BE1033112B1 - Method for determining a material property of a magnetizable metallic object - Google Patents

Abstract

The invention relates to a method for determining at least one material property and/or a magnetizable metallic object, in particular a method for determining a material stress in steel components, by means of a micromagnetic sensor arrangement (4) comprising at least one coil, in particular wound around a magnetic core, for generating magnetic field pulses and at least one receiver, comprising the steps of generating at least one magnetic field pulse based on at least one excitation pulse of the sensor arrangement for magnetizing the object, wherein the resulting magnetic field (H-field) undergoes a change of 150 kA/m/ms - 550 kA/m/ms, preferably 350 kA/m/ms - 550 kA/m/ms, recording measurement data in the form of at least one signal waveform resulting from the magnetization of the object in the receiver during the at least one excitation pulse, and determining the material property and/or the material state of the object, in particular a relative stress state of the object, based only on at least parts of the signal waveform during the at least one excitation pulse. recorded measurement data and an arrangement for determining a material characteristic and/or a material state of a magnetizable metallic object.

Description

-1- BE2024/5802

Method for determining a material property of a magnetizable metallic object

The invention relates to a method for determining at least one material property of a magnetizable metallic object and/or for determining a

The method describes the determination of the material state of a magnetizable metallic object using one or the same micromagnetic sensor arrangement. In particular, it is a method for determining material stress as a material state in steel components used to construct tanks or pipelines. One material property is the tensile strength of the object. The method can thus be used to investigate and determine material stresses in metallic objects, especially in the form of pipes or pipelines. The method is carried out using a micromagnetic sensor arrangement comprising at least one coil, in particular a coil wound around a magnetic core, for generating magnetic field pulses, and the sensor arrangement includes at least one receiver. The sensor arrangement is, in particular, part of a pipeline pig. It is understood that the device used in the method is further equipped with electronic or electrical means for controlling the coil and for recording and storing the measurement data via the receiver.

The measurement data is evaluated. Computer tools, in particular in the form of a [system/device], are used as means for controlling the sensor arrangement and for evaluation.

-2- BE2024/5802 or several electronic data processing units or systems (EDP unit or system) are used, which are appropriately trained to carry out the procedure.

Micromagnetic measurement methods, i.e., measurement methods using micromagnetic sensor arrangements, have been used for approximately 40 years for the non-destructive determination of

Material properties and stress states in ferromagnetic materials were developed and used. Micromagnetic measurement methods are non-destructive.

Testing methods that, unlike eddy current measurements, which are not intended to induce any changes in the magnetic state of the object being tested, are based on the ferromagnetic behavior of materials. These methods utilize the characteristic domain structure of ferromagnetic materials, in which the

Magnetization vectors of the domains are aligned by external magnetic fields.

From EP 4 080 232 A1, a measuring device and a method for determining the magnetic properties of a magnetizable test specimen using a high-current pulse are known. US 4,931,730 discloses a device and a method for generating and recording magnetization responses of ferromagnetic, ferrimagnetic, paramagnetic, and diamagnetic materials using an alternating magnetic field, wherein the pulses intended for magnetizing and demagnetizing have the same energy content.

-3- BE2024/5802

CN 1150318786A discloses a method for stress testing of ferromagnetic materials, in which the Barkhausen effect is utilized by means of square wave excitation. CN 104316594A describes a system for the non-destructive testing of steel parts for defects, which uses a special test head with an E-shaped magnetic core.

EP 1 792 173 B1 discloses a device and a method for detecting defects in objects or for locating metallic objects, in which a demodulation of an amplitude- and/or phase-modulated received signal is carried out, the demodulation being carried out not only in the vicinity of the carrier frequency but also at its harmonics.

Previous micromagnetic applications such as the HACOM method or 3MA technology typically use a sinusoidal excitation in the triple digits.

Frequency range with an amplitude in the ampere range around the ferromagnetic

The material is alternately magnetized. To measure the properties of the magnetic hysteresis, methods such as Barkhausen noise analysis or the analysis of the higher harmonics of the excitation(s) are used. Typically, 1-3 additional excitations with frequencies in the range of < 20 kHz are also used to generate additional parameters for determining the desired target quantity. The standard measurement process is therefore relatively lengthy.

