WO2022097091A1

WO2022097091A1 - Detection method for detecting electrical partial discharges in an electrical device, and such detector device

Info

Publication number: WO2022097091A1
Application number: PCT/IB2021/060277
Authority: WO
Inventors: Francesco FIAMBERTI; Angelo Pariani; Maurizio MOSTARDA
Original assignee: E.D.C. Electrical Dynamic Company S.R.L.
Priority date: 2020-11-06
Filing date: 2021-11-05
Publication date: 2022-05-12

Abstract

A detector device for detecting electrical partial discharges in an electrical apparatus; the device (1) comprising two output terminals (10); a pulse generator (11), preferably high voltage, connected to the two output terminals (10); a detecting unit (13) connected to the two output terminals (10); and a processing unit (15) connected to the detecting unit (13); preferably the device (1) comprising a coupling unit (12) disposed between the output terminals (10) and the detecting unit (13); preferably the device (1) comprising an acquisition system (14) disposed between the detecting unit (13) and the processing unit (15).

Description

"DETECTION METHOD FOR DETECTING ELECTRICAL PARTIAL DISCHARGES IN AN ELECTRICAL DEVICE, AND SUCH DETECTOR DEVICE"

CROSS-REFERENCE TO RELATED APPLICATIONS

This patent application claims priority from Italian patent application no. 102020000026515 filed on November 6, 2020, the entire disclosure of which is incorporated herein by reference.

TECHNICAL FIELD

The present invention relates to a detection method for detecting partial discharges in an electrical device, for example in a rotating electrical machine, and a detector device for detecting partial discharges in an electrical device, for example in a rotating electrical machine.

BACKGROUND ART

A rotating electrical machine comprises a stator and a rotor. The stator comprises a plurality of stator windings. Each stator winding comprises a plurality of turns which are formed by a conductor wound a plurality of times around a portion of stator for forming the stator poles. Said conductor of the stator winding is insulated via an insulator layer. During the use of the rotating electrical machine partial discharges can occur, i.e. electrical discharges that involve only a part of the insulator disposed between different turns of the same conductor or between different conductors and consequently between different windings. The repeated appearance over time of partial discharges can cause damage to the insulating material.

In the field of electrical devices, in particular high voltage, for example electrical machines or disconnecting switches, it is important to measure the partial discharges, in particular the amplitude, the duration over time and the frequency of said discharge currents. For example, said measurements are carried out on a just produced electrical machine and prior to the setting in use so as to check possible manufacturing defects or said measurements are carried out on electrical machines already in use, preferably at regular intervals, so as to check the good functioning thereof and detect possible deteriorations or beginning of deterioration. The electrical machines can be rotating, i.e. electric motors and/or generators, or static, i.e. electrical transformers.

When a partial discharge is triggered, transient high frequency current pulses will appear and persist for an interval of time comprised between nanoseconds and a microsecond, to then repeatedly disappear and reappear. The currents of the partial discharges are difficult to measure given their exiguity and their brief duration. The event can be detected as a very small variation of the current absorbed by the sample being observed.

In the known art, partial discharges are measured by applying voltage pulses to the device being tested, the voltage pulse applied to the device being tested has to have a brief risetime, for example between 100 ns and 500 ns, so that the test may reproduce situations as similar as possible to real ones during use, for example if the device being tested is a rotating electrical machine, the real conditions are those to which it is subjected when it is driven by an inverter .

The brief risetime is a strict requirement for the design of the detector device for detecting the electrical partial discharges since it can generate an involuntary remainder which superimposes the real signal of the detector device and since its frequency spectrum extends up to various MHz with amplitudes in the order of various kV.

Consequently, the detection and the measurement of the partial discharge is altered by the voltage pulse which is injected in the device being tested.

