State Analysis of Electrical Operating Devices

Abstract

According to a method for analyzing the state of an operating device for electricity, a detection voltage is applied to the operating device. Subsequently, a measurement signal is acquired at a connection point (AS) of the operating device. A transmission parameter is determined from the measurement signal, and the transmission parameter characterizes the signal transmission from the location of partial discharge in the operating device to the connection point (AS). Based on the transmission parameter, at least one characteristic parameter of the partial discharge is determined. The determination of the transmission parameter includes processing the measurement signal according to a method for linear predictive coding.

Description

Technical Field

[0001] The present invention relates to a method for analyzing the state of an electrical operating device and a detection device for analyzing the state of an electrical operating device. Background Art

[0002] A locally restricted discharge is called partial discharge TE, which only partially bridges the insulation between conductors and can but does not have to occur in the vicinity of the conductors. Partial discharge may be attributed to a fault source, such as a defect or inhomogeneity in the electrical insulation. Partial discharge causes aging of the electrical insulation and thus adversely affects its dielectric quality, especially the breakdown strength. In order to ensure the dielectric quality of electrical operating devices, partial discharge measurements are performed to check for the presence of partial discharge and, if necessary, to determine its specific characteristics.

[0003] Naturally, the signal generated by the TE in response to the detection voltage must pass through a certain transmission path between the location of the TE and the measurement location. Depending on the distribution of the inductive, capacitive, and resistive elements of the operating device (cable, transformer, etc.), the original TE pulse attenuates and deforms. Usually, only a small part of the discharge energy released at the defect can be measured at the measurement point. Therefore, this is called the so-called apparent charge (Scheinbare Ladung). Thus, only a small sensitivity is achieved with known measurement methods. The greater the distance between the location of the TE and the measurement location, the greater the error in determining the apparent charge. In the quality assurance and monitoring of operating devices, a TE value that is not allowed to be exceeded is usually defined. This definition here relates to the apparent charge measurable at the measurement point, which in turn is only a small part of the actual discharge energy. The quality assurance of operating devices can thus only be achieved to a limited extent.

[0004] In order to locate possible faults in operating devices, the time domain reflectometry method can be used. Here, the time offset between the input detection pulse and the measured reflection corresponds to twice the distance to the fault location. Technically, this is significantly difficult due to signal attenuation and signal dispersion, because the location according to this method is only possible with a sufficient signal amplitude. In addition, this method presupposes a reflection at the fault location, which is not always the case due to the wave impedance at the fault location. In the grid operation of specific operating devices, such as gas-insulated switchgear, GIS, or gas-insulated lines, GIL, it is often not possible to perform fault location in the above manner, because a large number of discharges cover possible reflections or the amplitude of the discharge to be located is too small to be measured after reflection. Summary of the Invention

[0005] Accordingly, the object of the present invention is to provide an improved solution for the TE measurement of an electrically operated device, which enables a higher accuracy of TE analysis for the electrically operated device.

[0006] This object is solved by a method for the condition analysis of an electrically operated device according to the invention and a detection device for the condition analysis of an electrically operated device according to the invention.

[0007] In a first aspect, the present invention relates to a method for the condition analysis of an electrically operated device, wherein the method comprises:

[0008] - applying a detection voltage to the operating device;

[0009] - collecting a measurement signal at a connection point of the operating device;

[0010] - determining a transmission parameter from the measurement signal, the transmission parameter characterizing the signal transmission from the location of partial discharge in the operating device to the connection point;

[0011] - determining at least one characteristic parameter of the partial discharge according to the transmission parameter,

[0012] wherein the determination of the transmission parameter includes processing the measurement signal according to a method for linear predictive coding.

[0013] In a second aspect, the present invention relates to a detection device for the condition analysis of an electrically operated device, the detection device comprising an evaluation unit, the evaluation unit being configured to:

[0014] - collect a measurement signal at a connection point of the operating device;

[0015] - determine a transmission parameter from the measurement signal, the transmission parameter characterizing the signal transmission from the location of partial discharge in the operating device to the connection point; and

[0016] - determine at least one characteristic parameter of the partial discharge according to the transmission parameter,

[0017] wherein the determination of the transmission parameter includes processing the measurement signal according to a method for linear predictive coding.

