The present invention relates to a machine and a method for working sheet metal, plates, strips, slabs of metal and the like to obtain semi-finished and/or finished products. More specifically, the invention relates to a working machine for punching and/or bending and/or cutting, mechanically or by means of a laser, a sheet metal, which is provided with a sensing device for characterizing, i.e., for determining the properties or characteristics of the sheet metal or metal piece, and capable of setting optimal working parameters. The invention further relates to the related method for working sheets, for example punching or bending or laser cutting, based on the characteristics of the sheets detected by the sensing device.

With reference to the laser cutting process of metal sheets, known laser cutting machines generally comprise an optical source suitable for emitting a laser beam, an optical guide system capable of guiding and focusing in a work area the laser beam emitted by the source, and a work head which determines the power density of the laser beam required to cut the piece, typically a sheet metal. A moving system allows the relative movement between the laser beam exiting the work head and the piece to be cut, in particular by adjusting speed and direction of such movement.

A laser work head is generally provided with a nozzle through which a working gas is emitted which creates a gas cloud around the work area to activate a chemical reaction in the piece and to remove slag and molten material generated by the laser cutting. The pressure, shape, and speed of the working gas flow on the piece may be controlled by appropriately varying the nozzle shape, the distance thereof from the sheet metal surface, and the gas pressure at the nozzle outlet.

The quality and stability of the laser cutting process depend directly on a plurality of working parameters including, for example, shape and position of the laser beam on the piece, optical power of the laser radiation, flow and pressure of the working gas on the piece, and travel speed of the laser beam with respect to the piece.

With reference to the sheet metal punching process, punching machines are known which are provided with single or multiple press punching heads, i.e. comprising a single tool or punch or a plurality of tools or punches which are linearly actuated so as to interact with the piece (sheet metal) which must be worked by relative presses, typically consisting of linear hydraulic actuators (hydraulic cylinders).

To properly perform the punching process on the piece, it is necessary to check the working parameters which include an operating stroke and a speed of the punching tool, as well as a punching force exerted by the latter on the piece.

With reference to the sheet metal bending process, bending machines are known, also called bending presses, comprising a mechanically or hydraulically driven press capable of moving an upper tool or punch so that it rests against a lower tool or die on which the piece is positioned. The punch exerts a force on the piece which can deform and bend the piece according to an angle determined by the configuration of the tools themselves.

Typically, the punch includes a blade with rounded edges which is capable of deforming the piece along a predefined bending line. The piece is locked in position on the work surface of the machine by suitable clamping means.

Also in these machines, the quality and precision of the bending process depend on an accurate control of the working parameters, which mainly include a position and an operating stroke of the punch and a bending force exerted by the latter on the piece.

For all the working machines described above, i.e. for punching, bending or laser cutting sheet metal, it is known that the working parameters must be set and adjusted according to the physical and/or chemical properties or characteristics of the sheet metal to be worked, such characteristics including material type (steel, stainless steel, aluminium, copper, brass) and material composition (alloy composition), thickness, surface conditions (treatment, finishing) of the sheet metal.

A common approach in the industry is to find the optimal working parameters for most of the typical sheet metal used in production, conducting in-depth empirical testing and/or with theoretical methods, and storing these parameters in a material database of the machine or the CAD/CAM software which generates working programs performed by the machine. The operator of the machine or CAD/CAM software selects such a group of parameters.

In particular, the process parameters are set by the operator in the machine before the working, based on the generic type of material and the thickness of the piece or sheet metal to be worked known a priori. However, due to human error and/or limited knowledge of the operator, these parameters may be inadequate or inaccurate as they may not take into account all the characteristics of the piece, in particular the material thereof. Therefore, the working process is not optimized for the piece, to the detriment of the quality and stability of the entire working process. In addition, undesirable consequences may include damage to the piece and machine and reduced productivity due to downtime for the measures necessary to correct the error, as well as a high number of rejects.