In order to overcome this problem, a solution in the known art is to use a multipole filter at the input of the detector of partial discharges, the drawback of said solution is that in order to completely remove the residual pulses deriving from the pulse injected in the device being tested a cut-off frequency in the order of a few hundreds of MHz should be chosen, but the detector device of the partial discharges is less sensitive to said frequencies and anyway the detection system is highly sensitive to external electromagnetic noise not correlated with the activity of the partial discharges, on the other hand if the cut-off frequency is chosen in the interval in which the detection of the partial discharges is more effective, i.e. under 100 MHz, the cancelling of the residual pulse is not complete.

Another solution of the known art in order to overcome the interferences created by the test pulse is to interrupt the input of the detector of partial discharges so that the input signal cannot enter, for a given interval of switchoff time after injecting the voltage pulse in the device being tested. Said switch-off interval has to be long enough for ensuring the complete extinction of the residual pulse. The drawback of said method is that it makes the detector device not sensitive to the partial discharges that occur during the switch-off interval, in particular to the partial discharges that occur during the risetime of the pulse or immediately after the peak of the pulse.

For some types of electrical apparatuses such as, for example, certain stators having a very low inductance, this drawback is particularly problematic since the detector device does not manage to detect discharges that occur at a given voltage, in particular the detector device manages to determine only the partial discharges that occur at a higher voltage with respect to a lower voltage where instead they already actually take place.

DISCLOSURE OF INVENTION

An object of the present invention is to provide a detector device for detecting partial discharges capable of reducing at least one of the drawbacks of the known art.

According to the present invention, a detector device for detecting electrical partial discharges in an electrical apparatus is provided according to one of the claims 1 to 8.

Thanks to the present invention, a device for detecting the partial discharges is obtained which is more sensitive and accurate with respect to the devices of the known art and detects all the partial discharges that occur in an apparatus being tested without the drawbacks of the known art. In fact, the partial discharges are events that have a certain degree of randomness and, consequently, do not repeat exactly in the same form, amplitude, time and/or frequency. In this manner, by subtracting from each other two test signals obtained by injecting two high voltage pulses it is possible to eliminate the common components and detect the partial discharges in a more sensitive and performing manner with respect to the known art and eliminate all the drawbacks of the known art. In other words, thanks to the present device it is possible to detect partial discharges that cannot be detected with the devices and the methods of the known art .

Another object of the present invention is to provide a detection method for detecting partial discharges capable of reducing at least one of the drawbacks of the known art.

According to the present invention, a detection method for detecting electrical partial discharges in an electrical apparatus is provided according to one of the claims 9 to 15.

BRIEF DESCRIPTION OF THE DRAWINGS

Further characteristics and advantages of the present invention will be apparent from the following description of a non-limiting example embodiment thereof, with reference to the accompanying drawings, wherein:

- Figure 1 is a block scheme of a detector device for detecting partial discharges in an electrical apparatus manufactured according to the present invention;

- Figure 2 is a scheme of a detail of Figure 1; and

- Figure 3 is a flow diagram relative to a detection method for detecting partial discharges embodied according to the present invention;

- Figure 4 illustrates the course of two test signals detected by the detector device of the present invention; and

- Figure 5 illustrates the course of a detection signal defined by the detector device of the present invention.

BEST MODE FOR CARRYING OUT THE INVENTION

With reference to Figure 1, reference numeral 1 indicates a detector device for detecting electrical partial discharges in an electrical apparatus 2.

The term electrical partial discharges means electrical discharges involving a part of an insulator of an electrical apparatus, preferably only a part of an insulator of an electrical apparatus.

In particular, electrical discharges involving a part of insulator of an electrical apparatus, preferably only a part of an insulator of an electrical apparatus, disposed between different turns of the same conductor of the electrical apparatus 2 being tested or between different conductors and consequently between different windings of the electrical apparatus 2 being tested. The detection of said electrical partial discharges is necessary because the repeated appearance over time of electrical partial discharges can cause damage to the insulating material.

The assembly comprising the detector device 1 and the electrical apparatus 2 defines a test system for testing electrical apparatuses.

The detector device 1 comprises two output terminals 10; a pulse generator 11, preferably high voltage; a coupling unit 12; a detecting unit 13; an acquisition system 14, preferably fast; and a processing unit 15.