[0018] The improved solution is based on the idea of identifying the influence of the transmission path on the signal from the measurement signal, especially the measurement signal collected in response to the applied detection voltage. This influence is described by a transmission parameter, which characterizes a partial discharge or a fault source or multiple fault sources. Therefore, the characteristic parameters of the partial discharge can be inferred from the transmission parameter.

[0019] According to an improved solution, a method for state analysis of electrical operating devices is proposed. According to this method, a detection voltage is applied to the operating device, for example injected into the operating device, in particular applied to the injection location of the operating device. Then, a measurement signal is collected at the connection part of the operating device. Transmission parameters are obtained based on the measurement signal, and the transmission parameters characterize the signal transmission from the location of partial discharge in the operating device to the connection part. Based on the transmission parameters, at least one characteristic parameter of the partial discharge is determined.

[0020] The measurement signal can be understood as the response to the detection voltage caused by partial discharge. As a response to the detection voltage, for example, multiple partial discharges at different locations of the operating device can also jointly cause the measurement signal. This is included in the expression "location of partial discharge" and is understood as "location of partial discharge" if necessary or accordingly understood.

[0021] According to at least one embodiment of the method, obtaining the transmission parameters includes discretizing and / or digitizing the measurement signal, thereby generating a discrete measurement signal.

[0022] According to at least one embodiment of the method, the transmission parameter is a parameter of a transfer function or a transmission function, and the parameter characterizes the signal transmission from the location of partial discharge to the connection part. The measurement signal or the discrete measurement signal can be understood here as the mapping of the input signal at the location of partial discharge, or as the mapping of the input signal at the apparent location of partial discharge in the case of multiple partial discharges, where the transfer function is used as the mapping operator. The transfer function can always also represent an approximation of the exact transfer function here and in the following.

[0023] The detection voltage is in particular an alternating voltage, for example having a frequency in the range of 0 to 500 Hz. However, in various embodiments, a DC voltage can alternatively be used as the detection voltage. The detection voltage is in particular provided by a high-voltage source, which is connected to the injection location directly or indirectly, in particular through at least one blocking impedance and / or through an input filter. The blocking impedance and / or the input filter can be used to decouple the high-voltage source from the rest of the test equipment.

[0024] According to at least one embodiment, the transmission parameters are obtained by means of a method for linear predictive coding (LPC) or a part of the method for LPC. The LPC method is currently used in audio signal and speech processing. Here, the signal value at a determined time point is determined by a linear combination of the signal values at previous discrete time points and is thus predicted to a certain extent.

[0025] According to at least one embodiment, the determination of transmission parameters, in particular for LPC, includes determining the filter coefficients of a virtual filter, where the filter approximately simulates signal transmission.

[0026] The filter is called virtual here because it does not have a physical counterpart in the sense of a dedicated filter component, but is effectively constituted by operating devices, measurement circuits, and optionally one or more fault sources.

[0027] For example, the filter according to LPC will be described by a recursive formula, such as the following

[0028]

[0029] Here, k is a discrete time variable, that is, a natural number greater than zero, y(k) is the value of the discrete measurement signal at the discrete time point k, and N is the approximate order. Here, if (k - i) is less than or equal to zero, then y(k - i) is defined as 0. Here, a k is the so-called linear prediction factor of order N, and e(k) is the prediction error.

[0030] According to at least one embodiment, the prediction factor a k represents the transmission parameter.

[0031] Since the values of the discrete measurement signals are known, the transmission parameters can be determined by minimizing the total squared error. The total squared error is given by qE according to the following equation

[0032]

[0033] For this purpose, the total squared error of a i can be differentiated, the corresponding result can be set equal to zero, and the system of equations obtained from N linear equations can be solved.

[0034] According to at least one embodiment, the operating device is uniformly constructed, in particular, the resistance elements, inductance elements, and / or capacitance elements of the operating device are uniformly distributed. A uniformly constructed operating device can be, for example, a cable, in particular a shielded cable, such as a shielded high-voltage cable, GIS, or GIL.