These drawbacks are more evident and critical when the working machine is integrated into an automated production line in which vertical warehouses and loading systems are provided upstream of the machine which supply the latter with the raw sheet metal, while unloading systems are provided downstream of the machine to remove the finished pieces from the latter and move them further along the line. The entire manufacturing process is fully automated and does not require any periodic intervention by the machine operators, the production line being remotely controlled. Therefore, due to the lack of machine operators, an incorrect or inaccurate setting of the process parameters may go unnoticed for a long or even indefinite period of time.

In addition, errors may occur in this automated production line during the programming and/or loading of the warehouse or within the supply system which may cause the loading of a sheet metal not suitable for the intended working process. Again, because there are no machine operators, errors of this type may go unnoticed for a long or even indefinite period of time. The result is the production of waste and/or a complete stoppage of the production line.

Analysis methods are known which allow the characterization of the material, i.e. the detection of properties and characteristics of the material in relation to chemical composition, structure and surface conditions. However, such analysis methods, which typically include chemical processes or mass spectrometry analysis, have several drawbacks which limit the integration thereof into automated machines or production lines.

A first drawback is the high cost of the devices capable of performing such analysis methods. A second drawback is that the detection must be completed within a period of time of a few seconds in order not to interrupt the normal operating flow of the machine. Third, due to the measurement mode the piece may suffer slight or even substantial damage. Fourth, the analysis of the material must take place directly on the working machine, shortly before the working itself, otherwise any further manipulation of the material may lead to the loading of the incorrect material.

WO 2011/134805 describes a method for determining the laser workability of metal sheets using atomic emission spectroscopy as an analytical method. In atomic emission spectroscopy, a small sample of the piece is ionized into a plasma. During the plasma cooling, the recombination of charge vectors causes the emission of radiation with characteristic wavelengths. The spectral distribution of the light emitted allows to deduce the elementary composition of the piece. Typically, the plasma excitation and the ablation of the material to be analysed are performed remotely, with a laser pulse of the length of a few nanoseconds or with an electrical spark. Atomic emission spectroscopy is a known method for remote material analysis and has the capacity to determine the elementary composition of a piece as well as the mass or volumetric proportions of the elementary components thereof. Such method is very expensive, as it requires a high resolution spectrometer and a laser source or a precise current source to create a reproducible spark. Furthermore, the laser source of the working machine described in the aforementioned document is designed to emit laser radiation as a high-power continuous wave, and is thus incapable of creating reproducible optical pulses with high response rates required to create the plasma necessary for atomic emission spectroscopy. The plasma created by the laser source of the machine can be too hot at the time of measurement, thus emitting an almost continuous spectrum which hides the emission spectra characteristic of the elementary composition of the piece. Thus, a second laser source near the high power source of the machine may be required to perform the measurements.

Finally, a major drawback of this method is that an accurate analysis of the composition of the raw material is only possible if a portion of material is removed prior to measurement. Therefore, this method causes considerable damage to the piece.

Another analysis method known for the material characterization is fluorescence X- spectroscopy. In fluorescence X- spectroscopy, the piece particles are excited by X-rays, and then emit secondary or fluorescent X-rays characteristic of each element of the material composition. An energy resolution detector subsequently measures such fluorescent X-rays. This method allows to quickly determine the elemental composition of a piece only of the surface layers of the piece if no material is removed before the measurement.

Nevertheless, the use of a fluorescence X-spectroscopy system in a sheet metal working machine is problematic and expensive due to the need to meet radiation protection requirements (radiation protection means and procedures).

Other more or less complex solutions have been developed to allow the characterization of sheet metal material and therefore the optimization of the working parameters in the working machine. However, these solutions are generally quite complex and expensive, and difficult to use in a sheet metal working machine, and in any case they are not very reliable due to the characterization, sometimes even approximate, only of the surface layer of the piece.