With reference to Figure 1, the pulse generator 11 is connected to the two output terminals 10 for injecting pulses I at the output terminals 10 to which the electrical apparatus 2 to be tested is connected.

More specifically, high voltage pulse generator means a pulse generator having a voltage greater than: 100 Volt.

The electrical apparatus 2 can be any electrical apparatus to be tested.

In a preferred embodiment, the electrical apparatus 2 to be tested is an electrical machine, in particular a transformer or a rotating electrical machine, for example an electric motor or an electric generator.

In a preferred embodiment, the electrical apparatus 2 to be tested is a stator or a rotor of an electrical machine. Furthermore, in another embodiment, the electrical apparatus 2 to be tested is a coil or an inductor or a disconnecting switch.

Furthermore, in another embodiment, the electrical apparatus 2 to be tested is another type of electrical or electronic apparatus.

In particular, the electrical apparatus 2 comprises at least one conductor, preferably wound, with insulator around it. The electrical partial discharges to be measured are the electrical discharges that develop by means of the electrical insulator .

With reference to Figure 1, the detecting unit 13 is connected to the output terminals 10 via the coupling unit 12. The detecting unit 13, by means of the coupling unit 12, is connected to the output terminals 10 in parallel with the pulse generator 11 so as to detect a test signal ST at the output terminals 10 after the pulse generator 11 has injected one or more voltage pulses I, preferably high voltage.

In particular, the detecting unit 13 comprises a multipole filter, a fast amplifier and an envelope detector, in particular a diode envelope detector (not illustrated in the accompanying figures) . In particular, the envelope detector is configured to increase the duration over time of the signal peaks so as to facilitate the following processing of the signal, in particular so as to facilitate the following sampling.

In a preferred but non-limiting embodiment of the present invention, the detector device 1 comprises two calibration terminals 24 (Figure 2) . The calibration terminals 24 are configured to be connected to a calibration unit 25 which is preferably a unit external to the detector device 1.

The coupling unit 12 is connected to the output terminals 10 and is a high-pass filter. In particular, the coupling unit 12 is an RC high-pass filter comprising a resistor and a capacitor so as to protect the detecting unit 13 from the high voltage pulses.

In particular, the coupling unit 12 also protects the calibration terminals 24 and possible circuits connected thereto from the high voltage pulses, for example the calibration unit 25.

The coupling unit 12 is connected between the output terminals 10 and the detecting unit 13.

As mentioned above, the calibration unit 25, which is usually external to the detector device 1, is connected to the calibration terminals 24 and, consequently, between the coupling unit 12 and the detecting unit 13.

In particular, since the calibration unit 25 is connected to the calibration terminals 18, it is disposed upstream of the detecting unit 13.

In particular, the detector device 1 comprises a buffer unit 18, in a particular non-limiting example embodiment thereof a resistor. The coupling unit 12 and the detecting unit 13 are connected to each other via the buffer unit 18. The calibration terminals 24 are connected in parallel to the buffer unit 18 and between the coupling unit 12 and the detecting unit 13.

The calibration unit 25 will be connected to the calibration terminals 24 and, consequently, in parallel with the buffer unit 18 and between the coupling unit 12 and the detecting unit 13.

Thanks to the buffer unit 18, the calibration unit 25 can be connected as described above to the detecting unit 13 and in this manner, excepting an attenuation factor independent of the electrical apparatus 2 being tested, the values measured during the calibration are the same that would have been obtained injecting the calibration signal directly at the output terminals 10 with the advantage that the calibration unit 25 is protected from the high voltage pulses thanks to the coupling unit 12.

With reference to Figure 1, the processing unit 15 is connected to the detecting unit 13 for receiving values of the received signals.

In particular, the processing unit 15 is connected to the detecting unit 13 via the acquisition system 14.

In other words, the acquisition system 14 is disposed between the detecting unit 13 and the processing unit 15.