-4- BE2024/5802 (- 10 -100 ms) and energy-intensive ( - 1-10 J/measurement) and therefore suitable for long

Pigs driven by pipelines with a length of dozens or hundreds of kilometers are not well suited for this method. The same applies, at least with regard to its length, to the method described in DE 10 2019 133799 A1.

It is therefore an object of the present invention to improve a previously described method with regard to the aforementioned disadvantages and in particular to make it faster and more energy-efficient.

According to the invention, at least one magnetic field pulse is generated by means of at least one excitation pulse of the sensor arrangement to magnetize the object, whereby the resulting magnetic field (H-field) undergoes a change of 150 kA/m/ms - 550 kA/m/ms, preferably 350 kA/m/ms - 550 kA/m/ms. It has surprisingly been found that the generation of significant measurement data depends solely on the magnitude of the change in the magnetic field. This previously unknown relationship is used according to the invention to generate precisely defined and comparatively short, and therefore low-energy, magnetic field pulses that enable the analysis of the material under consideration in the narrowband range.

The recording of measurement data in the form of at least one from the magnetization of the

The resulting signal waveform in the receiver is determined only during at least one excitation pulse. The determination of the material properties and/or the material condition of the object, in particular a relative [missing information], is not possible.

-5- BE2024/5802

The determination of the object's stress state is based only on at least parts of the measurement data recorded during at least one excitation pulse, i.e., in particular on at least parts of the measurement data recorded during one or two excitation pulses. The magnetic circuit created with the object is partly determined by the object's material properties, so that conclusions about the material properties can be drawn from the signal recorded in the receiver.

A special feature of the present invention is therefore that it has a specific functional range in which the desired information can be obtained. The specification of a voltage or current amplitude of the excitation pulse often cannot be generally used to determine the resulting magnetic flux, which, as is known, depends strongly on the geometry and the precise structure of the sensor arrangement. As a general, inventive

The functional range is therefore measured by directly measuring the change in the magnetic pulse(s) at the sensor assembly, for example, at the pole pieces of the sensor assembly. This results in the particularly preferred functional range for the change in the H-field of 150 kA/m/ms - 550 kA/m/ms and 350 kA/m/ms - 550 kA/m/ms. These values were obtained from a very large number of investigations in which satisfactory performance, i.e., a signal-to-

A noise ratio of, for example, >15 dB was obtained. It is noteworthy that the measurement performance of the inventive method depends on the maximum change in the magnetic field and, to a first approximation, is not determined by

-6- BE2024/5802 the value of the maximum field or the total energy, although these can further specify the invention. This finding is not known from the prior art.

The determination of the material characteristic value and/or the material condition is carried out by means of an evaluation device, which may at least partially be already formed by the sensor arrangement and/or an associated control unit or may at least partially be assigned to it.

The preceding or following indication of ranges (“from” and “to”, “between” and “and”) always includes, for the purposes of the registration, the boundaries of these ranges.

Preferably, two, and in particular only two, magnetic field pulses are generated by two excitation pulses of the sensor arrangement to magnetize the object, wherein the two excitation pulses are formed by two voltages with opposite polarity and/or currents in opposite directions. The two excitation pulses follow each other directly insofar as a time offset caused by switching is not taken into account or is used for the

Changes in the magnetic field are irrelevant to the measurement. In particular, the amplitudes of the voltage or currents remain nearly constant during this excitation. The required gradient of the magnetic field can be generated particularly well by using opposing excitations or magnetic field pulses.

“7 - BE2024/5802

During the first and second excitation pulses, measurement data in the form of at least one signal waveform resulting from the magnetization of the object are recorded in the receiver. This data is then used to determine the material properties and/or material condition. The measurement data can also be recorded during the switching of the excitation pulse. Specifically, the measurement begins with the first excitation pulse, particularly at the start of the excitation pulse, and ends with the second excitation pulse, particularly at the end of the second excitation pulse.

The coils usable for the invention can generally have any shape suitable for generating the desired magnetic pulse, in particular a “classic” coil wound around a magnetic core.

Excitation pulses represent a brief excitation of the sensor arrangement with a

This involves applying a current or voltage to generate a magnetic field pulse. Therefore, when using two (voltage) excitation pulses, it is assumed above or below that one is positive and the other is negative, regardless of whether a positive or negative pulse is used first. In particular, no measurement data recorded before the first and after the second excitation pulse are used to determine the material properties.