[0035] According to at least one embodiment, the operating device is non-uniformly constructed, in particular, the resistance elements, inductance elements, and / or capacitance elements of the operating device are non-uniformly distributed. A "non-uniformly constructed operating device" can be, for example, a power transformer, a tap-changer of a power transformer, or other components. A "non-uniformly constructed operating device" can also be a power switch, a circuit breaker, a disconnector for use in GIS or air-insulated switchgear, a measurement transducer, an overvoltage arrester.

[0036] According to at least one embodiment, the at least one characteristic parameter includes one or more fault sources that cause partial discharge. This can be advantageous not only for uniformly constructed operating devices but also for non-uniformly constructed operating devices.

[0037] In particular, such an embodiment can be advantageous in DC applications or in applications for continuously monitoring or supervising operating devices.

[0038] According to at least one embodiment, the at least one characteristic parameter of the one or more fault sources includes the number of fault sources or the relative contribution of one of the fault sources to the partial discharge, in particular the relative contribution to the total charge value of the partial discharge (Teilungsladung).

[0039] According to at least one embodiment, at least one characteristic parameter of the fault source is used to identify a fault pattern.

[0040] With known methods for TE measurement, such identification is not possible, so the improved concept allows for the evaluation of partial discharge. Depending on this evaluation, for example, it can be decided whether operations such as maintenance, repair, shutdown, or replacement of the operating device or a part of the operating device are required. Unnecessary operations can be avoided in particular.

[0041] In particular, the identification can be carried out by the distribution of transmission parameters. The fault pattern can be regarded to a certain extent as the "fingerprint" of the fault source.

[0042] According to at least one embodiment, the fault pattern is associated with known characteristic fault patterns.

[0043] For example, the known characteristic fault patterns can be extracted from a fault catalog or generated by machine learning.

[0044] According to at least one embodiment, the method further includes generating a response signal based on a theoretical input signal and transmission parameters, the theoretical input signal being in particular time-discrete.

[0045] When there is a theoretical input signal at the location of the partial discharge, the response signal corresponds to the theoretical measurement signal here.

[0046] According to at least one embodiment, the theoretical input signal is a pulse signal approximating a partial discharge pulse. For example, the theoretical input signal can be non-zero only at discrete time points. This corresponds to the usually very short rise time of the partial discharge pulse.

[0047] According to at least one embodiment, the generation of the response signal includes determining a transfer function based on the transmission parameters and applying the transfer function to the theoretical input signal.

[0048] Here, "apply" is understood in the sense of applying a mathematical operator. For example, a transfer function can be applied to a theoretical input signal in the z-space. That is, the z-transform of the theoretical input signal can be generated and multiplied by the transfer function in the corresponding z-space representation. Optionally, the result can be transformed back to the discrete time period by an inverse z-transform to obtain a response signal in the discrete time period. Alternatively, the calculation can be performed in the discrete time period or in the frequency space.

[0049] According to at least one embodiment, at least one characteristic parameter of the partial discharge includes the value of the apparent charge of the partial discharge. In particular, the apparent charge is the apparent charge at the location of the partial discharge.

[0050] According to at least one embodiment, the value of the apparent charge of the partial discharge is determined based on the response signal.

[0051] According to at least one embodiment, the determination of the value of the apparent charge includes, in particular, the integration of the response signal in the discrete time period or a signal related to the response signal. The signal related to the response signal can be generated, for example, by filtering the response signal. Here, the filtering can correspond to a band-pass filter, in particular having a frequency band according to the international standard IEC 60270:2000, the content of which is incorporated herein by reference. The frequency band can be, for example, in the range of 100 - 900 kHz or, for example, corresponding to 100 - 400 kHz.

[0052] Since the response signal is not significantly affected by noise or other interference, a reliable value of the apparent charge at the location of the TE can be determined despite signal attenuation. According to an improved scheme, the apparent charge can thus be obtained with increased accuracy and independently of the distance from the connection point to the location of the partial discharge.

