relating to readings taken by the m sensor nodes 903 of the network 900.

[0092] For each of the m received signals, the central processing system performs dispersion compensation (step 1102) repeating same N times for the N different predetermined angular directions selected during the characterization of the structure. At the end of this phase, there will therefore be N compensated signals for each of the m signals transmitted by the m sensor nodes.

[0093] Subsequently, for each sensor node, the central processing system (step 1103) calculates the autocorrelation for each of the N compensated signals and extracts N second autocorrelation peaks for the N predetermined angular directions. From among these N second autocorrelation peaks the central system selects the one with the maximum amplitude and, on the basis of this value, determines (step 1104) the direction of origin of the wavefront with respect to each of the m sensor nodes. In this case also, it is possible to provide modules which carry out the dispersion compensation and calculate the direction of origin of the wavefront within the sensor nodes themselves.

[0094] Finally, from the position of the m sensor nodes, the direction and the energy of the wavefront detected by each sensor node, the central processing system determines (step 1105) the point of impact and the energy of the impact.

[0095] In light of the above it is clear how the piezoelectric sensor, the system and method for monitoring the integrity of a structure described above permit the attainment of the proposed aims, allowing the structure to be monitored effectively.

[0096] It is therefore clear that numerous variants of the sensor and of the system and method described above can be implemented by a person skilled in the art, without departing from the scope of protection as defined by the attached claims.

[0097] For example, in one embodiment, the two sensing areas which constitute one of the electrodes of the piezoelectric sensor can be electrically separated, that is to say that there is no electrical connection 112 as in the example of FIG. 1.

[0098] In this case, shown in FIG. 12, the piezoelectric sensor 100 includes two sensing areas 110 and 111 which are electrically separated and constitute independent electrodes, to which the common electrode 12 is added. The piezoelectric sensor 100 will thus have two output terminals U1 and U2, at the ends of which there will be two different sensing signals resulting from the same incident elastic wave.

[0099] In order to verify the direction of origin and the energy of the wavefront, it is necessary to calculate (in the processing unit 2 or in the central system 901, depending on the embodiment) the cross-correlation between the two signals outputted by the sensor 100. The cross-correlation of these signals has a maximum whose position varies according to changes in the direction of origin of the wavefront, so in order to determine the direction and energy of the wavefront it is necessary to calculate the cross-correlation between these signals and extract the maximum cross-correlation.

[0100] The methods for monitoring the integrity of a structure described above with reference to FIGS. 8, 10 and 11 will thus be modified by providing for the calculation of the maximum of the cross-correlation between the pairs of signals generated by the sensor 100 instead of the calculation of the second autocorrelation peak of the signal generated by the sensor 1.

[0101] Also the cases of the dispersive and anisotropic propagations will be addressed with solutions similar to those described above for the solution based on the areas not electrically separated, taking care to replace the processing steps based on autocorrelation with those based on the cross-correlation of the pairs of signals collected.

[0102] It will therefore be possible to monitor a structure in an anisotropic and dispersive material by means of a piezoelectric sensor in which the two sensing areas of the first electric contact element are mutually electrically isolated, in which the piezoelectric sensor generates a pair of electrical signals upon the detection of an elastic wave, and by including the following steps: [0103] characterizing the structure by identifying, for N different preselected directions within the structure, different dispersion curves for different modes of propagation of an elastic wave in the N preselected directions, [0104] applying N times a compensation algorithm from the dispersion to the pairs of signals generated by the piezoelectric sensor in such a way as to generate N pairs of compensated signals, each time considering the dispersion curves of the more energetic of the modes of propagation of the wave along the preselected direction taken into consideration, [0105] calculating N cross-correlation signals of the N pairs of compensated signals, [0106] for each of said N calculated cross-correlation signals, extracting the maximum peak of its envelope, [0107] from among the N extracted maximum peaks, determining the peak with the greatest amplitude, [0108] determining the direction of origin of the elastic wave as a function of the position of the peak of maximum amplitude.

[0109] FIG. 13 shows the trend of the cross-correlation envelope (I×1 and I×2) of the two signals generated by the wavefronts F1 and F2 of the type illustrated in FIG. 3. As can be appreciated from FIG. 13, Ix1 and Ix2 both have a main peak whose distance from the origin depends on the point of impact of the wavefront with the sensing area 111.

[0110] Although the examples described above have presented a piezoelectric sensor capable of detecting guided waves in a sensing interval of 90°, it is clear that by following the principles explained above piezoelectric sensors with different sensing ranges can be realized.

[0111] Although in the embodiments described above the electrode 12 has been described as a conductive layer of constant thickness, this requirement is not essential and the electrode 12 may have other shapes and also be composed of two or more electrically connected parts. In every case, the electrode 12 is preferably disposed on a surface area that is greater than or equal to that occupied by the first contact element 11.