An object of the present invention is to improve the machines and methods known for working sheet metal, plates, strips, metal bars and the like, and in particular the machines and methods for punching and/or bending and/or cutting sheet metal, in particular by means of a laser.

Another object is to provide a machine and a method for working a sheet metal which allow to characterize, i.e. to determine the properties or characteristics of the sheet metal to be worked, in particular of the material thereof, and to automatically set the optimal working parameters based on the characteristics detected, without any intervention by the operator. A further object is to provide a machine and a method for working a sheet metal, in which the characteristics of the sheet metal to be worked, in particular type of material, composition, thickness, finishing and surface treatment, can be accurately and quickly detected without damaging the piece.

In a first aspect of the invention a machine for working sheet metal according to claim 1 is provided.

In a second aspect of the invention a method for working sheet metal according to claim 11 is provided.

The invention will be better understood and implemented with reference to the attached drawings, which illustrate an exemplary and non-limiting embodiment thereof, in which: figure 1 is a schematic front view of the machine of the invention for working a sheet metal, provided with a sensing device for characterizing a piece or sheet metal supplied in the machine; figure 2 is a schematic view of the components of a preferred embodiment of the sensing device of figure 1 associated with the piece, shown in cross section; figure 3 is a schematic view of the components of an alternative embodiment of the sensing device of figure 1 associated with the piece, shown in cross section.

Referring to figures 1 and 2, the sheet metal working machine 1 of the invention comprises operating means 2 for performing at least one working on a piece 50 of metal, a sensing device 3 capable of non-destructively detecting a plurality of characteristics of said piece 50, in particular physical, chemical, mechanical, dimensional characteristics, and providing the related data, and a control unit 4 connected to the operating means 2 and the sensing device 3, and suitable for configuring and/or adjusting the working parameters to control the operating means 2 acting on the piece 50 based on the data received from the sensing device 3, i.e. based on the characteristics of the piece 50.

The piece 50 is a sheet metal, plate, band, slab, plate, or the like, made of metal material, such as non-alloy or alloy steels, aluminium, copper, brass, etc.

In the embodiment illustrated in figure 1 as a non-limiting example, the sheet metal working machine 1 is a laser cutting machine and the operating means 2 comprise a laser cutting head 20 which is movable along three orthogonal movement directions X, Y, Z by means of moving means 21 of a known type and therefore not described in further detail.

Referring to figure 2, the sensing device 3 comprises at least one eddy current sensor 5 for detecting electrical and magnetic characteristics of the piece 50, at least one electromagnetic acoustic sensor (EMAT) 6 for detecting at least one thickness of the piece 50 and at least one optical sensor 8 for detecting characteristics of a surface 50a of the piece 50.

The sensing device 3 may also comprise a temperature sensor 10, for example a pyrometer or an infrared thermometer, for detecting a temperature of the piece 50.

The sensing device 3 further comprises a processing unit 11 configured to coordinate the measurement process performed almost simultaneously by the sensors 5, 6, 8, 10, receive from the sensors a group of data related to the plurality of detected characteristics of the piece 50, process the said group of data to identify the material and a set of process characteristics of the piece 50 (i.e. characteristics which influence and condition the working or working process of the piece) and send identifying data of the material (e.g. type of metal alloy) and data related to the set of process characteristics (e.g., thickness, surface finishing, rolling direction, material supply status, Young's module, shear modulus, surface with protective film, etc.) to the control unit 4 so that the latter is capable of configuring and/or appropriately adjusting the working parameters of the operating means 2, in particular based on the information on the piece 50 (identifying data of the material and process characteristics of the piece) provided by the sensing unit 3, to perform a correct settled working on the piece 50.

In a preferred embodiment of the machine 1 of the invention, the processing unit 11 is configured to identify, or detect or determine, the material and the process characteristics of the piece 50 by means of a machine learning system or machine learning model of a known type, capable of processing the group of data related to the plurality of characteristics of the piece in order to identify the type of material and the process characteristics most corresponding to said group of data.