[0053] The charge value thus determined can be set with respect to the theoretical input signal. Thereby, it can be determined which component of the input signal can also be measured as the output signal. This enables the determination of the energy released at the discharge location independently of the distance from the measurement point to the fault location. For the condition assessment of electrical operating devices, this has significant advantages because it is thereby possible to judge the danger caused by defects.

[0054] According to an improved scheme, the value of the apparent charge at the location of the TE can be determined for operating devices with uniform and non-uniform structures.

[0055] According to at least one embodiment, at least one characteristic parameter of the partial discharge includes the fault location of the partial discharge. The fault location of the partial discharge is obtained here based on the response signal.

[0056] According to at least one embodiment, determining the fault location of a partial discharge includes fitting a reference function to the response signal, wherein the fitting is performed by adapting at least one parameter of the reference function.

[0057] According to at least one embodiment, a numerical optimization algorithm (e.g., the Levenberg - Marquardt algorithm) is used to perform the fitting.

[0058] According to at least one embodiment, the reference function is the solution of a differential equation for describing an electrical oscillation circuit. This is because the partial discharge excites an electrical oscillation circuit between the fault location of the partial discharge and the connection part.

[0059] According to at least one embodiment, the determination of the fault location is performed based on the capacitance and / or inductance of the operating device, wherein the capacitance and / or inductance are in particular corresponding approximate values.

[0060] According to at least one embodiment, the capacitance and / or inductance are obtained by fitting the reference function to the response signal. In particular, the capacitance and / or inductance are parameters of the reference function to be fitted.

[0061] According to at least one embodiment, the location of the partial discharge is determined from at least one adapted parameter of the reference function and the geometry of the operating device.

[0062] The location of the partial discharge can be determined according to an improved scheme especially for uniformly constructed operating devices. Since the response signal is not significantly affected by noise or other interference, very precise localization of the fault source can be achieved despite signal attenuation, regardless of the distance from the connection part to the fault location. The test enables an accuracy of at least 5% with respect to the length of the operating device to be expected.

[0063] For non - uniformly constructed operating devices, at least the parameters of the reference function, especially the capacitance and / or inductance, can be obtained, whereby a qualitative evaluation can be performed even without detailed inclusion of the geometry of the operating device. For example, the relative positions of different fault sources with respect to each other and to the connection part can be obtained.

[0064] According to this improved scheme, a detection device for the condition analysis of an electrical operating device is also provided. The detection device has an evaluation unit which is arranged to acquire a measurement signal at the connection part of the operating device. The evaluation unit is furthermore arranged to determine a transmission parameter from the measurement signal, the transmission parameter characterizing the signal transmission from the location of the partial discharge in the operating device to the connection part. In addition, the evaluation unit is arranged to determine at least one characteristic parameter of the partial discharge based on the transmission parameter.

[0065] According to at least one embodiment of the detection device, the detection device also includes a high-voltage source for applying a detection voltage to the operating device.

[0066] According to at least one embodiment, the detection device includes a coupling unit that can be coupled to the connection part and to the evaluation unit, in particular can be directly or indirectly electrically connected, and is arranged to provide a measurement signal for the evaluation unit.

[0067] According to at least one embodiment, the coupling unit includes a measurement impedance, in particular an inductive element having a settable inductance.

[0068] According to at least one embodiment, the coupling unit includes a filter element or a filter network connected downstream of the inductive element. Here, the filter element has, for example, a high-pass characteristic or a band-pass characteristic.

[0069] The possible limit frequencies for high-pass filtering can be, for example, 30 kHz or 100 kHz. For a band-pass, for example, 100 kHz as the lower limit frequency and / or 400 kHz or 500 kHz as the upper limit frequency are also possible. Depending on the specific requirements, other limit frequencies are possible or required. For monitoring applications, for example, a band-pass characteristic with an upper limit frequency from a few MHz to a few 10 MHz may be required.

[0070] According to at least one embodiment, the coupling unit is arranged to convert a signal, in particular a current signal, at its input into a measurement signal, in particular a voltage signal, at its signal output.

[0071] According to at least one embodiment, the detection device includes a coupling capacitor that can be connected between the coupling unit and the connection part, in particular can be directly or indirectly electrically connected to the connection part, and can be directly or indirectly electrically connected to the coupling unit. The coupling capacitor is, for example, a high-voltage capacitor. The coupling capacitor can, for example, be responsible for the recharge of the fault source or the maintenance of the electric field at the defect.