-8- BE2024/5802

The invention uses one or two short-term excitation pulses, in particular

Voltage pulses are used to stimulate magnetization. The measurement parameters are determined, in particular, by measuring the current (during the voltage pulse) in the

Receiver detected. The measurement is therefore significantly faster (in the range of a few milliseconds or microseconds) and significantly more energy-efficient than previous methods (< 0.1 J / measurement).

At the same time, the broadband operation with nearly constant amplitudes results

Excitation, particularly via a voltage pulse, excites a significantly larger number of frequencies within the material than excitation with sinusoidal oscillations using standard methods. This leads to a multitude of higher harmonics, which can be determined as a superposition of the response in the current signal. It has been shown that this additional information results in a higher signal-to-noise ratio for distinguishing between different steel materials than with standard sinusoidal excitation.

In comparison to eddy current techniques previously used on pipeline pigs, the use of a micromagnetic sensor arrangement has a similarly low energy consumption and a similar measurement speed, but additionally offers a significantly larger information content due to the defined change in magnetization.

Instead of imprinting a large number of magnetic field pulses, which essentially conditions the material, only two are used, in particular with at least partial

-9- BE2024/5802 uses magnetic pulses generated in the opposite direction. Furthermore, the evaluation focuses solely on the measurement data recorded exclusively during the excitation pulse(s) – which are immediately consecutive – so that excitation and measurement can be performed frequently in direct succession. For the evaluation, after the switch-off of the

No measurement data was recorded after the voltage generator, i.e., after the first excitation pulse when using one magnetic field pulse or after the second excitation pulse when using two magnetic field pulses.

For the evaluation of the measurement data stored in the associated sensor arrangement, this data can be evaluated, at least partially, directly within the sensor arrangement or a part of the evaluation device formed by or associated with it. For example, the sensor arrangement and part of the evaluation device can be located in a sensor head of a pipeline pig that moves along the inside of a pipeline wall.

Further components of the evaluation unit can be located in or formed within a central body of the inspection device, e.g., a pipeline pig, for example, in its control unit. Furthermore, for example, after an inspection has been carried out, the respective data set, including unprocessed measurement data or data that has already been processed at least partially, can be transferred to another electronic data processing system for further or complete evaluation.

- 10 - BE2024/5802

In particular, the excitation comprises only one positive and one negative excitation pulse with a total duration of no more than 10 ms, so that at a typical pipeline pig speed of 1 to 5 m/s, a resolution of at least 1 to 5 cm is achieved in the direction of the pipeline axis. A large number of sensor arrays arranged circumferentially around a pipeline pig body then provide a high-resolution representation of the stress states of a pipeline wall in both the longitudinal and circumferential directions. Accordingly, the measurement data then yields profiles of the respective values over the length of the object under consideration.

In particular, the voltage is measured by means of one of the sensor arrangement's associated sensors.

A voltage generator produces a time-varying current signal and measures it as the measurement signal, preferably using the coil (used for excitation) as the receiver, wherein the time-varying current signal due to the excitation is measured.

The current response is recorded. The setup is then correspondingly simple and requires less installation space. The current measured in the coil during excitation thus reflects the magnetic properties of the (ferromagnetic material) object. Alternatively or additionally, a second receiver can be used.

A sensor coil can be used, which can be arranged on the same core. Here again, the measurement of the [missing information] is preferred as the measurement parameter.

Current is used during the voltage pulse. The total duration of the

The pulse sequence is in the micro- to millisecond range. As already described, the information required for determining the pulse size is present in the temporal changes in current.

- 11 - BE2024/5802 can be used to determine the material condition, in particular the mechanical stress state, and the material properties of the material to be tested.

In general, the measurement data is derived from the measurement signals, regardless of whether voltages or currents are being measured.

Preferably, the magnetic field pulses are excited with an amplitude in the range of 5 V to 15 V, particularly with an amplitude between 11 V and 13 V. For example, the magnetic field pulse of a respective excitation pulse is excited with +12 V and -12 V or +10 V and -10 V.