The processing unit 15 can be a personal computer or a dedicated microcontroller or a programmable logic device usually called with the acronym FPGA. An analysis software runs on the processing unit 15 for carrying out the detection of partial discharges on an electrical apparatus 2 to be tested as illustrated in the following.

Furthermore, the detector device 1 comprises a control device connected to the pulse generator 1 for controlling the pulse generator 1.

The control device is connected to the processing unit 15 for receiving data from the processing unit 15.

In a non-limiting embodiment of the present invention, the control device controls the pulse generator based on the data received from the processing unit 15.

In the following of the patent application, the term signal peak means the portion of the signal which is greater than a set signal threshold. In particular, peak means the portion of the signal comprised between the instant when the signal exceeds said signal threshold, in particular passing from values lower than said signal threshold to values higher than said signal threshold, and the following instant when the signal returns below said signal threshold, in particular passing from values above said signal threshold to values below said signal threshold.

Figure 5 illustrates peaks PK of a detection signal SR that will be introduced in the following. In said figure, the threshold value is indicated with a horizontal broken line .

Figure 4 illustrates two test signals ST, wherein each test signal ST is generated by a pulse, and the two pulses that generated the test signals ST are equal to each other. As apparent in Figure 4, although the pulses that generate the test signals are equal to each other, in this case the test signals ST are different from each other at least for one portion of the test signals ST. This difference indicates that a partial discharge occurred and the detector device uses this difference between the two test signals ST for detecting the partial discharge as illustrated in the following .

Figure 5 illustrates a detection signal SR which is generated based on the two test signals ST of Figure 4, in particular the detection signal SR of Figure 5 is generated by subtracting the two test signals ST of Figure 4 from each other .

With reference to Figures 3, 4 and 5, in use, the detector device 1 performs the following steps of a detection method for detecting the partial discharges:

100) Generating a first high voltage reference pulse I, injecting it at the output terminals 10 to which an electrical apparatus 2 to be tested is connected and detecting a first test signal ST (Figure 4) at the output terminals 10, in particular the first reference pulse is generated by the pulse generator 11 and the first test signal ST is detected by the detecting unit 13 which sends it to the processing unit 15 by means of the acquisition system 14;

110) injecting a high voltage pulse I at the output terminals 10 to which an electrical apparatus 2 to be tested is connected, in particular the detector device 1 injects the pulse I via the pulse generator 11;

120) detecting a test signal ST (Figure 4) at the output terminals 10 relative to the injected pulse, in particular the test signal ST is detected by the detecting unit 13, which sends it to the processing unit 15 by means of the acquisition system 14; and determining a detection signal SR (illustrated in Figure 5) by subtracting the test signal ST (Figure 4) relative to the injected pulse and the test signal ST (Figure 4) relative to the injected reference pulse, in particular the detection signal SR is defined by the processing unit 15 which subtracts the two test signals ST; in particular, the test signal ST is sampled by the acquisition system 14 and the values are sent to the processing unit 15; in particular, the detection signal SR is defined by the processing unit 15 which subtracts the two test signals ST in the entire sampling interval or only in a portion of the sampling interval which is of interest and which can be set via a selection of an operator;

130) selecting a portion of the detection signal SR to be analyzed, in an alternative embodiment the entire detection signal SR is considered and analyzed and is not limited to a single portion, consequently in said embodiment this step can be omitted and it is not an indispensable step of the detection method and of the detector device 1;

140) analyzing the detection signal SR so as to detect components of the detection signal SR, in particular so as to detect the peaks of said detection signal SR, and detect the amplitude values and/or value and/or duration over time for each given detected component of the detection signal SR, in particular detect the amplitude values and/or value and/or duration over time for each peak PK of the detection signal SR; preferably the detection signal SR is analyzed by the processing unit 15 via a software that detects the peaks PK of the detection signal SR and preferably extracts the amplitude value, value and duration over time for each peak PK of the signal, in an embodiment of the present invention this step is omitted;