As is known, a machine learning system comprises a plurality of mathematical and statistical algorithms which, exposed to a certain set of data in an initial step defined “training” and passing through a second evaluation step of the results with optimization of the parameters, are capable of independently obtaining a function or correlation capable of identifying in a different set or group of data (e.g. data related to the characteristics of the piece), the value most likely corresponding thereto (type of material, process characteristics).

In particular, the processing unit 11 is configured to process the group of data related to the characteristics of the piece 50, in particular reducing the complexity of the data, by means of a Principal Component Analysis (PCA) method and/or by means of a Linear Discriminant Analysis (LDA) method thus obtaining a set of processed data and is configured to identify the material and the process characteristics of the piece 50 by classifying the group of processed data by means of a classification algorithm of the machine learning system. Alternatively, the processing unit 11 may comprise a database containing groups of characteristics of a plurality of respective reference materials and is configured to identify the material and the process characteristics of the piece 50 by comparing the plurality of characteristics detected almost simultaneously by the sensors 5, 6, 7, 8 of the sensing device 3 with the groups of characteristics contained in the database.

In the embodiment illustrated in figure 1, the sensing device 3 is positioned in the machine 1 upstream of the operating means 2 with respect to a supplying direction A of the piece 50 entering the machine 1, and is close to and facing a surface 50a, in particular a top surface, of the piece 50. The piece, for example, is positioned on a working plane 31 of the machine 1 by means of a loading system 30 of the machine 1.

The sensing device 3 is positioned in an on-board portion of the machine 1 which is close to the loading system 30, and fixed, by means of supporting means 12, to the working plane 31 or to a supporting frame 32 of the machine 1. Alternatively, the sensing device 3 may be associated with the loading system 30 of the machine 1.

The supporting means 12, which support the sensing device 3, can be provided with moving means 13 arranged to move the sensing device 3 along an adjusting direction B, in particular orthogonal to the working plane 31, to adjust a distance of the sensing device 3 from the piece 50.

The moving means 13 of the supporting means 12 can be controlled by the control unit 4 so as to maintain a constant distance of the sensing device 3 from the piece 50. The distance between the sensing device 3 and the piece 50 can be automatically detected by a distance sensor, which is connected to the control unit 4.

The eddy current sensor 5 is capable of measuring the electrical conductivity and the magnetic permeability of the material of the piece.

More precisely, the eddy current sensor 5 is configured to generate an alternating primary excitation magnetic field MF1, which creates eddy currents EC in the portion of the piece 50 facing the sensing device 3; the eddy currents EC in turn generate a secondary magnetic field MF2, which provides feedback or retroaction to the eddy current sensor 5 to measure the electrical conductivity and the magnetic permeability of the material of the piece.

An excitation coil of the eddy current sensor 5 generates the primary excitation magnetic field MF1, while the secondary magnetic field MF2 is detected by a dedicated receiving coil or by the excitation coil itself. The value of the eddy currents EC is measured by measuring the amplitudes and phase shift of two signals relating to the two magnetic fields MF1, MF2, which can be represented as a complex value of the voltage.

Since the value or strength of the eddy currents EC created in the piece 50 depends on the electrical conductivity and permeability of the material of the piece, the eddy current sensor 5 allows a precise and accurate measurement of such characteristics of the piece 50.

It should be noted that some common types of materials, in particular non-alloy and/or low- alloy steels, differ only slightly in terms of electrical conductivity and magnetic permeability in the presence of a specific field force and/or measurement frequency, and may therefore be difficult to distinguish. However, the permeability curve often differs with increasing field forces and/or frequencies of the excitation magnetic field.

Thus, the eddy current sensor 5 is also configured to perform the eddy current EC measurement not only at a fixed frequency and/or field force of the primary excitation magnetic field MF1, but at several separate test frequencies and/or test field forces, or within a range of test frequencies and/or test field forces.