The inventive method works particularly well with specially defined

Excitation pulses, particularly in the form of voltage pulses, which can be defined by their amplitude, the duration of the first (especially positive) and the duration of the second (especially negative) excitation pulse. Preferably, the first excitation pulse is longer than the second excitation pulse, in particular where the two excitation pulses have a pulse length ratio (first:second) between 2.5:1 and 0.75:1, in particular from 2:1 inclusive to 1:1 inclusive. This functional range was surprisingly discovered and determined through extensive experimentation.

“12 - BE2024/5802

According to a further development of the invention, the measurement data are obtained from two measurement sequences recorded separately during each excitation pulse, so that a measurement sequence generated during a switching of the voltage generator is obtained.

The signal is not recorded or used. Therefore, a zero crossing of the signal is not necessarily detected. It is understood that a measuring amplifier is used to condition the measurement signals, if necessary. Alternatively, measurements can also be taken during switching.

When using two magnetic field pulses, it is advantageous for the excitation to reverse the direction of the magnetic pulse, which has a positive effect on the magnitude of the change in the magnetic field.

According to a beneficial further training, the evaluation includes a decomposition of the

The measurement signal, or the measurement data, is transformed into its frequency components, or their frequency components, preferably by means of a wavelet transform, a Fourier transform, a Hilbert transform, autoregressive modeling, or eigenvalue analysis, in particular wherein information in a frequency range of 0.5 to 50 kHz of the transformed signal is used. Preferably, the information is used in a frequency range of a fast Fourier transform.

The advantage of these evaluation methods is the discretization and compaction of the information and the separation of the information from the noise of the electronics or the measurement. In particular, the determination of the material property value is carried out from

“13 - BE2024/5802 at least one component of the transformation function based on the maximum

Change in the magnetic field signal.

Preferably, a Fourier transform, in particular a Fast Fourier transform, is used.

Transformation is already performed on an integrated circuit (FPGA - Field Programmable Gate Array) within the sensor array, so that only the usable portion of the recorded data needs to be stored. This allows for initial signal processing, for example, directly in the sensor head of a pipeline pig. Simultaneously, the number of potential sources of interference is reduced by shortening the cable lengths required to transmit the measurement signal or data to the evaluation device. The sensor array can therefore incorporate part of the evaluation device or be designed in conjunction with it.

The compacted information based on the decomposition of the measurement signal or the

The measurement data, when used with this excitation method, yields approximately 50 to 200 usable parameters, a large number of which are used to determine the material properties.

Parameters can be used. Typically, the use of two to four of these parameters is sufficient to determine the voltage state. Additionally, the partial evaluation already performed within the sensor assembly significantly reduces the amount of data to be stored, as the raw data does not need to be saved.

“14 - BE2024/5802

Preferably, the stress state and/or tensile strength is determined using the parameters with the lowest amplitude (especially zero crossings in the frequency domain) obtained from frequency analysis. Relative changes in the investigated characteristic or state value can be derived from the measurement data without calibration. For an absolute determination, it is advisable to use a calibration on the material under investigation or a corresponding database from which the characteristic or state values can be read, for example, using machine learning methods.

In particular, the sensor's lift-off values are also derived from at least one

Component of the decomposition is derived in the form of a frequency and/or amplitude signal, so that an internal calibration of the lift-off is performed based on the measurement data, thereby improving the usable measurement data.

The measurement signal, and therefore the measurement data, is also improved if the electronics used to generate the magnetic field pulse are located close to the metallic object during the measurement, for example, on a

The sensor head is positioned against the pipe wall, allowing for the shortest possible cable lengths. At least parts of the control unit are then formed by this electronics.

Preferably, the sensor arrangement is part of an inspection device, which is in particular designed in the form of a pipeline pig.

-15- BE2024/5802

According to a further development of the invention, the total energy of one or both magnetic field pulses is 1 mJ - 150 mJ, in particular 1 mJ - 50 mJ, and more preferably 1 mJ - 20 mJ. These values have proven to be preferable for a good signal-to-noise ratio in numerous experiments.

Preferably, the first magnetic field pulse has at least one and in particular both of the following properties (maximum field strength or time duration): 1) maximum field strength of 50 kA/m - 450 kA/m, in particular of 80 kA/m - 300 kA/m 2) time duration of the first magnetic field pulse of 1 µs - 3000 µs, preferably of 100 µs - 1300 µs, further preferably of 200 µs - 800 µs.