150) filtering the detection signal SR so as to remove certain components of the signal, in particular certain peaks PK, whose amplitude is less than an amplitude threshold or whose area defined by the component of the signal, in particular by the peak PK, is less than an area threshold, in particular the processing unit 15, via the software, filters the detection signal SR so as to remove the peaks PK whose amplitude is less than an amplitude threshold or whose area defined by the peak PK is less than an area threshold, in an embodiment of the present invention this step is omitted;

160) detecting the maximum value of the preferably filtered detection signal SR, in particular the processing unit 15 detects the maximum value of the detection signal SR; preferably the processing unit 15 detects, via the software, the greater value of the amplitude of the greater peak PK of the filtered detection signal SR; preferably detecting the number of peaks PK in the filtered detection signal SR above the set threshold, in particular the processing unit 15 detects the number of peaks PK in the preferably filtered detection signal SR above the set threshold;

170) comparing the maximum value of the preferably filtered detection signal SR with a maximum threshold value and detecting a partial discharge if the maximum value is greater than the maximum threshold value, in particular the processing unit 15 compares the maximum value of the preferably filtered detection signal SR with a maximum threshold value and detects a partial discharge if the maximum value is greater than the maximum threshold value, preferably the maximum threshold value can be set by a user via a user interface connected to the processing unit 15; in an alternative embodiment, comparing the greater value of the amplitude of the greater peak PK with a peak threshold and detecting a partial discharge if said greater value of the amplitude is greater than the peak threshold, in particular the processing unit 15, via the software, compares the greater value of the amplitude of the greater peak PK with a peak threshold and detects a partial discharge if said greater value of the amplitude is greater than the peak threshold, preferably the peak threshold can be set by a user via a user interface connected to the processing unit 180) increasing a counter of injected pulses I and detecting if the number of injected pulses I is equal to a number of pulses I to be injected defined and/or selectable by a user via the user interface; furthermore, detecting if the number of partial discharges detected up to that moment is greater than a partial discharges threshold defined and/or selectable by a user via the user interface; repeating the steps from 110 to 180 if both the following conditions are met: the number of injected pulses I is less than the number of pulses to be injected and the number of partial discharges detected up to that moment is less than the partial discharges threshold; if even only one of the two above- mentioned conditions is not met, the sequence of steps is ended 190 and a message is sent, preferably via a user interface to an operator, indicating if there were partial discharges, in particular the processing unit 15 sends a message indicating if the partial discharges were or were not detected, and preferably the number of pulses that gave place to partial discharges and the voltage of the pulse that generated partial discharges; in particular, step 180 is performed by the processing unit 15.

In a preferred embodiment of the present invention, the number of total injected pulses I is greater than or equal to two. In particular, in order to detect a partial discharge, at least one reference pulse I and a following pulse I are necessary, or at least two pulses I are necessary, or at least one reference pulse I and a plurality of following pulses I are necessary.

In an embodiment of the present invention, it is counted how many high voltage pulses I, at the same voltage, of the number of pulses I produced electrical partial discharges, if said value is greater than a discharges threshold a repetitive electrical partial discharge at the relative voltage of the high voltage pulse I is detected, and preferably a message is sent preferably via a user interface to an operator, indicating the presence of repetitive electrical partial discharges and the voltage value that generated the repetitive electrical partial discharges. The electrical partial discharges threshold is preferably equal to half of the number of pulses I injected at a given high voltage. In particular, the step just described is performed by the processing unit 15.

Furthermore, the steps of the method from 100 to 180 can be repeated for different voltages, preferably greater, of the high voltage pulses I to be injected and the minimum voltage at which there are electrical partial discharges is detected.

More specifically, the detector device 1 can be used in two manners: in laboratory tests or in production line tests. In laboratory tests the procedure is as described above, performing cycles of steps from 100 to 180 for different voltages of the injected pulses I so as to determine the minimum voltage at which the partial discharges occur.

In production line tests, the steps from 100 to 180 are performed for a number of pulses I set at a given voltage, so as to check if the partial discharges are or are not present at the given voltage.

