Monitoring system and method for monitoring a voltage network

Abstract

The invention relates to a monitoring system for monitoring a voltage network (12), having a protection device (16) which is to be incorporated into the voltage network (12) and comprises at least one triggering protection component (18). The protection device (16) has at least one communications interface (26) designed to transmit a status value of the protection device (16) to an evaluation unit (30) which is separate from the protection device (16) and is designed to perform an evaluation so that an outsourced evaluation is provided. The invention also relates to a method for monitoring a voltage network (12).

Description

[0001] Monitoring system and method for monitoring a voltage network

[0002] The invention relates to a monitoring system for monitoring a voltage network. Furthermore, the invention relates to a method for monitoring a voltage network using a monitoring system.

[0003] Various protective devices are known from the state of the art, such as surge protection devices (SPDs), also known as surge arresters. Such protective devices are integrated into a voltage network to be monitored and protected, for example, a low-voltage network, which is also referred to as the building network.

[0004] In the event of an overvoltage, for example, due to a lightning strike, the protective device serves to divert the existing overvoltage in order to protect devices in the power grid from the effects of the overvoltage. For this purpose, the protective device typically has at least one triggering protective component that responds in the event of an overvoltage and transitions to a conductive or quasi-conductive state, thereby diverting the overvoltage because an associated current flows. The triggering protective component is therefore also referred to as a disconnecting device and is designed, for example, as a spark gap.

[0005] Typically, such protective devices have an indicator element that indicates the current state of the protective device, for example, by means of a mechanically variable indicator element that is green or red, with the respective color depending on the state of the protective device, in particular the state of the triggering protective component. Such a surge device is known, for example, from DE 10 2019 110 745 B3.

[0006] WO 2016 / 026522 A1 discloses a surge protection device that, like the protection device in DE 10 2019 110 745 B3, performs an evaluation or diagnosis. For this purpose, the surge protection device in WO 2016 / 026522 A1 has a diagnostic circuit that performs an evaluation or diagnosis and then sends the result of the diagnosis or evaluation to a controller.

[0007] However, the current state-of-the-art protective devices are only designed to safely and reliably reduce the existing overvoltage in order to prevent overloading the power grid and to restore operation as quickly as possible once the overvoltage event has cleared. Furthermore, the current status of the protective device is displayed locally, so that the user is informed of the current status. However, such protective devices do not allow for additional information.

[0008] The object of the invention is therefore to further develop such protective devices so that more information is provided.

[0009] The object is achieved according to the invention by a monitoring system for monitoring a voltage grid, wherein the monitoring system comprises a protective device to be integrated into the voltage grid, which protective device comprises at least one triggering protective component. The protective device has at least one communication interface configured to transmit a status value of the protective device to an evaluation unit configured separately from the protective device. The evaluation unit is configured to perform an evaluation, thus providing an evaluation separate from the protective device.

[0010] The basic idea of ​​the invention is to provide a separate evaluation unit, whereby the protective device itself can be designed to save space. The intelligence or computing power required for the evaluation can, however, be implemented in an evaluation unit designed separately from the protective device, thereby creating external logic that communicates with the at least one protective device to be integrated into the voltage grid, for example via a computer network such as the Internet. The separate evaluation unit therefore ensures that the evaluation of the status value transmitted by the protective device takes place outside the protective device. This evaluation is also referred to as diagnostics. Accordingly, the status value is transmitted in the form of data to the separately designed evaluation unit. Transmission protocols such as TCP, in particular TCP / IP, can be used for this purpose.

[0011] In any case, the separately designed evaluation unit provides correspondingly high computing power, allowing even complex evaluations to be performed that could not be implemented with a conventional protective device. This is due, among other things, to the limited space available at the installation site, typically in the area of ​​a DIN rail.

[0012] In principle, the monitoring system can comprise multiple protective devices that are integrated into the same voltage network or into different voltage networks. The multiple protective devices can send data to the same evaluation unit, in particular, they can communicate with the evaluation unit. This makes it possible for the separately designed evaluation unit to receive multiple status values ​​from different protective devices, giving the evaluation unit more data and information available for evaluation.

[0013] The analysis can thus be carried out based on a sufficiently large database, allowing correlations between different data to be identified.

[0014] For example, correlations can be identified based on the joint evaluation of different condition values ​​obtained from different protection devices.

[0015] The triggering protective component can be a (horn) spark gap, which changes into a conductive or quasi-conductive state when an overvoltage is present.

