EP4639596A1

EP4639596A1 - Method for monitoring an operating condition of a component of a high voltage device

Info

Publication number: EP4639596A1
Application number: EP23837707.1A
Authority: EP
Inventors: Felipe MACEDO; Thomas Christen; Angelos Garyfallos; Lennart MERKERT; Kai YUAN; Caio DOREA; Bernhard Deck
Original assignee: Hitachi Energy Ltd
Current assignee: Hitachi Energy Ltd
Priority date: 2022-12-22
Filing date: 2023-12-22
Publication date: 2025-10-29
Also published as: WO2024133895A1

Abstract

The invention relates to a method for monitoring an operating condition of a component (2) of a high voltage device (1), comprising the steps of collecting, during an operation of the high voltage device (1), a vibration signal of the component (2) with a wireless sensor (3), pre-processing the vibration signal by the wireless sensor (3) by determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component (2), transmitting, by the wireless sensor (3), the pre-processed parameters from the vibration signal to a processing device (11) external to the wireless sensor (3), and post-processing the transmitted pre-processed vibration signal by the processing device (11).

Description

Method for monitoring an operating condition of a component of a high voltage device

Technical Field

The invention relates to a method for monitoring an operating condition of a component of a high voltage device, comprising the steps of collecting, during an operation of the high voltage device, a signal of the component with a sensor, and transmitting the signal to a processing device external to the sensor. The invention further relates to an arrangement comprising a high voltage device having at least one component and sensor configured for collecting, during an operation of the high voltage device, a signal of the component.

Background Art

Most major failures of high voltage devices such as high voltages circuit breakers, HVCB, occur due to malfunctions of the HVCB's mechanical operating mechanism. Applying in-situ sensors connected to an edge device with analogue, binary and/or digital inputs are offered as monitoring systems by different manufacturers for detecting such operating mechanism failures as well as for identifying and counting a type of HVCB operations. However, installation of such type of monitoring system is normally invasive and requires a substantial cabling effort, which leads to adjustments of planning of substations’ cabling in case of a retrofit installation.

Usually, several sensor signals are collected during HVCB monitoring, such as motor current, trip coil current, contact travel curve and others. Condition-relevant features are derived from these signals, such as the number and type, for example close/open, pump starts, motor starts etc., of operations, travel time, speed etc. From such estimated features, HVCB heath condition can be assessed. HVCB operating mechanism systems are usually made of many different components, for example including trip coil, latch, springs, damper, pull-rod, joints etc. Ideally, in terms of condition assessment, each component would require a dedicated monitoring sensor. However, installing such dedicated monitoring sensors would economically and technically not feasible.

Summary of invention

It is therefore an object of the invention to provide a low-cost and quick-and-easy- to-install monitoring solution for detection of types of operation, such as for example close/open, pump start, motor start etc., and/or malfunctions of high voltage devices.

The object of the invention is solved by the features of the independent claims. Preferred implementations are detailed in the dependent claims.

Thus, the object is solved by a method for monitoring an operating condition of a component of a high voltage device, comprising the steps of collecting, during an operation of the high voltage device, a vibration signal of the component with a wireless sensor, pre-processing the vibration signal by the wireless sensor by determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component, transmitting, by the wireless sensor, the pre-processed parameters from the vibration signal to a processing device external to the wireless sensor, and post-processing the transmitted pre-processed vibration signal by the processing device.

A key point of the invention is therefore that signal processing consists of said preprocessing step comprising determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component being carried out by the wireless sensor as well as said post-processing step, which takes place in the external processing device. In particular for a battery powered wireless sensor large data transfers from the wireless sensor to the processing device would lead to a high battery energy consumption, which might make a battery powered sensors not feasible for various application. However, by using said preprocessing data transfer volume can be significantly reduced such that battery life is increased, which also increases applicability of the proposed solution. Further, installation of monitoring system is much simpler compared to prior art solutions, as installation is not invasive and does not require cabling effort, such that even retrofit installations at substations become possible.