The eddy current sensor 5 also allows to detect inhomogeneity in the material resulting from layer structures which may influence the working performed by the operating means 2.

As is known, the frequency of excitation of the primary excitation magnetic field MF1 generated by the eddy current sensor 5 determines the penetration depth of the primary excitation magnetic field MF1 in the piece 50. Thus, by varying the excitation frequency of the primary excitation magnetic field MF1 it is possible to control the penetration depth of the primary excitation magnetic field MF1 and subsequently the measurement depth, the variation of this penetration depth allowing to detect material inhomogeneity.

According to the embodiment illustrated in figure 2, the sensing device 3 comprises an electromagnetic acoustic sensor (EMAT) 6, which is capable of exciting, i.e. generating, and receiving ultrasonic waves in the conductive metal piece.

As is known, the excitation or generation of ultrasonic waves takes place by means of induction of an alternating eddy current in an object to be measured, which is simultaneously penetrated by a static magnetic field. The static field exerts a Lorentz force on the eddy currents, which causes the excitation of mechanical vibrations in the object, the characteristic of said mechanical vibrations being correlated to the physical and mechanical characteristics of the material of the object, in particular the elastic modulus thereof and intrinsic mechanical tension thereof.

The static magnetic field may be generated by a permanent magnet or preferably an electromagnet. The field generated by an electromagnet may be static in relation to the measurement duration only. At the times in which no measurement is in progress, the electromagnet may be switched off, thereby reducing the power consumption and heating of the device. The alternating magnetic field for the induction of alternating eddy currents can be generated by a coil. The arrangement of a coil in a static magnetic field can also detect the mechanical vibrations in the object of measurement. In this case, the movement (vibration) of the object in a static magnetic field leads to the induction of alternating eddy currents in the same object, whose magnetic field is converted into a voltage detectable by the coil.

By suitably choosing the coil assembly structure and the static magnetic field, it is possible to excite and detect various ultrasound wave modes and propagation modes, comprising longitudinal and transverse waves, as well as surface waves, for example Rayleigh and Lamb waves.

By measuring the propagation time of the ultrasound waves along the thickness of the object according to the impulse-echo principle (ultrasound waves generated and received by the electromagnetic acoustic sensor 6), a thickness of the object can be measured by knowing the speed of sound in the specific material of the object. The distinctive sound velocity of a material may be determined by means of prior empirical studies on reference materials or by data reference books.

Alternatively, in the embodiment illustrated in figure 3, the sensing device 3 may comprise a first electromagnetic acoustic sensor, or first EMAT 6, acting as an electromagnetic acoustic transducer with transmitter function, and a second electromagnetic acoustic sensor, or second EMAT 7, acting as an electromagnetic acoustic receiver, said first electromagnetic acoustic sensor 6 and said second electromagnetic acoustic sensor 7 being positioned at a settled distance from each other.

The first EMAT 6 is arranged to generate first ultrasound waves in the piece 50 along a first direction L in the extension of the piece (first ultrasound waves SW) and second ultrasound waves along a second direction M, through the thickness of the piece (second ultrasound waves LW).

In more detail, the first EMAT 6, particularly in a first step of the measurement procedure, generates first ultrasonic waves SW along the first direction L, particularly with a low vibration mode which minimizes undesirable effects due to dispersion.

The first ultrasound waves SW generated by the first EMAT 6 (transducer or transmitter) are detected by the second EMAT 7 (receiver) and a propagation time of said first ultrasound waves SW is measured from the transducer to the receiver. Based on the propagation time, knowing the settled distance d and the propagation mode of the first ultrasonic waves generated by the first EMAT 6, it is possible to accurately calculate the actual propagation speed of the waves or a speed of the sound which is distinctive of the material of the object. As is known, the distinctive speed of sound of the material depends largely on the modulus of elasticity and the intrinsic mechanical tension of the material. Thus, by measuring the distinctive speed of sound of the material by means of the two EMAT’s 6, 7 it is also possible to indirectly measure the modulus of elasticity and/or mechanical tension of the material and thus the type of material.