[0016] One aspect provides that the status value of the protective device is a measured value detected by the protective device or a value representative of a tripping state of the tripping protective component of the protective device. The protective device can, for example, have a measuring and / or sensor unit via which the corresponding measured value is detected. In particular, the measured value is a value of a characteristic parameter of the voltage network. Based on the characteristic parameter, the status of the voltage network can be determined.

[0017] In principle, the characteristic quantity can be a voltage, a current, a power, a frequency, a distortion, a harmonic, a reactive power and / or an energy value of the voltage network, in particular of a phase of a multi-phase voltage network.

[0018] In particular, the protective device is designed to have multiple measuring and / or sensor units. This allows multiple characteristic parameters of the voltage network to be determined and transmitted to the evaluation unit, thus enabling more precise evaluation using the separately designed evaluation unit.

[0019] The value representative of a tripping state of the tripping protection component of the protection device can be a binary value. Thus, "0" can represent the non-tripped state of the protection component, whereas "1" represents the tripped state of the tripping protection component. Therefore, the protection device can use the representative value to transmit via the communication interface whether the tripping protection component has tripped or not, i.e., whether an overvoltage event has occurred.

[0020] Another aspect provides for the protective device to be a surge protection device. In this respect, the protective device can be designed as a surge protection device (SPD). The protective device therefore serves to divert a surge, so that devices and components integrated in the power grid are safely and efficiently protected from the surge. Since the protective device also has the communication interface with which the protective device transmits data to the separately designed evaluation unit, it is ensured that corresponding information regarding the status of the protective device, i.e., a tripping condition, and / or a measured value recorded by the protective device is or will be transmitted to the separately designed evaluation unit, which is configured to perform a corresponding evaluation in order to conduct more in-depth analyses.In addition to the tripping protective component, the protective device can also include a voltage-limiting protective component. For example, the voltage-limiting protective component is a metal oxide varistor. The tripping protective component and the voltage-limiting protective component can be connected in series.

[0021] According to a further aspect, the system comprises an evaluation unit which is designed separately from the protective device and has at least one communication interface which is configured to communicate with the communication interface of the protective device. The communication interface, i.e. the communication interface of the protective device and / or the communication interface of the evaluation unit, can be designed as a bidirectional communication interface. In this respect, the monitoring system comprises not only the protective device but also the separately designed evaluation unit which communicates with the protective device. The evaluation unit and / or the protective device can each have a bidirectional communication interface so that data can be both received and sent. This results in bidirectional communication between the protective device and the evaluation unit.

[0022] Alternatively, the protective device can have two separate communication interfaces and / or the evaluation unit can have two separate communication interfaces, whereby unidirectional communication or data transmission takes place between the corresponding communication interfaces. This can be important if data transmission in at least one direction is to have a higher safety requirement than data transmission in the other direction.

[0023] In principle, the evaluation unit can be a server-based evaluation unit, for example a cloud system. In this respect, it is possible for the protective device to transmit the at least one status value to a cloud that is in communication with multiple devices, in particular multiple protective devices. In particular, it is therefore possible for the evaluation unit to receive data from numerous devices, i.e. also devices other than the protective devices, which can be used for the evaluation. Furthermore, it is possible for separate access to the server-based evaluation unit to be possible, so that an authorized user has access to corresponding data that has been transmitted from the at least one protective device to the server-based evaluation unit. In other words, it is possible for a user or operator of the monitoring system to have access to the evaluation results that are stored in the cloud.It is therefore possible for the evaluation results determined by the evaluation unit to be stored in the cloud so that they can be accessed. However, it is also possible for the evaluation results to be transmitted from the evaluation unit, for example, to the protection device, so that the evaluation results are available directly on-site.

[0024] According to a further aspect, the evaluation unit is configured to receive input data from at least one database or a measuring device. This is particularly important if the evaluation unit is a server-based evaluation unit that communicates with multiple devices installed at different locations. In particular, it is possible for data from a database to be included in the evaluation, providing additional information, thereby improving the evaluation of the status value of the protective device.

[0025] This can include weather, climate, or general environmental data. Lightning events or similar events can be included from a database, thereby improving the evaluation of the protection device's status value. Time data (weekday, weekend, holiday, day and / or night) or usage data from nearby systems, such as a train timetable or usage data from nearby electric charging systems, can also be obtained from the database.