Furthermore, when using two magnetic field pulses, the second pulse, designed as a magnetic field pulse, according to a further development of the invention, has at least one and in particular both of the following properties (maximum field strength or temporal duration): 1) maximum field strength of 50 kA/m - 450 kA/m, in particular of 80 kA/m - 300 kA/m, 2) temporal duration of the second magnetic field pulse of 1 µs - 5000 µs, in particular of 1 µs - 2000 µs, and further in particular of 400 µs to 1200 µs.

Particularly good measurement results are obtained especially from using the claimed quantity of the magnetic field change in combination with a length

- 16 - BE2024/5802 of the first and second magnetic field pulses from the two previously described areas. The magnetic field pulse is the pulse that arises during a voltage or current excitation. Any decay is not considered here; in particular, measurements are only taken during active excitation.

Preferably, both a measurement cycle and an initial evaluation, including the storage of the data obtained from the raw or measurement data by means of an FFT, are performed within a 10 ms cycle.

The results are further improved if, in addition to the magnitude of the change in the magnetic field and the excitation durations as well as the temporal ratio of the durations, the field strength of the two pulses is also in the preferred range.

In particular, the voltages used to generate the excitation, especially the voltages thus applied to the coil, are at least nearly constant for generating the first or the second magnetic field pulse, such that a change in voltage over 75% of the duration of the respective excitation pulse, in particular over 90% of the duration of the excitation pulse, is in particular <30%, preferably <10%, and more preferably <5% of the amplitude of the voltage, so that quasi-rectangular, or at least more or less constant,

Voltages are used for excitation.

Preferably, the short-term excitation of the magnetic field pulses is in the form of a rectangular signal.

“17 - BE2024/5802

An absolute voltage state can be determined, in particular, through calibration of the

The sensor arrangement is determined on the material of the object under investigation or on a material corresponding to the material of the object under investigation.

Further material properties can be determined in particular using machine learning methods, in which the determined components of the decomposition are compared with a previously created database containing comprehensive sensor data from material samples with defined material properties.

The task stated at the beginning is also accomplished by an arrangement for determining a material property of a magnetizable metallic object and/or a

The invention addresses the material state of a magnetizable metallic object, and in particular relates to an arrangement for determining material stress in steel components. The arrangement according to the invention comprises an inspection device, which is designed in particular as an inspection pig for a pipeline. Such an inspection device comprises at least one micromagnetic sensor arrangement, which has at least one coil, in particular wound around a magnetic core, for generating magnetic field pulses, and at least one receiver. Furthermore, the inspection device has a control unit for controlling the sensor arrangement and means for receiving and storing the sensor.

Store any measurement and analysis data. Analysis data corresponds, at least partially, to evaluated data. For example, this includes data that

"18 - BE2024/5802 can be obtained via the FPGA. These means for recording and storing data can already be part of the control unit. The arrangement is also provided with an evaluation device, which can be completely separate from the inspection device or which can be partially formed by parts of the inspection device, in particular by its sensor arrangement and/or its control unit. The arrangement is designed to perform the procedure described above or below."

The evaluation device is primarily equipped with electronic data processing equipment and appropriate hardware and software. Furthermore, it is understood that the inspection device, in particular, has a power supply, for example in the form of accumulators.

At least parts of the control unit, the sensor arrangement and the evaluation device can be integrated together, so that, for example, the control and recording of measurement data is realized at least partially via the same electronic unit as at least part of the evaluation and/or part of the measurement.

Preferably the coil is wound around a soft magnetic yoke, in particular wherein the yoke is a nanocrystalline yoke, in particular which has crystallites with grain dimensions in the range of 10 nm - 15 nm.

-19- BE2024/5802

In particular, the control unit has a generator for producing a square wave voltage, which, if necessary with the aid of an amplifier, is used to drive the coil. The time-varying signal detected in the receiver

Current is measured via a current meter in the control unit.

The arrangement is specifically designed to access remote and previously written databases using standard means for the purpose of data comparison.

Further advantages and details of the invention can be found in the following description of the figures. The schematic representation shows:

Fig. 1 shows an arrangement according to the invention,

Fig. 2 shows a partial view of the arrangement according to Fig. 1.