Generally, since vibration data contain information for example about a whole mechanical switching process, application of wireless vibration sensors for high voltage devices such as high voltages circuit breakers, HVCB, monitoring is an advantageous approach. Vibration is usually present from a start to an end of the high voltage device operation; therefore, vibration monitoring covers all components that contribute to the mechanical signal in a single measurement. Even arcing information of HVCB is available in the vibration signal due to an arc pressure build up effects on the vibration. Thus, while theoretically multiple wireless sensors can be installed, for example individually positioned for individually determining trip coil, latch, springs, damper, pull-rod, joints etc. condition assessment of different components of the HVCB, the proposed solution also allows installing only one wireless sensor for determining trip coil, latch, springs, damper, pull-rod, joints etc. condition assessment of all different components.

Although detailed high voltage device respectively HVCB diagnostics can be achieved from the vibration signal with a relatively high resolution respectively bandwidth, collected data can be rather large. Since data transfer is one of the main issues in terms of energy consumption, this aspect is disadvantageous for wireless data transmission by a battery-powered sensor. Therefore, the proposed smart vibration sensor with integrated data analysis respectively pre-processing provides an advantageous solution to this problem, since data transmission between the vibration sensor and the external processing device can be limited to a few features that contain only relevant information for condition assessment of the high voltage device operating mechanism. In contrast to machines that are continuously running, like motors, a typical operation of the high voltage device respectively a switching event of the HVCB often leads to a finite vibration signal in time. Action of different components of the operating mechanism leads to mechanical events that leave distinct signatures in the vibration signal. With the proposed solution condition health assessment of different operating mechanism components, using as reference the vibration signal measured on its structure, is achieved.

The step collecting, during the operation of the high voltage device, the vibration signal of the component with the wireless sensor shall preferably be understood that the vibration signal characterizes a status, in particular a health assessment, of said component. The term component preferably refers to the operating mechanism of the high voltage device. The term pre-processing shall preferably not be understood in that simply the vibration signal is converted, for example from current to voltage, without a data reduction. Rather, the term pre-processing shall preferably be understood that a data reduction takes place in respect to the vibration signal. Thus, the term pre-processing shall preferably be understood that a data analysis takes place on the vibration signal resulting in a reduced data volume of the vibration signal.

The wireless sensor can generally be provided as any wireless sensor known from prior art configured for determining for example a health status of the component. The vibration signal preferably comprises a series of signals, in particular a series of signals over time. In this respect, also a plurality of sensors can be present. One of said plurality of sensors may collect the individual vibration signals and may conduct the pre-processing of all vibration signals and subsequently transmit the pre- processed vibration signals to the processing device. Alternatively, all wireless sensors may collect individual vibrations signals and pre-process the individually collected vibration signals.

According to a preferred implementation, physical properties comprise vibration peak intensities and/or time intervals between vibration peaks, signal characteristics comprise total or partial signal characteristics and/or number of vibration pulses, signal energy and/or energy-based signal entropy. Preferably, such determining comprises reducing a resolution, bandwidth and/or data volume of the vibration signal. Said physical quantities preferably relate to mechanical parts as components of the high voltage device, such as for example trip coil, latch, springs, damper, pullrod, joints of a HVCB.

In another preferred implementation, determining the vibration signal characteristic parameters in respect to physical properties comprises determining a time difference between distinct mechanical events and/or determining distinct mechanical events of the component and/or determining the vibration signal characteristic parameters in respect to signal characteristics comprises determining theoretical quantities that characterize the vibration signal. Such a distinct mechanical event can be for example travel start and travel end in case of a circuit breaker. Signal characteristics can be for example peaks of the vibration signal. Determining means in particular filtering.

According to a further preferred implementation, determining the vibration signal characteristic parameters in respect to physical properties related to the component comprises detecting amplitude peaks of the vibration signal as a function of time and/or determining the vibration signal characteristic parameters in respect to signal characteristics of the vibration signal comprises mapping the vibration signal to a real number by integration.