In an alternative embodiment of the present invention, step 120 is modified in the following manner with step 120' : 120' ) detecting a test signal ST at the output terminals 10 relative to the injected pulse I, in particular the test signal ST is detected by the detecting unit 13, which sends it to the processing unit 15 by means of the acquisition system 14; and determining a detection signal SR by subtracting the test signal ST relative to the injected pulse and the test signal ST relative to the pulse I injected in the previous cycle, in particular the detection signal SR is defined by the processing unit 15 which subtracts the two test signals ST; in particular, the test signal ST is sampled by the acquisition system 14 and the values are sent to the processing unit 15; in particular, the detection signal SR is defined by the processing unit 15 which subtracts the two test signals ST in the entire sampling interval or only in a portion of the sampling interval which is of interest and that can be set via a selection of an operator.

Practically, in this embodiment, the detection signal SR is defined by subtracting two test signals ST detected consecutively and relative to two consecutive pulses I from each other.

Also in said embodiment, the number of total injected pulses I is greater than or equal to two, in other words in order to detect an electrical partial discharge at least two pulses injected between reference pulse I and pulses I are necessary .

Variants can be applied to the above-described method or the latter can be simplified, for example the method can be carried out in a simplified form in the following manner: injecting a number of high voltage pulses I at the terminals of an electrical apparatus to be tested; detecting a test signal ST at the terminals 10 of the electrical apparatus 2 to be tested for each injected pulse I of the number of pulses; determining a detection signal SR based on two test signals ST, preferably by subtracting two test signals ST from each other; analyzing the detection signal SR so as to detect partial discharges for example by means of the detection and the processing of given components of the detection signal SR.

Furthermore, by way of non-limiting example of the present invention, the step of analyzing the detection signal SR can be modified so as to determine the amplitude and/or the value over time and/or the duration over time and/or the frequency of one or more components of the detection signal SR, in particular analyzing the detection signal SR so as to detect the peaks PK of the detection signal SR and determining the amplitude and/or the value over time and/or the duration over time and/or the frequency of said detected peaks PK.

Furthermore, by way of non-limiting example of the present invention, the step of filtering the test signal ST can be modified so as to remove certain components of the signal whose amplitude is less than a first threshold value and/or the duration over time or the frequency band of said component is less than a second threshold value; in particular, the step of filtering the test signal ST can be performed for removing the peaks PK of the signal whose amplitude is less than the first threshold value and/or the duration over time or the frequency band is less than the second threshold value.

Furthermore, by way of non-limiting example of the present invention, the method comprises the step of detecting the maximum value of the amplitude of the greater peak PK of each detection signal SR for each injected high voltage pulse and comparing the amplitude of the detected greater peak PK with a third threshold and detecting the presence of a partial discharge if said amplitude is greater than the third threshold. In particular, detecting the peak PK having maximum amplitude for each detection signal SR; comparing said maximum amplitude with the third threshold value and detecting a partial discharge if said amplitude is greater than the third threshold. In a preferred embodiment, the detection method comprises the step of detecting the maximum value of each test signal ST; comparing it with a third threshold value and detecting a partial discharge if said amplitude is greater than the third threshold.

Furthermore, by way of non-limiting example of the present invention, the method comprises the step of defining a fourth threshold, counting how many partial discharges have been detected following the number of injected pulses, detecting a repetitive partial discharge when the number of detected partial discharges relative to different pulses of the number of injected pulses is greater than the fourth threshold .

As mentioned above, a variant of the method provides for determining a detection signal SR by subtracting two test signals relative to two pulses of the number of injected pulses preferably consecutive between each other.

In another alternative embodiment, instead, the method comprises the step of sending a high voltage reference pulse, detecting a test signal ST at the terminals of the electrical apparatus to be tested for the high voltage reference pulse, and wherein the step of determining a detection signal SR is obtained by subtracting a test signal ST obtained with a high voltage pulse from the test signal ST obtained by injecting the high voltage reference pulse.