[0072] Other design ways and implementation ways of the detection device are directly derived from the different design ways of the method according to the improved solution, and vice versa. In particular, the single or multiple components described for the detection device and / or the devices for performing the method can be implemented accordingly. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The present invention will be described in detail below by way of exemplary embodiments with reference to the drawings. Components that are the same, functionally the same, or have the same effect may be provided with the same reference numerals. The same components or components with the same function may be explained only with reference to the drawings in which they first appear. The explanation is not necessarily repeated in the subsequent drawings.

[0074] Here, shown

[0075] Figure 1 A schematic diagram showing an example implementation of a detection device and an example operating device according to an improved solution;

[0076] Figure 2a , 2b a diagram showing an example transfer function of an example implementation of a method according to an improved approach;

[0077] Figure 3a , Figure 3b , Figure 3c a diagram showing a measurement signal, a theoretical input signal and a response signal according to another example implementation of the method according to the improved approach; and

[0078] Figure 4 A diagram showing a reference function and a response signal of another example implementation of the method according to the improved approach is shown. DETAILED DESCRIPTION

[0079] Figure 1 A schematic diagram of an exemplary embodiment of a detection device and an exemplary operating device according to the improved solution is shown. The operating device is shown as a shielded cable, which has a shield SC and a core SE. However, the operating device can be any electrical operating device that is uniformly or non-uniformly constructed and has components that are electrically insulated from each other.

[0080] The core SE and the screen SC are replaced by electrically insulated parts of the operating components. In a GIS or GIL, the operating components are, for example, one of the insulated conductors and the pressure vessel of the GIS / GIL. For a transformer, the operating components are, for example, the transformer winding and the housing of the transformer. In principle, coupling out is always possible as long as there is a corresponding capacitance in order to be able to recharge at least a part of the TE and thus collect the at least a part of the TE.

[0081] The detection device can have a high-voltage source HV, for example a high-voltage generator, which can be connected to a connection point AS of the operating device, for example a core line SE. In addition, the detection device can have a coupling capacitor KK and a coupling unit KE connected in series with each other. The coupling capacitor KK can be connected to the connection point AS, for example. Alternatively, the high-voltage source HV and the coupling capacitor KK can be connected to different positions of the operating device, in particular the high-voltage source HV can be connected to an injection position of the operating device, in particular can be connected to the core line SE, and the coupling capacitor KK can be connected to the connection point AS.

[0082] Alternatively, the high-voltage source can also be a component which is required for energy generation or distribution anyway, for example a generator or a grid transformer.

[0083] According to this improved solution, the detection device has an evaluation unit AE, which is electrically connected to the coupling unit KE, for example.

[0084] During detection, the detection voltage provided by the high-voltage source HV can be injected into the operating device at the connection point AS. Subsequently, signals generated, for example, by partial discharges in the insulation part of the operating device can be collected via the coupling capacitor KK and the coupling unit KE and output as measurement signals to the evaluation unit AE.

[0085] The evaluation unit AE can, for example, digitize the measurement signal and thereby generate a discrete measurement signal. Figure 3a An exemplary discrete measurement signal is shown in.

[0086] The evaluation unit AE can then determine transmission parameters based on the measurement signal, in particular the discrete measurement signal, which characterize the signal transmission from the location of the partial discharge in the operating device to the connection point AS, and determine one or more characteristic parameters of the partial discharge based on the transmission parameters.

[0087] The transmission parameter can here be the prediction factor a of a virtual filter defined by the recursive formula (1) i . Thus, the evaluation unit AE can determine the transmission parameter by minimizing the total squared error qE according to equation (2).

[0088] The transmission parameter can be understood as a parameter of a transfer function that correlates the theoretical input signal of the partial discharge with the discrete measurement signal. In particular, the transmission parameter can correspond to the poles of the transfer function in the complex z-plane.

[0089] Figure 2a The results of an exemplary determination of the transmission parameter are shown. Here, each cross represents a pole of the associated transfer function.