In another preferred implementation, the physical properties related to the component comprise a time difference between a trip coil armature start and travel onset and/or a time difference between a damper action onset and travel of the component and/or the signal characteristics comprise a number of vibration pulses, signal energy and/or energy-based signal entropy of the vibration signal.

According to a further preferred implementation, the processing device comprises a computer and/or a cloud-based computing device. The processing device can be installed, for example, in a data centre or distant to the high voltage device, for example close to an operator of the high voltage device, and/or can be provided cloud based. The term external to wireless sensor means that the processing device is not part of the wireless sensor, for example arranged distant to the high voltage device close an operator or within a data centre.

In another preferred implementation, post-processing is based on experimental data measured for seeded defects of the high voltage device and/or comprises machine learning. Thus, vibrations can be measured at high voltages devices provided with seeded defects, or from collected data of high voltages devices operating in a field. Thereby, in particular in case of sufficiently large data machine learning can be involved. Machine learning help to increase robustness of the proposed failure detection solution.

According to a further preferred implementation, the wireless sensor is battery powered. Said battery preferably provides power to the wireless sensor for a couple of days, a couple of weeks are a couple of months. The battery is preferably provided rechargeable. The wireless sensor may further comprise an indication, for example a LED, for indicating that the battery needs to be charged or replaced. In this respect the battery can also be provided replaceable. The battery may provide a voltage of 3 or 5 to V DC. Alternatively, it is possible that the wireless sensor is configured for receiving electrical power from a cabling connection. The high voltage device is preferably not battery powered.

In another preferred implementation, the wireless sensor communicates via Bluetooth Special Interest Group published Bluetooth Core Specification Version 5.3 for transmitting the pre-processed vibration signal. IEEE standardized Bluetooth as IEEE 802.15.1. Bluetooth Core Specification Version 5.3 was published on 14 July 2021. In a further preferred implementation, the wireless sensor communicates via LoRaWAN, "long range" WAN, LoRaWAN. LoRaWAN is typically understood as a cloud-based medium access control, MAC, layer protocol, preferably acting as a network layer protocol for managing communication between LPWAN gateways and end-node devices as a routing protocol, maintained by the LoRa Alliance. Preferably, version 1 .0.4 as published in October 2020 or a newer version is used for LoRaWAN. According to a further preferred implementation, the wireless sensor comprises a microprocessor configured for pre-processing the vibration signal and/or a wake-up function configure for wakening-up the microprocessor upon receipt of the vibration signal. The microprocessor may comprise a memory and/or of feature calculation means. The micro sensor can be provided as 8-, 16- or 32-bit microprocessor. The microprocessor can be configured for temporary storing the vibration signal, for example until a switching event of the HVCB is finished, for subsequently pre-processing the vibration signal. The wake-up function preferably shuts down the microprocessor after having pre-processed the vibration signal. For transmitting the pre- processed vibration signal the wireless sensor may comprise a wireless modem. The wireless sensor may be configured for transmitting the pre-processed vibration signal via Bluetooth, in particular Bluetooth Core Specification Version 5.3, and/or a W-LAN, GSM, LTE or 5G mobile communication network standard.

In another preferred implementation, the wireless sensor comprises an acceleration sensor and/or a microphone for collection the vibration signal, and preferably a current sensor, preferably a hall effect sensor. The vibration signal, the acceleration sensor and/or the microphone is preferably arranged adjacent to or touching with the component for such wise collecting the vibration signal. The wireless sensor can also include means for monitoring additional, different physical quantities such as electric current, for example by said hall effect sensor, and/or temperature. Said additional different physical quantities can be included in the vibration signal and/or pre-processed and transmitted as additional signal.

According to a further preferred implementation, the high voltage device is provided as high voltage circuit breaker, HVCB, as tap changer of a transformer and/or as a disconnector.