Fig. 3 shows an illustration of an excitation and measurement signal,

Fig. 4 shows a sequence of the method according to the invention,

Fig. 5 Signal-to-noise ratios for a number of different excitation pulses with a representation of the determined relationships,

_ 20 - BE2024/5802

Fig. 6 Measurement results for the measured strain compared to the actual strain, as well as measurement results for the measured tensile strength compared to the actual tensile strength.

Fig. 7 mechanical stresses in the elastic range with a signal-to-

Noise ratios of 21 dB, 15 dB and 9 dB.

Individual technical features of the embodiments described below can also be combined with embodiments described above, as well as the

Features of one of the independent claims and any further claims are combined to form the subject matter of the invention. Where appropriate, functionally equivalent elements are provided with identical reference numerals.

An arrangement according to the invention comprises an inspection device 2 in the form of an inspection pig and an evaluation unit 50, which, for the purpose of downloading measurement and/or analysis data from the inspection device 2 after its inspection run, can be connected to it via a data connection 52 and which is part of an evaluation device that is formed by the inspection device itself. The inspection device 2 has a plurality of sensor assemblies 4, which are arranged circumferentially around a central body 6 in sensor heads 5. Only two of several sensor heads 5, evenly distributed circumferentially, are shown.

- 21 - BE2024/5802

The central body 6 contains a power supply 8 for the sensor arrangements 4 and

Means are available for further control of the process and, in particular, for data storage. Specifically, a control unit 10 is available, in which a

The voltage generator for exciting the magnetic field pulses may be included. The voltage generator for generating the excitation pulses is shown here in the form of...

Rectangular voltages are already formed in the sensor head 5 containing the sensor arrangement 4.

Furthermore, the inspection device has a data storage device 12 in which the data, already partially processed in the sensor head 5, are stored. Each sensor head 5 also contains an FPGA, which is used to initially evaluate the measurement data, so that the data storage device 12 contains a significantly reduced amount of data compared to the recorded measurement data.

The data set is stored. After completion of the measurement run, this data is transferred to another electronic data processing system and used there for the final evaluation of the material properties, the material condition, or the progression of the respective values along the considered section of the object. The control unit 10 can also, or alternatively, provide means for evaluating the

measurement data or for further processing of the data that has already been processed, and thus also form at least part of the evaluation device.

The inspection device 2 is held centrally in a pipeline 18 (shown only with dashed lines) by means of the usually existing disks 14 and cups 16 and connected to

“22 - BE2024/5802 moves through the medium. Odometers 20 serve to determine the position of the inspection device 2 in the pipeline 18. The control unit 10 can evaluate the data of the odometer 20 and, for example, trigger the measurement by the sensor arrangement 4 depending on a position in the pipeline or pipe, and, for example, store the measurement and/or already at least partially evaluated data with corresponding position and/or time information in the data memory 12.”

Depending on the design of the invention, the data storage device can also be part of the control unit.

The sensor arrangements 4 have a soft magnetic yoke 22, which is wound with a coil 24. The coil 24 is supplied with a square wave voltage in the form of a single positive and a single immediately following negative amplitude via a voltage generator 26 in the sensor head 5 and triggered by the control unit 10. The resulting time-dependent current in the coil 24 is measured by an ammeter 28 of the sensor arrangement 4 (Fig. 2). The excitation pulse is thus generated by the coil 24. At the same time, the coil 24 also serves as a receiver for detecting the signal waveform and, accordingly, for receiving the measurement data.

For an alternative embodiment of the invention, the dash-dot-lined straight lines 35 indicate an alternative receiver in the form of a second coil, which is then in particular identical to the first coil.

-23- BE2024/5802

A time-dependent variation of the current is detected due to the magnetic circuit 30 coupled through the material of the pipeline wall 18. The excitation signal 32 of the voltage generator 26 is shown in arbitrary units in Fig. 3. Preferably, the voltage, which can also be a maximum of 60 V, is, for example, 10 V+ and 10 V-. The resulting measurement signal in the coil 24, or the

Measurement data result from a common signal waveform or from two signal waveforms 34, which may be separated in the area of the sign change, but are evaluated together.

For example, the coil has 240 turns with a wire diameter between 0.3 mm and 0.4 mm, with the pole shoes having dimensions of 25 mm x 25 mm, a total height of 66 mm, a total depth of 25 mm and a total width of 74 mm.