[0026] The additional measuring device can be a device that is directly integrated into the voltage grid but does not have a protective function for the voltage grid, for example, a simple current or voltage measuring device. Furthermore, the measuring device can be a device that measures a value that is independent of the voltage grid, for example, a humidity sensor or an ambient temperature sensor, or similar. The monitoring system can therefore include at least one second value in the evaluation in addition to the status value transmitted by the protective device integrated into the voltage grid. The second value can be a value from a database, a measured value, or a characteristic parameter of the voltage grid.

[0027] In this respect, the monitoring system may comprise at least a database and / or an additional measuring device.

[0028] Furthermore, the evaluation unit can have at least one processor unit configured to process at least the state value received from the protective device. The processor unit can be configured to execute algorithms to determine an evaluation result based on the received state value, for example, an output value or state, in particular a state characterizing the voltage network or the protective device.

[0029] This can be a (current) aging state of the protective device or a (current) state of the voltage network, i.e. the network state of the voltage network.

[0030] The corresponding condition can also be predicted, meaning that a future aging state of the protection device or a future state of the voltage grid can be predicted by at least processing the status value of the protection device using the evaluation unit. This allows, for example, future grid quality and aging-related problems to be predicted, allowing appropriate countermeasures to be initiated in a timely manner.

[0031] In particular, the processor unit comprises artificial intelligence configured to predict a state of the voltage grid and / or a state of the protection device based on the state value received from the protection device. This can be a failure prediction for this protection device or a state prediction of the voltage grid based on previously acquired data relating to the voltage grid, which has been transmitted to the evaluation unit by the at least one protection device, in particular additionally by other devices or the database. Artificial intelligence allows large amounts of data, particularly of different types, to be processed and correlations between the different data to be identified, enabling more in-depth analyses.

[0032] In this respect, the evaluation unit, in particular the processor unit comprising artificial intelligence, can predict future conditions based on historical data, thus enabling the user or operator to initiate appropriate countermeasures in a timely manner to counteract a predicted poor power quality (condition) of the voltage network or a predicted fault in the protection device. For example, it is possible that an impending failure of the protection device is detected early, allowing the protection device to be replaced in a timely manner, resulting in no or only minimal downtime, since the replacement device is already in stock or the protection device is replaced before the actual failure.

[0033] Furthermore, the separately designed evaluation unit can be configured to determine an evaluation result, in particular an output value and / or a state, based on the state value of the protective device. The evaluation unit thus processes the state value transmitted by the protective device in order to obtain the evaluation result understandable to the user, for example, the output value or the state.

[0034] In addition, the separately designed evaluation unit can be configured to transmit the determined evaluation result to the protective device or a display device. This makes it possible for the evaluation result (easily understandable for the user) to be displayed on the protective device itself or on a display device located in the installation area of ​​the protective device. This ensures that the user at the location of the protective device is informed about the protective device, especially its status.

[0035] A further aspect provides that the protective device has at least one display that is configured to show an evaluation result from the evaluation unit. The display can be a digital display, for example in the form of a screen or at least one LED. There can be multiple LEDs, in particular multi-colored ones, for example red, yellow or orange and green. In particular, the display is connected to the communication interface of the protective device in a signal-transmitting manner, so that the data received via the communication interface that corresponds to the evaluation result is converted by the protective device to control the display, i.e. the screen or the LED(s). The evaluation unit therefore transmits the evaluation result to the protective device so that the evaluation result can be displayed at the location of the protective device.

[0036] Alternatively, the display can also be a mechanical display, such as a flap or slider, which is moved mechanically to indicate the evaluation result, particularly by making a red or green area visible. The mechanical display is also (indirectly) controlled based on the data received from the communication interface that corresponds to the evaluation result, as the protective device implements this accordingly, for example, to control an actuator that interacts with the mechanical display.

[0037] In addition, the monitoring system can have a display that is designed separately from the protective device, for example in a separately designed display unit. The display unit can be assigned to the voltage network in which the protective device is integrated. This makes it possible for a display or message to be made in a location that is easier to access or more frequently visited than, for example, a technical room in which the protective device is usually installed. This ensures that an operator of the monitoring system receives the relevant information from the evaluation unit in a timely and reliable manner, for example a corresponding recommendation for action to counteract a failure of the protective device or a predicted poor condition of the voltage network.