In an embodiment, the control device of the detector device 1 is connected to the pulse generator 11 and to the processing unit 15 for controlling said devices so as to perform the steps of the method from 100 to 180.

Furthermore, the control device is configured also to perform the above-illustrated modified steps of the method.

Finally, it is apparent that modifications and variants can be made to the device and to the method described herein without departing from the scope of the appended claims .

Claims

23 CLAIMS

1. A detector device for detecting electrical partial discharges in an electrical apparatus, in particular electrical partial discharges in an electrical insulator of the electrical apparatus to be tested; the device (1) comprising two output terminals (10) ; a pulse generator (11) , preferably high voltage, connected to the two output terminals (10) ; a detecting unit (13) connected to the two output terminals (10) ; and a processing unit (15) connected to the detecting unit (13) ; wherein the pulse generator (11) is configured to inject a number of high-voltage pulses at the output terminals (10) , wherein preferably the number of pulses is greater than or equal to two; the detecting unit (13) is configured to detect a test signal (ST) at the output terminals (10) for each pulse of the number of pulses injected by the pulse generator (11) and to send them to the processing unit (15) ; the processing unit (15) is configured to: determine a detection signal (SR) based on at least two of the test signals (ST) , in particular by subtracting two of the test signals (ST) from each other; analyze the detection signal (SR) ; and detect partial discharges based on the analyzed detection signal (SR) .

2. Detector device according to claim 1, comprising calibration terminals (24) upstream of the detecting unit (13) and configured to be connected to a calibration unit (25) , preferably external; the calibration terminals (24) being connected between the output terminals (10) and the detecting unit (13) ; preferably the device (1) comprising a coupling unit (12) disposed between the output terminals (10) and the detecting unit (13) ; preferably the coupling unit (12) and the detecting unit (13) being connected to each other via a buffer unit (18) , in particular a resistor, and the calibration terminals (24) being connected in parallel to the buffer unit (18) .

3. A detector device of claim 1, comprising a coupling unit (12) disposed between the device terminals (10) and the detecting unit (13) wherein the coupling unit (12) is a high- pass filter, in particular an RC high-pass filter comprising a resistor and a capacitor so as to protect the detecting unit (13) , and preferably the calibration terminals (24) , from high voltage pulses.

4. Detector device according to claim 1, the device (1) comprising a coupling unit (12) disposed between the output terminals (10) and the detecting unit (13) ; preferably the device (1) comprising an acquisition system (14) disposed between the detecting unit (13) and the processing unit (15) ; in particular the calibration terminals (24) being connected between the coupling unit (12) and the detecting unit (13) .

5. Detector device according to any one of the preceding claims, wherein the processing unit (15) is configured to analyze the detection signal (SR) to determine the amplitude and/or the value over time and/or the duration over time and/or the frequency of one or more components of the detection signal (SR) , in particular it is configured to analyze the detection signal (SR) to detect peaks (PK) of the detection signal (SR) and determine the amplitude and/or the value over time and/or the duration over time and/or the frequency of said detected peaks (PK) .

6. A detector device according to any one of the preceding claims, wherein the processing unit (15) is configured to filter the detection signal (SR) to remove certain components of the detection signal (SR) whose amplitude is less than a first threshold value and/or the duration over time or the frequency band of said component is less than a second threshold value; in particular, the processing unit (15) is configured to filter the detection signal (SR) to remove peaks (PK) of the detection signal (SR) whose amplitude is less than the first threshold value and/or duration over time or the frequency band is less than the second threshold value.

7. Detector device according to any one of the preceding claims, the processing unit (15) is configured to: detect the maximum value of each detection signal (SR) ; compare it with a third threshold value; and detect a partial discharge if said amplitude is greater than the third threshold value; in particular, the processing unit (15) is configured to detect the peak (PK) having a maximum amplitude of each detection signal (SR) , compare said maximum amplitude with the third threshold value, and detect a partial discharge if said amplitude is greater than the third threshold value.