Circuit breakers are usually used for high voltage switching applications and are predominantly used for interrupting a current, when an electrical fault occurs. Thereby, circuit breakers have the task of opening contact elements and keeping them apart from one another for avoiding a current flow even in case of high electrical potential originating from the electrical fault itself. Such high voltage circuit breakers typically break high currents at voltages of 72 kV and up to 1200 kV and are arranged in the respective electrical circuits which are intended to be interrupted based on some predefined event occurring in the electrical circuit. The tap changer is preferably provided as a mechanism in a transformer allowing for variable turn ratios to be selected in distinct steps.

A disconnector or earthing switch, also known a grounding switch, is a protective device included in switchgear components like circuit breakers and isolators. When circuit breakers are removed and racked out, earthing switches automatically ground a part of a bus bar adjacent to the circuit breakers. For isolators, the earthing switches make contact with the bus bar when the isolator isolates the circuits, discharging any charges that may have gathered there. For example, an earthing switch in switchgear is used to ground a remaining change in a power line after the power line has been removed from its source. A residual charge often remains in a circuit after it has been severed or opened by the circuit breaker and isolator. An earthing switch is usually provided to discharge the charge.

The term high voltage preferably relates to voltages ranging from above 72,5 kV to 1200 kV, like 145 kV, 245 kV or 420 kV. Nominal currents of the circuit breaker can be preferably in the range from 2,5 kA to 5 kA with fault currents of 25 to 63 kA or 10 kA to 500 kA. The high voltage device and/or the high voltage circuit breaker can be provided as a gas-insulated circuit breaker, for example including an encapsulating housing which defines a volume for the gas. The circuit breaker can include a gas blowing system configured to extinguish the arc during a stage of the current interruption operation.

The object is further solved by an arrangement comprising a high voltage device having at least one component, a wireless sensor and a processing device external to the wireless sensor, whereby the wireless sensor is configured for collecting, during an operation of the high voltage device, a vibration signal of the component, configured for pre-processing the vibration signal by determining vibration signal characteristic parame- ters in respect to physical properties and/or signal characteristics related to the component (2) and configured for transmitting the pre-processed parameters from the vibration signal to the processing device, and the processing device is configured for post-processing the transmitted pre- processed vibration signal.

According to a preferred implementation, the wireless sensor comprises a microprocessor configured for pre-processing the vibration signal and/or a wake-up function configures for wakening-up the microprocessor upon receipt of the vibration signal.

In another preferred implementation the high voltage device is provided as high voltage circuit breaker, HVCB, as tap changer of a transformer and/or as a disconnector.

Further implementations and advantages of the arrangement are directly and unambiguously derived by the person skilled in the art from the method as described before.

Brief description of drawings

These and other aspects of the invention will be apparent from and elucidated with reference to the implementations described hereinafter.

In the drawings:

Fig. 1 shows an arrangement comprising a high voltage device having at least one component, a wireless sensor and a processing device external to the wireless sensor for carrying out a method for monitoring an operating condition of the component in a schematic view according to a preferred implementation,

Fig. 2 shows the wireless sensor of Fig. 1 in a schematic view according to the preferred implementation, Fig. 3 shows travel curve and vibration signal signatures given by physical phenomena for close and open operations of a circuit breaker as high voltage device according to the preferred implementation,

Fig. 4 shows a 3D graph plotting of different loads of an operating mechanism spring and different dynamic viscosity of a damper oil according to the preferred implementation, and

Fig. 5 shows a further 3D graph plotting of different loads of the operating mechanism spring and different dynamic viscosity of the damper oil according to the preferred implementation.

Description of implementations

Fig. 1 shows an arrangement comprising a high voltage device 1 , which is provided as a high voltage circuit breaker. The high voltage device 1 comprise as component 2 an operating mechanism for disconnecting two electrical contacts, which are not shown.