[0038] In principle, it is possible for the display to provide additional information, particularly recommended actions. This can be shown on the display, allowing a user or operator of the monitoring system to be instructed to implement appropriate measures. There is no direct signal transmission between the at least one triggering protective component and the display, i.e. there is no wired, signal line, or wireless connection via which signals are transmitted. Ultimately, the measured value or the value representative of the tripping state of the triggering protective component of the protective device is recorded and transmitted by the protective device via the communication interface to the separately designed evaluation unit, which communicates with the protective device to transmit the evaluation result to be output via the display.In this respect, there is no direct data transmission between the triggering protective component of the protective device and the display of the protective device, but only an indirect one via the separately designed evaluation unit, which communicates with at least one communication interface.

[0039] The protective device may have a housing in which at least the triggering protective component and the communication interface are at least partially housed. In this respect, the protective device is a single component that encompasses the respective components and protects them from the environment.

[0040] The invention further provides a method for monitoring a voltage network by means of a monitoring system, the method comprising the following steps:

[0041] Recording a status value using a protective device that is integrated into the voltage network to be monitored,

[0042] Transmitting the recorded status value to an evaluation unit designed separately from the protective device, and

[0043] Processing the transmitted status value using the evaluation unit so that an outsourced evaluation takes place.

[0044] This ensures that the status value recorded by the protection device can be processed using complex evaluation techniques to determine an output value or predict the status of the protection device or the voltage grid. The evaluation result, for example, the correspondingly determined output value or the predicted status, can be transmitted to the protection device or to a display device designed separately from the protection device, which can be assigned to the voltage grid in which the protection device is integrated. If the evaluation result is transmitted to the protection device, corresponding bidirectional communication results between the protection device and the evaluation unit, even if this occurs via unidirectional communication interfaces.

[0045] For example, the evaluation result or a data representation thereof is displayed on the protection device itself or on a separate display device, informing a user or operator of the monitoring system. The operator can then initiate appropriate measures to prevent a failure of the protection device or minimize the downtime, or to stabilize the voltage grid to counteract a predicted poor power quality.

[0046] Furthermore, the evaluation unit can determine an evaluation result, in particular an output value and / or a status. The evaluation unit transmits the evaluation result back to the protective device or a display device, where the evaluation result is output. This ensures that the evaluation result (easily understandable for the user) is output, for example, at the location of the protective device, for example, by the protective device itself, in particular a display of the protective device, or by a separately designed display device located in the vicinity of the protective device.

[0047] In principle, it is therefore possible to estimate or predict future grid quality or the condition of the protective device based on historical and / or current data, enabling the grid operator of the voltage grid and / or the electricity customer / user to detect future disruptions at an early stage. In other words, the grid operator and / or the electricity customer / user can predict the time of an impending grid disruption and its intensity. This makes it possible to initiate appropriate countermeasures to avert the impending grid disruption, at least reduce its intensity, or generally minimize the impact of the impending grid disruption on certain areas. Likewise, the aging state (wear) of the protective device can be predicted, allowing an impending fault in the protective device to be detected at an early stage.

[0048] Due to the volatile nature of voltage grids, the future behavior of the voltage grid in terms of its network quality (network status) is often more important for the grid operator and / or the electricity customer / user than the current grid status, which cannot be responded to anyway. This applies analogously to the protection device, whose operational capability must be ensured for the future. Looking ahead makes it possible to initiate appropriate measures in a timely manner to counteract the predicted grid disruption, i.e., the deterioration of the voltage grid condition, or the failure of the protection device.

[0049] In other words, the grid operator and / or the electricity customer / user will be able to make the voltage grid more resilient to expected future disruptions or to replace the protective device.

[0050] The future state of the voltage grid or the protection device can be predicted for the next few hours, days or weeks, giving the grid operator and / or the electricity customer / user sufficient time to initiate any countermeasures in a timely manner to counteract any predicted disruption or deterioration.

[0051] The countermeasures to be taken regarding the voltage grid could, for example, consist of avoiding further loads during a predicted weak phase of the voltage grid, thereby ensuring the fail-safe operation of devices and / or machines whose operation should not be disrupted. In this respect, the countermeasure could consist of disconnecting non-essential machines or devices from the grid during the predicted weak phase of the voltage grid or not connecting them to the voltage grid in order to reduce the load or prevent it from increasing it further. With regard to a predicted failure of the protective device, the countermeasures could consist of procuring and installing a replacement device in a timely manner before the protective device actually fails.In principle, the voltage grid can be a local network, also known as a building network, assigned to a private household or an industrial building. The voltage grid can therefore be a low-voltage network, for example, a building network connected to a utility grid.