The first EMAT 6, in particular in a second step of the measurement procedure, also generates second ultrasound waves LW along the second direction M through the thickness of the piece 50, and measures a propagation time required of the second ultrasound waves LW to be reflected from a bottom surface of the piece 50 (echo effect) and to be received by the same first EMAT 6. The bottom surface is the surface opposite the (upper) surface 50a facing the sensing device 3.

Based on the speed of sound calculated previously and the propagation time of the second ultrasonic waves LW, a thickness of the object can be calculated by the processing unit 11 or by the control unit 4.

The optical sensor 8 is configured to detect the characteristics of the surface 50a of the piece 50, in particular a surface finishing, the presence of a protective plastic film, a colour.

In particular, the optical sensor 8 is configured to detect a colour of the piece surface 50a and/or to measure a magnitude of the optical reflectance of the piece surface 50a. Alternatively, the sensing device 3 can comprise a first optical sensor configured to detect the colour of the surface 50a and a second optical sensor configured to measure a magnitude of the optical reflectance of the surface 50a.

The sensing device 3 further comprises a light source apparatus 9 adapted to illuminate the piece 50 and cooperate with the optical sensor 8 to detect the characteristics of the surface 50a.

The light source apparatus 9 can emit white or multicoloured light which actively illuminates the surface 50a of the piece 50.

Preferably, the light source apparatus 9 comprises multiple white or multicoloured light sources surrounding the optical sensor 8 with a ring arrangement and are individually controllable in order to illuminate the piece from different angles with respect to the optical sensor 8, to measure the optical reflectance of the surface with different illumination angles and to better detect the characteristics of the surface 50a.

The use of the optical sensor 8 makes it possible to detect the characteristics of the surface, i.e. the surface finishing comprising mechanical surface treatments or finishing or surface coatings. The mechanical surface treatments include, for example, brushing, sandblasting, sanding, and rolling a material. The surface coatings can be laminations, anodizations, anti corrosive coatings, metal coatings and foils, but also films adhering to the surface after rolling/laminating .

It should be noted that the eddy current sensor 5 can also detect some metallic, and thus conductive, surface coatings and the thicknesses thereof by comparing the measurement values of the coated piece with the reference values known for the corresponding pure, uncoated piece.

The temperature sensor 10 is configured to detect a temperature of the piece 50 and the processing unit 11 is configured to calculate, based on said detected temperature, corresponding compensation parameters for the characteristics detected by the eddy current sensor 5, the electromagnetic acoustic sensor 6, 7 and the optical sensor 8.

In particular, the temperature measured by the temperature sensor 10 is used to compensate for the effects of the temperature in the eddy current sensor 5 and the electromagnetic acoustic sensor 6, 7.

The operation of the sheet metal working machine 1 of the invention provides an initial step in which the sheet metal or piece 50 to be worked is introduced into the machine along the supplying direction A by the loading system 30.

During the loading movement, the sensing device 3 - which is positioned in the machine 1 upstream of the operating means 2, close to and facing a surface 50a of the piece 50 - simultaneously detects a plurality of characteristics of the piece 50, in particular physical, chemical, mechanical, dimensional characteristics, by means of the sensors described above and comprising an eddy current sensor 5, capable of detecting the electrical conductivity and the magnetic permeability characteristics of the material of the piece, at least one electromagnetic acoustic sensor 6, 7 (EMAT) for detecting at least one thickness of the piece 50, and an optical sensor 8 associated with a light source apparatus 9 for detecting characteristics of a surface 50a of the piece 50, in particular surface finishing comprising mechanical surface treatments or finishing and surface coatings. The sensing device 3 may also comprise a temperature sensor 10, for example a pyrometer or an infrared thermometer, for detecting a temperature of the piece 50.