Firmly attached to the high voltage device 1 is a first wireless sensor 3, which is configured for collecting, during an operation of the high voltage device 1 , a vibration signal of the component 2. Said first wireless sensor 3 shown on the right side of Fig. 1 is provided as an accelerometer 4 connected to a housing of the high voltage circuit breaker. Fig. 1 shows a second wireless sensor 3, which is provided as microphone arranged on a stand a few centi-meters or meters distant to the high voltage device 1 . While Fig. 1 shows two wireless sensors 3, only one or both wireless sensors 3can present.

The wireless sensor 3 is shown in a schematic and more detailed view in Fig. 2. The wireless sensor 3 comprises said accelerometer 4 for data acquisition of the vibrations via one to three channels. The so acquired analogous sensor data is converted by a 16-bit AD-converter 5 to a digital vibration signal, which is provided to a hybrid 8- or 32-bit microprocessor 6 comprising a memory as storage and respective calculations means for pre-processing the vibration signal by determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component 2.

The pre-processed parameters from the vibration signal are then wirelessly transmitted by a wireless modem 7 by using Bluetooth Special Interest Group published Bluetooth Core Specification Version 5.3 as communication network protocol. Alternatively, protocols such as WLAN, GSM, LTE or 5G can be used, as indicated with cell towers 9 for the transmission in Fig. 1 . The so by cell towers 9 transmitted pre- processed vibration signal reaches via a cloud 10 a processing device 11 such wise external to the wireless sensor 3. The processing device 11 then post-processes the transmitted pre-processed vibration signal. Fig. 2 shows a computer as processing device 11 , while the cloud 10 may also form the processing device 11 .

The microprocessor 6 comprises a wake-up function, which wakens-up the microprocessor 6 and the wireless modem 7 upon receipt of the vibration signal and shuts-down the microprocessor 6 and the wireless modem 7 after having pre-processed the vibration signal. The wireless sensor 3 is battery powered by a battery 8 as power source delivering 5 V DC to the accelerometer 4, the AD-converter 5, the microprocessor 6 and the wireless modem 7. The battery 8 is rechargeable and allows operation of the wireless sensor 3 for a couple of months. In addition or alternatively to the battery 8 the wireless sensor 3 can comprise an energy harvesting device for supplying the accelerometer 4, the AD-converter 5, the microprocessor 6 and the wireless modem 7 with electrical energy, thereby avoiding recharging the battery 8.

As mentioned, signal processing consists of said pre-processing step carried out by the wireless sensor 3, as well as a post-processing step, which takes place in the processing device 11 , provided as an external computer or in the cloud 10. Due to the pre-processing step large data transfers from the wireless sensor 3 to the processing device 11 leading to a high battery energy consumption of the battery powered wireless sensor 3 can be avoided. Pre-processing generates just a few features from the vibration signal, which can be divided in two classes:

1. Physical properties, in particular quantities, related to components 2 such as mechanical parts of the circuit breaker, for example time difference between a trip coil armature start and travel onset, time difference between damper action onset and travel end etc., and

2. Total or partial signal characteristics, such as number of vibration pulses, signal energy, energy-based signal entropy etc.

Regarding the first class of features, Fig. 3 shows an example of some signatures of the vibration signal as amplitude peaks as a function of time given by mechanical events, which occur during HVCB open and close operations. Specifically, Fig. 3 shows travel curve, above, and vibration signal signatures, below, given by physical phenomena for close, left side, and open, right side, operations of a circuit breaker as high voltage device 1 in a schematic view according to a preferred implementation. Thereby, a few of the possible time differences are indicated in the graph.

In particular, At roughly characterizes the time difference between travel start and end, which is used as a reference as one of the features (K = At) to be considered for pre-processing. Further time-related features can be associated with other mechanical events, which are not considered here. In Fig. 3, only the vibration occurring during the corresponding operating mechanism travel time is shown. Nevertheless, longer vibration monitoring times could also provide information regarding further operations of a drive of the circuit breaker, such as the motor spring charging.

Regarding the second class of features, which correspond to signal characteristics of a whole vibration pattern of the vibration signal, said second class of features are generally functions that map a total vibration signal to a real number by integration. Some examples are the vibration energy,

K₂ = f A²dt or a quantity which is, up to the sign and a constant, related to a Shannon entropy.