[0090] Figure 2b Shows Figure 2a the magnitude (solid line) and phase angle (dashed line) of the transfer function in the frequency domain in. For this purpose, the transfer function can be transformed from the z-plane to the frequency domain by inverse z-transform and Fourier transform.

[0091] Based on the transmission parameter and its distribution, in particular based on the arrangement of the poles of the transfer function, it is already possible to directly characterize one fault source or multiple fault sources that cause one or more partial discharges. For example, a fingerprint of the fault source can be generated to a certain extent based on the transmission parameter and compared, for example, with known characteristic fault images. In particular, the number of fault sources and their relative contribution to the total partial discharge can be determined in this way.

[0092] Figure 3b Shows a theoretically, especially digitally generated input signal (solid line). For example, since partial discharge pulses have a very fast rise time, e.g. in the range of a few nanoseconds or one nanosecond, the theoretical input signal can correspond to a digital pulse of minimum width. The area of the theoretical input signal is e.g. equal to 1.

[0093] Figure 3b Also shows a response signal (dashed line), which corresponds to the mapping of the theoretical input signal by means of a transfer function. Figure 3c Shows Figure 3b a reduced local part.

[0094] The response signal is similar to Figure 3a the discrete measurement signal in, but has little noise. The evaluation unit AE can e.g. filter and integrate the response signal in a determined range, e.g. given in IEC 60270:2000, e.g. in the range of 100 - 400 kHz. The result of the integration is a measure of the apparent charge of the partial discharge at the location of the TE. In the Figure 3b and Figure 3c example, the digital input signal has an area of 1, and the integration of the filtered response signal gives e.g. 0.55. If this value is compared with a reference value, the value of the apparent charge can be determined, and the reference value can e.g. be determined by feeding a defined charge with a TE calibrator.

[0095] By partial discharge, an oscillating circuit can be excited from the fault location to the connection part AS. The oscillation can generally be described e.g. by the differential equation

[0096]

[0097] Here, U is the voltage in the oscillating circuit, R is the resistance, L is the inductance and C is the capacitance. The differential equation for the corresponding current has the same form and can be used similarly.

[0098] The solution of equation (3) can then be used as a reference function and has a known form in the case of oscillation

[0099] U(t) = U0exp(-tR / 2L)·sin(t / LC + φ) if (R / 2L) 2 <1 / LC (4)

[0100] Here, U0 and φ are parameters depending on the initial conditions of the oscillation. For the creep case and the non - periodic limiting case, the corresponding solutions of equation (3) are known. From the adaptation or fitting of the reference function to the response signal, the capacitance C or the inductance L can be determined in particular.

[0101] Figure 4 An exemplary response signal (dashed line) and a reference function (solid line) are shown. For a clearer illustration, the two curves are shown shifted from each other in time.

[0102] In the case of uniformly constructed operating devices, such as cables, GIS or GIL, the distance between the fault location and the connection point AS can be determined by means of the geometry of the operating device, such as the length, cross-section of the cable, cross-section of the core, and, if necessary, by means of the insulating material used. Thereby, the fault location can be determined.

[0103] In non-uniformly constructed operating devices, a qualitative characterization of one or more fault sources can be achieved based on the inductance L and / or the capacitance C. For example, the relative distance between the fault location of different fault sources and the connection point AS can be determined or limited.

[0104] Using the method or the detection device according to the improved scheme, a more accurate TE analysis of electrical operating devices can be achieved. The charge released at the fault location can be determined in all types of electrical operating devices, especially in cables, which allows a significant improvement in the condition assessment. In the method according to the improved scheme, only a single excitation is required, such as a TE pulse at the fault location. Thereby, an oscillating circuit is constructed from the inductive, capacitive and resistive elements of the operating device and the measuring circuit. The current or voltage in this circuit is completely described. Therefore, the fault location can also be determined, especially in the case of cables, without signal reflection occurring at the fault location. Therefore, faults can be identified at a low partial discharge level, i.e., a small amount of released charge, and thus particularly early. In addition, the improved scheme allows different partial discharge sources to be separated based on the determined transmission parameters ("fingerprints").