The processing unit 11 of the sensing device 3 coordinates the measurement process carried out almost simultaneously by the sensors 5, 6, 8, 10, receives from the sensors a group of data related to the plurality of detected characteristics of the piece 50, processes said group of data to identify the material and a set of process characteristics of the piece 50 and sends the identifying data of the material (e.g. type of metal alloy) and of the set of process characteristics (e.g. thickness, surface finishing, rolling direction, material supply status, Young's module, shear module G, surface with protective film, etc.) to the control unit 4 of the machine 1 which is thus capable of configuring and/or adjusting the working parameters to control the operating means 2 in order to optimally and correctly perform a required working on the piece 50, i.e. based on the identified material of the piece and the process characteristics of the piece 50 itself.

In particular, material and process characteristics of the piece 50 are identified by the processing unit 11 by means of a machine learning system or machine learning model of a known type, which processes the group of data related to the characteristics of the piece in order to identify the type of material and the process characteristics most corresponding to said group of data.

Alternatively, the processing unit 11 can comprise a database containing groups of characteristics corresponding to a plurality of respective reference materials and is configured to identify the material and process characteristics of the piece 50 by comparing the group of characteristics detected by the sensors 5, 6, 7, 8 of the sensing device 3 with the groups of characteristics contained in the database.

Therefore, the sheet metal working machine 1 of the invention allows to automatically characterize the piece 50 to be worked, i.e. to identify type of material and process characteristics thereof in order to automatically set and/or adjust the working parameters to control the operating means 2 and perform the settled working on the piece 50 with high quality and precision, without any intervention by the operators.

The method of the invention for working sheet metal in a sheet metal working machine 1 , which is provided with operating means 2 for performing at least on settled working on a piece 50 of metal and a sensing device 3, comprises the following steps: introducing a piece 50 to be worked and positioning the piece 50 close to the sensing device 3 that comprises at least a eddy current sensor 5, an electromagnetic acoustic sensor 6, 7 and an optical sensor 8; non-destructively detecting by means of the sensing device 3 a plurality of characteristics, in particular physical, chemical, mechanical, dimensional characteristics, of the piece 50; processing a group of data related to the plurality of characteristics of the piece 50 detected almost simultaneously by the sensing device 3 and identifying a material and a set of process characteristics of the piece 50; sending identifying data of the material and data related to the set of process characteristics of the piece 50 to a control unit 4 of the machine 1, said control unit 4 being connected to the operating means 2 to control the operating means while performing a settled working on the piece 50; configuring and/or adjusting working parameters to control the operating means 2 in order to perform the settled working based on the identifying data of the material and the data related to the set of process characteristics of the piece 50. The method provides processing the group of data and identifying the material and the process characteristics of the piece 50 by a machine learning system or machine learning model of a known type, capable of processing the group of data related to the characteristics of the piece 50 in order to identify the type of material, for example metal or metal alloy type, and process characteristics, for example thickness, surface finishing, rolling direction, Young's module, shear module, which most correspond to said group of data.

In particular, there is provided to process the group of data related to the plurality of characteristics of the piece 50, in particular reducing the complexity of the data, by means of a Principal Component Analysis (PCA) method and/or by means of a Linear Discriminant Analysis (LDA) method, obtaining a group of processed data and to identify the material and the process characteristics of the piece 50 by classifying the group of processed data (by PCA and/or LDA) by means of a classification algorithm of the machine learning system. Alternatively, the method provides identifying the material and process characteristics of the piece by comparing the plurality of characteristics of the piece 50 detected almost simultaneously by the sensors 5, 6, 7, 8 of the sensing device 3 with groups of characteristics of a plurality of respective reference materials contained in a database, in particular included in the processing unit 11.