Specifically, after the inspection device is started (1), a measurement is initiated in step 3 (Fig. 4). In a first measurement step (15), a positive voltage pulse is applied to the coil (24) for a time T1 (substep 7). Simultaneously, in substep 9, the coil current is measured for the same duration. In a further measurement step (11), a second measurement (15) of the coil current is then performed for a second time T2, parallel to the application (13) of a negative voltage pulse. The measurement is then terminated by ending the excitation, and in step 17, the evaluation of the measurement data is started.

“24 - BE2024/5802

In the evaluation 19 of the measurement data, a Fourier transformation 21 of the entire current measurement for times T1 and T2 is first carried out, for example in a

FPGA in sensor head 5. The (considered relevant) Fourier components can then be stored and, if necessary, output in a single step 27.

Furthermore, based on the aforementioned two to four relevant Fourier components, stress and strain states can already be determined in step 25, which are also output and/or stored in step 31.

For further data evaluation, the Fourier components can be used in a component comparison 23 with data from a (material) database 36 to determine the material properties, such as the tensile strength or toughness of the material. From this, stress and strain states can be derived in step 25. This is done, for example, after the data from the inspection device 2 have been transferred after a measurement run to the evaluation unit 50, which is designed as an electronic data processing system, with a corresponding database 36 or in conjunction with a corresponding database 36.

The material properties determined in step 23 can also be used again in a

Step 29 will be output and/or saved.

The process ends in step 33.

_ 25 - BE2024/5802

In the figures in Fig. 5, the signal-to-noise ratio is plotted for a variety of measured values for pipeline steels. The preferred regions described above can now be identified from the added and dashed envelopes. The figures show the values characterizing the magnetic field pulse with their respective signal-to-noise ratios.

Noise ratio. In particular, the maximum H-field change and the length of the magnetic field or excitation pulses in the three figures on the right clearly illustrate the regions claimed according to the invention. The optimum of

The strength of the H-field is clearly visible.

Results according to the invention at signal-to-noise ratios >20 dB are shown for exemplary measurements in Fig. 6. In the left-hand figure of Fig. 6, the actual tensile strength of the material in MPa is plotted against the results of the reference measurement. In the right-hand figure of Fig. 6, the mechanical stress is plotted as a function of the measured stress, representing the material state of the material under investigation. The inventive principle is evident.

Methods can be used to determine material properties and conditions.

Measurement data acquired outside the preferred ranges exhibit a significantly worse signal-to-noise ratio (Fig. 7, right) than the excitations from the preferred ranges (Fig. 7, left). Intermediate values are plotted in the center of Fig. 7, which already provide comparatively meaningful results.

- 26 - BE2024/5802

The measured voltage values in MPa are plotted against the

Reference values.

Claims (18)