[0105] The improved scheme makes use of the knowledge that an operating device can be regarded as a filter, for example as a low-pass filter. Therefore, methods for LPC can be used to determine the transmission parameters. Here, the accuracy of the transmission parameters depends in particular on the order of the approximation involved, and on the available data, such as the sampling rate or the signal-to-noise ratio of the discrete measurement signal.

[0106] The method according to the improved scheme can also be used for the condition assessment of electrical energy transmission components, for example in the sense of continuous or ongoing monitoring during normal operation of the components.

[0107] List of reference signs

[0108] HV High-voltage source

[0109] AE Evaluation unit

[0110] KK Coupling capacitor

[0111] KE coupling unit

[0112] AS connection part

[0113] SC shielding

[0114] SE core wire

Claims

1. A method for analyzing the state of an operating device for electricity, wherein, The method includes: - Applying a detection voltage to the operating device; - Collecting a measurement signal at a connection point (AS) of the operating device; - Determining a transmission parameter based on the measurement signal, the transmission parameter characterizing signal transmission from a location of partial discharge in the operating device to the connection point (AS); - Determining at least one characteristic parameter of the partial discharge based on the transmission parameter, wherein the determination of the transmission parameter includes processing the measurement signal according to a method for linear predictive coding, and the method for linear predictive coding includes determining filter coefficients of a virtual filter, and the virtual filter approximately simulates the signal transmission.

2. The method according to claim 1, wherein The transmission parameter is a prediction factor of a virtual filter defined by a recursive formula.

3. The method according to claim 1 or 2, wherein The determination of the transmission parameter includes discretization and / or digitization of the measurement signal.

4. The method according to claim 1 or 2, wherein, The at least one characteristic parameter of the partial discharge includes at least one characteristic parameter of one or more fault sources causing the partial discharge.

5. The method according to claim 4, wherein, The at least one characteristic parameter of the one or more fault sources includes: the number of the fault sources; or the relative contribution of one of the fault sources to the partial discharge.

6. The method according to claim 1 or 2, wherein The method further includes generating a response signal based on a theoretical input signal and the transmission parameter.

7. The method according to claim 6, wherein, The at least one characteristic parameter of the partial discharge includes the value of the apparent charge of the partial discharge.

8. The method according to claim 6, wherein The at least one characteristic parameter of the partial discharge includes the fault location of the partial discharge.

9. The method according to claim 8, wherein, The fault location is determined based on the capacitance and / or inductance of the operating device.

10. The method according to claim 1 or 2, wherein, The operating device includes: - A cable; - A gas-insulated switchgear GIS; - A gas-insulated line GIL; - A power transformer; or - A grading switch of a power transformer.

11. The method according to claim 10, wherein The cable is a shielded cable.

12. The method according to claim 10, wherein, The cable is a high-voltage cable.

13. A detection device for state analysis of an electrical operating device, the detection device including an evaluation unit (AE), the evaluation unit being configured to: - Collect a measurement signal at a connection point (AS) of the operating device; - Determine a transmission parameter based on the measurement signal, the transmission parameter characterizing signal transmission from a location of partial discharge in the operating device to the connection point (AS); and - Determine at least one characteristic parameter of the partial discharge based on the transmission parameter, Among them, wherein the determination of the transmission parameter includes processing the measurement signal according to a method for linear predictive coding, and the method for linear predictive coding includes determining filter coefficients of a virtual filter, and the virtual filter approximately simulates the signal transmission.

14. The detection device according to claim 13, wherein, The detection device further includes a high-voltage source (HV) for applying a detection voltage to the operating device.

15. The detection device according to claim 13 or 14, wherein, The detection device further includes a coupling unit (KE), the coupling unit being capable of coupling with the connection point (AS) and the evaluation unit (AE), and the coupling unit being configured to provide the measurement signal to the evaluation unit (AE).

16. The detection device according to claim 15, wherein, The detection device further includes a coupling capacitor (KK), and the coupling capacitor can be connected between the coupling unit (KE) and the connection part (AS).

17. The detection device according to claim 13 or 14, wherein, The transmission parameter is a prediction factor of a virtual filter defined by a recursive formula.

Images