Detecting the plurality of characteristics comprises detecting electrical and magnetic characteristics of the piece 50 by means of the eddy current sensor 5, at least the thickness of the piece 50 by means of the electromagnetic acoustic sensor 6, the surface characteristics, in particular surface finishing, of a surface 50a of said piece 50, by means of the optical sensor 8.

According to the method, detecting the thickness of the piece 50 comprises: generating in the piece 50, by means of the electromagnetic acoustic sensor 6, ultrasonic waves which propagate along a thickness of the piece 50 and which are reflected; receiving the reflected ultrasound waves by means of the same electromagnetic acoustic sensor 6; calculating a thickness of said piece 50 based on a propagation time of the ultrasonic waves and knowing a distinctive speed of sound of the specific material of the piece 50. Alternatively, according to the method, detecting the thickness of the piece 50 comprises: generating within the piece 50 first ultrasonic waves SW along a first direction L parallel to a longitudinal extension of the piece, by means of a first electromagnetic acoustic sensor 6 of the sensing device 3 acting as an electromagnetic acoustic transducer or transmitter; receiving the first ultrasonic waves SW by means of a second electromagnetic acoustic sensor 7 of the sensing device 3 acting as an electromagnetic acoustic receiver and positioned at a settled distance d from the first electromagnetic acoustic sensor 6; calculating a distinctive speed of sound of the material of the piece based on a propagation time of the first ultrasonic waves SW from the first electromagnetic acoustic sensor 6 and the second electromagnetic acoustic sensor 7 and on the settled distance d; generating, by means of said first electromagnetic acoustic sensor 6, second ultrasound waves LW along a second direction M orthogonal to the first direction L and parallel to a thickness of the piece 50, which are reflected and received by the same first electromagnetic acoustic sensor 6; calculating the thickness of said piece 50 based on the calculated distinctive speed of sound and a propagation time of the second ultrasound waves LW.

According to the method, detecting the electrical and magnetic characteristics of the piece 5 comprises generating by means of the eddy current sensor 5 an alternating primary excitation magnetic field MF1, which creates eddy currents EC in the portion of the piece 50 facing the sensing device 3; the eddy currents EC in turn generate a secondary magnetic field MF2, which provides feedback or retroaction to the eddy current sensor 5 to measure an electrical conductivity and a magnetic permeability of the material of the piece 50.

The method also comprises detecting characteristics of the material of the piece, particularly the modulus of elasticity and/or mechanical tension, by means of the pair of electromagnetic acoustic sensors 6, 7, particularly based on the distinctive speed of sound of the material of the piece measured by the electromagnetic acoustic sensors.

The method also comprises detecting a temperature of the piece 50 by means of a temperature sensor 10 and calculating, on the basis of the detected temperature, corresponding compensation parameters for the plurality of characteristics detected by the eddy current sensor 5, the electromagnetic acoustic sensor 6, 7 and the optical sensor 8. Detecting the surface characteristics of a surface 50a of the piece 50 by means of the optical sensor 8 comprises detecting a surface finishing, the presence of a protective plastic film, a colour of the surface 50a.

Thanks to the machine and the method of the invention for working sheet metal, it is therefore possible to automatically characterize the piece 50 to be worked, i.e. identify the type of material and determine the process characteristics thereof (which influence and condition the working of the piece), so as to automatically set and/or adjust the working parameters to control the operating means 2 and perform the working on the piece 50 with high quality and precision according to the characteristics detected, without any intervention by the operators.

This avoids errors in setting the working parameters to control the operating means that are due to human error and/or limited knowledge of the operator in relation to the programming of the machine. At the same time, it is also possible to avoid errors and/or malfunctions of an automatic supplying system which may cause the loading of a piece unsuitable for the intended work process, when the machine 1 is integrated into an automated production line.

It should also be noted that thanks to the machine and the method of the invention it is possible to perform quick and accurate detections and measurements of a plurality of characteristics of the piece, in particular of the type of material, composition, thickness, finishing and surface treatment, in a non-destructive manner, i.e. without damaging the piece.