In an experimental trial, a HVCB operating mechanism was prepared with different seeded defects. A large number of HVCB operations were performed, while vibration signals were measured. Then the characteristics K_{l t} K₂, and K₃, as described before, were calculated, and plotted as shown in Fig. 4 as results for different spring loads and different damper oil dynamic viscosities, which are the adopted seeded defects. Said results show a clear distinction between the different HVCB operating conditions, just based on the features calculated out of the vibration signal. Such analysis can also be performed in a multi-dimensional space, considering several different features, which need to be selected in association with the particularity of each considered failure mode.

Fig. 4 shows a 3D graph plotting the calculated features K_lt K₂, and K₃ to the left with different loads of the operating mechanism spring and to the right for different dynamic viscosity of the damper oil. Regions in the multi-dimensional space as marked in Fig. 4 can be defined in association with HVCB working condition probabilities, which are derived from the vibration signal.

As an example, a probability can be estimated that the monitored HVCB has its spring stiffness in a certain value regime, using as reference the boundaries of the defined regions. From the referred features, a working condition of the whole operating mechanism and its components can be concluded, which are prone to degradation and exhibit malfunction risks. This can be done based on pre-knowledge, e.g., by measuring vibration signals of circuit breakers with seeded defects, or from collected data of circuit breakers operating in the field, which can possibly involve machine learning in case of sufficiently large data. Thus, having pre-processed the vibration signal by the wireless sensor 3, the wireless sensor 3 does not submit the vibration signal but rather one or all said calculated features K_lt K₂, and K₃ as pre-processed vibration signal to the processing device 11 , which then post-processes the so received one or all said calculated features K_{l t} K₂, and K₃. A final health assessment, which can be reported as statements on a condition of the different components of the operating mechanism or even the whole circuit breaker, is then post-processed by the processing device 11 , provided as an external computer, or as cloud 10.

Fig. 5 shows a further 3D graph plotting the calculated features K and K₂. Such wise a machine learning method can be used by detecting malfunctions by a “shift” of specific features in a certain direction. For example, using Fig. 5 as reference, if K1 increases this gives indication that there is a low spring load or stiffness. In contrast, if K2 increases it’s an indication that the damper is degraded. If both damper and spring have a malfunction, both K1 and K2 shift together. The highlighted clusters of points correspond to specific working conditions. The failure borderline of the clusters can be predefined based on given tolerances, or statistically determined; in other words, if feature values are detected out of a certain interval, it can trigger an alarm.

Experiments have been performed with “seeded defects” such that it became known what is going wrong with the operating mechanism, identifying changes in the feature values afterwards. In the field, it can work the other way around. Changes can be detected in feature values which were not seen before. In this case, an inspection is expected to be performed in the breaker to eventually identify some malfunction; in case a malfunction is identified, this information is added to the machine learning algorithm for further similar situations, which can be applied for breakers from the same “family” in a fleet.

Thus, the machine learning algorithms are able to detect a failure from an association of feature values with known former failure modes; however, this needs to be learned based on experience coming from lab tests, like the ones preformed, or directly from the field application. An additional important point is that the given example shows failure detections based on single features K and K₂, whereas other failures might be indicated by a combination of several features, preferably in a multi-dimensional space. This can be performed by any of the before cited machine learning methods, such as CART, random forests, KNN etc.

In sum, the proposed method can be used in combination with machine learning approaches to increase the robustness of the failure detection solution, using measurements collected from HVCBs operating in the field as training data. Application of machine learning methods in a cloud instance allow applying fleet information as training data for an identification of HVCB operations, while also being used as a storage place for collected vibration signals.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed implementations. Other variations to be disclosed implementations can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting scope.