- 27 - BE2024/5802 Claims 1. A method for determining at least one material property and/or a material state of a magnetizable metallic object, in particular a method for determining a material stress in steel components, by means of a micromagnetic sensor arrangement (4) comprising at least one coil (24) wound around a magnetic core (22) for generating magnetic field pulses and at least one receiver, comprising the steps of: - generating at least one magnetic field pulse based on at least one excitation pulse of the sensor arrangement (4) for magnetizing the object, - recording measurement data in the form of at least one signal waveform resulting from the magnetization of the object in the receiver during the at least one excitation pulse, - determining the material property and/or the material state of the object, in particular a relative stress state of the object, only based on at least parts of the measurement data recorded during the at least one excitation pulse, characterized in that the magnetic field (H-field) resulting from the excitation pulse exhibits a change of 150 kA/m/ms - 550 kA/m/ms, preferably from 350 kA/m/ms - 550 kA/m/ms, experiences. _ 28 - BE2024/5802 2. Method according to claim 1, characterized in that two magnetic field pulses are generated to change the magnetic field by means of two excitation pulses of the sensor arrangement (4), wherein the two excitation pulses are generated by voltages with opposite polarity or currents of opposite direction and the measurement data are recorded only during the first and second excitation pulse. 3. Method according to claim 2, characterized in that the two excitation pulses each have a nearly constant amplitude. 4. Method according to one of the preceding claims 2 or 3, characterized in that the excitation comprises only a positive and a negative excitation pulse of a total duration of not more than 5 ms, in particular a total duration between 0.2 ms and 5 ms, more preferably between 0.3 ms and 2 ms and more preferably from 0.4 ms to 1.2 ms. 5. Method according to one of the preceding claims, characterized in that the voltages are generated by means of a voltage generator associated with the sensor arrangement (4) and a time-varying current signal is measured as the measurement signal. 6. Method according to one of the preceding claims, characterized in that the measurement data to be used result exclusively from measurement data recorded during the sequence of the excitation pulse(s). -29- BE2024/5802 7. Method according to one of the preceding claims, characterized in that the evaluation involves a decomposition of the measurement data into their frequency components, preferably by means of a wavelet transformation, a Fourier transformation or an eigenvalue analysis, in particular wherein information in a frequency range of 0.5 to 50 kHz of the transform, in particular a Fourier transform, is used. 8. Method according to one of the preceding claims, characterized in that the total energy of one or the two magnetic field pulses has a value of 1 mJ - 150 mJ, in particular of 1 mJ - 50 mJ, and further preferably of 1 mJ - 20 mJ, 9. A method according to one of the preceding claims, characterized in that the first pulse designed as a magnetic field pulse has at least one and in particular both of the following properties: 1) a maximum field strength of 50 kA/m - 450 kA/m, in particular of 80 kA/m - 300 kA/m; 2) a time duration of the first magnetic field pulse of 1 µs - 3000 µs, preferably of 100 µs - 1300 ps, further preferably of 200 µs - 800 µs; and that, when using two excitation pulses, the second magnetic field pulse has at least one and in particular both of the following properties: 1) a maximum field strength of 50 kA/m - 450 kA/m, in particular of - 30 - BE2024/5802 80 kA/m - 300 kA/m, 2) time length of the second magnetic field pulse from 1 µs — 5000 µs, in particular from 1 µs — 2000 µs, further especially from 400 µs to 1200 µs. 10. Method according to one of the preceding claims including claim 2, characterized in that the first excitation pulse is longer than the second excitation pulse, in particular wherein the two excitation pulses have a pulse length ratio (first:second) between 2.5:1 and 0.75:1, in particular from 2:1 inclusive up to and including 1:1. 11. Method according to one of the preceding claims including claim 2, characterized in that the voltages used to generate the excitation, in particular the voltages applied to the coil, are at least nearly constant for generating the first or the second magnetic field pulse such that a change in the respective voltage over 75% of the duration of the respective excitation pulse, in particular over 90% of the duration of the excitation pulse, is less than 30%, preferably less than 10% and further preferably less than 5% of the amplitude of the voltage. 12. Method according to claim 10, characterized in that the excitation of at least one and in particular both magnetic field pulses is carried out by a rectangular signal. - 31 - BE2024/5802 13. Method according to claim 12, characterized in that the excitation is carried out with an amplitude from a range of 5 V to 15 V, in particular from 11 V to 13 V. 14. Method according to one of the preceding claims, characterized in that the coil (24) is used as the receiver, wherein the current response which changes over time due to the excitation is detected. 15. Method according to any one of the preceding claims 1 to 13, characterized in that a second sensor coil is used as the receiver. 16. Method according to one of the preceding claims, characterized in that, in order to determine material properties using machine learning methods, a comparison is made with a previously created database (36) comprising sensor data of material samples with defined material properties. 17. Method according to one of the preceding claims, characterized in that, in order to determine the absolute stress state, a calibration of the sensor arrangement (4) is carried out on the material of the object to be examined or on a material corresponding to the material of the object to be examined. 18. Arrangement for determining a material characteristic and/or a material condition of a magnetizable metallic object, in particular arrangement -32- BE2024/5802 for determining material stress in steel components, comprising an inspection device (2), which is designed in particular as an inspection pig for a pipeline and/or pipe (18), with at least one micromagnetic sensor arrangement (4) which has at least one, in particular around a magnetic core (22) a coiled coil (24) for generating magnetic field pulses and at least one receiver, wherein the arrangement includes a control unit (10) for controlling the sensor arrangement (4) as well as means for recording and storing any measurement and analysis data and an evaluation device that is completely separate from the inspection device (2) or partially separate from the inspection device- tung (2), in particular its control unit (10) or parts of the sensor arrangement (4), is characterized in that the arrangement is designed to carry out a method according to one of the previous claims.