8. Detector device according to any one of the preceding claims, the processing unit (15) is configured to define a fourth threshold value, count how many partial discharges have been detected following the number of injected pulses, detect a repetitive partial discharge when the number of detected partial discharges relative to the number of injected pulses is greater than the fourth threshold value.

9. A test system for detecting electrical partial discharges in an electrical insulator of an electrical apparatus to be tested, the system comprising the detector device according to any one of the preceding claims and an electrical apparatus, preferably the electrical apparatus being chosen in an assembly comprising the following electrical apparatuses: coil, inductor, disconnecting switch, electrical transformer, rotor, stator, rotating electrical machine, preferably an electric motor and/or an electric generator; wherein the electrical apparatus comprises at least one conductor having insulator around it.

10. A detection method for detecting electrical partial discharges in an electrical apparatus (2) , preferably the electrical apparatus being chosen in an assembly comprising the following electrical apparatuses: coil, inductor, disconnecting switch, electrical transformer, rotor, stator, rotating electrical machine, preferably an electric motor and/or an electric generator; in particular the electrical apparatus comprises at least one conductor having insulator around it; in particular the method detects the electrical partial discharges in an electrical insulator of the 27 electrical apparatus; the method comprising the steps of: injecting a number of high-voltage pulses at the terminals of an electrical apparatus (2) to be tested, preferably the number of pulses is greater than or equal to two; detecting a test signal (ST) at the terminals of the electrical apparatus (2) for each injected pulse of the number of pulses; determining a detection signal (SR) based on two of the detected test signals (ST) , in particular by subtracting two of the test signals (ST) from each other; analyzing the detection signal (SR) ; and detecting partial discharges based on the analyzed detection signal (SR) .

11. Method according to claim 10, comprising the step of analyzing the detection signal (SR) to determine the amplitude and/or the value over time and/or the duration over time and/or the frequency of one or more components of the detection signal (SR) , in particular analyzing the detection signal (SR) to detect peaks (PK) of the detection signal (SR) and determining the amplitude and/or the value over time and/or the duration over time and/or the frequency of said detected peaks (PK) .

12. Method according to claim 10 or 11, comprising the step of filtering the detection signal (SR) to remove certain components of the detection signal (SR) whose amplitude is less than a first threshold value and/or the duration over time or the frequency band of said component is less than a second threshold value; in particular filtering the detection signal (SR) to remove peaks (PK) of the detection signal (SR) whose amplitude is less than the first threshold value and/or the duration over time or the frequency band is less than the second threshold value.

13. Method according to any one of claims 10 to 12, comprising the step of detecting the maximum value of each detection signal (SR) ; comparing it with a third threshold value and detecting a partial discharge if said amplitude is greater than the third threshold; in particular detecting the peak (PK) having maximum amplitude of each detection signal (SR) ; comparing said maximum amplitude with the third threshold value and detecting a partial discharge if said amplitude is greater than the third threshold.

14. Method according to any one of claims 10 to 13, comprising the step of defining a fourth threshold, counting how many partial discharges were detected following the number of pulses injected; and detecting a repetitive partial discharge when the number of partial discharges detected following the number of pulses in ected is greater than the fourth threshold value.

15. Detection method according to any one of claims 10 to 14; wherein the step of determining a detection signal (SR) by subtracting two test signals (ST) from each other is performed by subtracting two test signals (ST) relative to two pulses of the number of pulses preferably injected consecutively from each other.

16. Method according to any one of claims 10 to 15, further comprising the step of sending a high voltage reference pulse, preferably prior to the steps described in claim 10; detecting a test signal (ST) on the terminals of the electrical apparatus to be tested for the high voltage reference pulse, preferably prior to the steps described in claim 10; wherein the step of determining a detection signal

(SR) is obtained by subtracting a test signal (ST) obtained with a high voltage pulse with the test signal (ST) obtained by injecting the high voltage reference pulse.

17. A computer program configured to run in a processing unit and implement the method steps of any one of claims 10 to 16.

18. A memory device on which the computer program of claim

17 is stored.