Reference signs list

1 high voltage device

2 component

3 wireless sensor

4 accelerometer

5 A/D-converter

6 microprocessor

7 modem

8 battery

9 cell tower

10 cloud

11 processing device

Claims

1 . Method for monitoring an operating condition of a component (2) of a high voltage device (1 ), comprising the steps of collecting, during an operation of the high voltage device (1 ), a vibration signal of the component (2) with a wireless sensor (3), pre-processing the vibration signal by the wireless sensor (3) by determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component (2), transmitting, by the wireless sensor (3), the pre-processed parameters from the vibration signal to a processing device (11 ) external to the wireless sensor (3), and post-processing the transmitted pre-processed vibration signal by the processing device (11 ).

2. Method according to the previous claim, whereby physical properties comprise vibration peak intensities and/or time intervals between vibration peaks, signal characteristics comprise total or partial signal characteristics and/or number of vibration pulses, signal energy and/or energy-based signal entropy.

3. Method according to any of the previous claims, whereby determining the vibration signal characteristic parameters in respect to physical properties comprises determining a time difference between distinct mechanical events and/or determining distinct mechanical events of the component (2) and/or determining the vibration signal characteristic parameters in respect to signal characteristics comprises determining theoretical quantities that characterize the vibration signal.

4. Method according to any of the previous two claims, whereby determining the vibration signal characteristic parameters in respect to physical properties related to the component (2) comprises detecting amplitude peaks of the vibration signal as a function of time and/or determining the vibration signal characteristic parameters in respect to signal characteristics of the vibration signal comprises mapping the vibration signal to a real number by integration

5. Method according to any of the three previous claims, whereby the physical properties related to the component (2) comprise a time difference between a trip coil armature start and travel onset and/or a time difference between a damper action onset and travel of the component (2) and/or the signal characteristics comprise a number of vibration pulses, signal energy and/or energy-based signal entropy of the vibration signal.

6. Method according to any of the previous claims, whereby the post-processing device (11 ) comprises a computer and/or a cloud-based computing device.

7. Method according to any of the previous claims, whereby post-processing is based on experimental data measured for seeded defects of the high voltage device (1 ) and/or comprises machine learning.

8. Method according to any of the previous claims, whereby the wireless sensor (3) is battery (8) powered.

9. Method according to any of the previous claims, whereby the wireless sensor (3) communicates via Bluetooth Special Interest Group published Bluetooth Core Specification Version 5.3 for transmitting the pre-processed vibration signal.

10. Method according to any of the previous claims, whereby the wireless sensor (3) comprises a microprocessor (6) configured for pre-processing the vibration signal and/or a wake-up function configured for wakening-up the microprocessor upon receipt of the vibration signal.

11 . Method according to any of the previous claims, whereby the wireless sensor (3) comprises an acceleration sensor (4) and/or a microphone for collection the vibration signal, and preferably a current sensor, preferably a hall effect sensor.

12. Method according to any of the previous claims, whereby the high voltage device (1 ) is provided as high voltage circuit breaker, HVCB, as tap changer of a transformer and/or as a disconnector.

13. Arrangement comprising a high voltage device (1 ) having at least one component (2), a wireless sensor (3) and a processing device (11 ) external to the wireless sensor (3), whereby the wireless sensor (3) is configured for collecting, during an operation of the high voltage device (1 ), a vibration signal of the component (2), configured for preprocessing the vibration signal by determining vibration signal characteristic parameters in respect to physical properties and/or signal characteristics related to the component (2) and configured for transmitting the pre-processed parameters from the vibration signal to the processing device (11 ), and the processing device (11 ) is configured for post-processing the transmitted pre-processed vibration signal.

14. Arrangement according to the previous arrangement claim, whereby the wireless sensor (3) comprises a microprocessor configured for pre-processing the vibration signal and/or a wake-up function configures for wakening-up the microprocessor upon receipt of the vibration signal.

15. Arrangement according to any of the two previous arrangement claims, whereby the high voltage device (1 ) is provided as high voltage circuit breaker, HVCB, as tap changer of a transformer and/or as a disconnector.