[0052] For forecasting purposes, the state value can be recorded multiple times, i.e., at least at two different points in time, and transmitted to the evaluation unit. The values ​​recorded at the different points in time can be fed into the evaluation unit's processor unit for evaluation. This allows the future time course to be forecast from a historically recorded history of the state values, which can extend up to the present. The future time course can refer to the recorded state value itself, so that its future time course is forecast, which corresponds to the forecast future state of the voltage grid or the protective device.

[0053] When evaluating additional data on the state value, a data fusion of two different variables can occur, allowing a glimpse into the future of the respective state. In other words, when the two different variables are processed together, a correlation between these variables is exploited, allowing conclusions to be drawn about future behavior.

[0054] The artificial intelligence can receive the state value recorded at at least two different points in time as an input and output the future state as an output. The artificial intelligence can therefore be trained to recognize the corresponding correlation between the state values ​​recorded at the different points in time and to learn the corresponding technical relationships, thus enabling reliable or highly accurate forecasts regarding the future state of the voltage grid or the protective device.

[0055] In particular, a method for training an artificial intelligence to predict a future state of a voltage grid or a protective device is therefore also provided. The training method comprises the steps of: providing a training data set for the artificial intelligence, which includes at least one state value at a first time, the state value at a second time, and an actual state of the voltage grid or the protective device at a third time, which is later than the first time and the second time,

[0056] Feeding the state value at the first time and the state value at the second time into a processor unit having the artificial intelligence to be trained, wherein the processor unit having the artificial intelligence jointly processes the state value recorded at the different times and outputs a predicted future state of the voltage network or the protective device at the third time,

[0057] Comparing the predicted future state of the voltage system or the protection device at the third time with the actual state of the voltage system or the protection device at the third time, which is part of the training data set, in order to determine a deviation between the predicted future state of the voltage system or the protection device at the third time and the actual state of the voltage system or the protection device at the third time, and

[0058] Feeding back the deviation between the predicted future state of the voltage system or the protection device at the third time point and the actual state of the voltage system or the protection device at the third time point in order to adjust weighting factors of the artificial intelligence to be trained, provided that the deviation lies outside a tolerance range.

[0059] The artificial intelligence can therefore be trained using at least one training data set, which includes the state value at both a first point in time and a second point in time, in particular a corresponding time sequence or time series of the state value, as well as an actual future state of the voltage grid or the protective device that exists at the third point in time, which is chronologically subsequent to the points in time at which the state value was determined or recorded. In particular, the first point in time and the second point in time do not coincide, so that the state value was recorded at two different points in time.

[0060] In principle, when training the artificial intelligence, it can also be provided that the artificial intelligence does not output the state of the voltage network or the protection device at a discrete point in time, but at an interval that includes the third point in time, i.e. predicts the future state of the voltage network or the protection device for a future period.

[0061] The corresponding tolerance range for the deviation can be predefined and / or set by a user, e.g. as a percentage or variance.

[0062] In principle, artificial intelligence trained using the method described above can be used to predict the future state of the voltage network or the protection device, i.e. for the evaluation of the state value that has been transmitted to the evaluation unit.

[0063] In other words, the processor unit comprises a trained artificial intelligence, in particular an artificial intelligence trained according to the method described above.

[0064] One aspect provides for the artificial intelligence to comprise at least one artificial neural network, e.g., a recurrent neural network (RNN) or a convolutional neural network (CNN). The artificial intelligence may comprise a long short-term memory (LSTM) network or a gated recurrent unit (GRU). Such neural networks allow a prediction of a future state based on time series. The artificial intelligence thus learns from past data (historical data), predicting the future state of the voltage grid or the protection device.

[0065] Further advantages and features of the invention will become apparent from the following description and the drawings, to which reference is made. Figure 1 shows a schematic representation of a monitoring system according to the invention, and

[0066] Figure 2 shows an overview of the method according to the invention for monitoring a voltage network.

[0067] Figure 1 shows a monitoring system 10 which is intended to monitor a voltage network 12, which is a building or low-voltage network which is connected to a supply network 14.

[0068] In the embodiment shown, the voltage network 12 is shown as a multi-phase building or low-voltage network.

[0069] The monitoring system 10 includes a protective device 16, which is integrated into the voltage network 12. The protective device 16 is designed as a surge protection device (SPD), also referred to as a surge arrester.

[0070] In this respect, the protective device 16 comprises a triggering protective component 18, which, when triggered, transitions to a conductive or quasi-conductive state to divert an overvoltage occurring during an overvoltage event such as a lightning strike, allowing a corresponding current to flow away, thus protecting devices integrated into the voltage network 12. After the overvoltage event has ended, the protective device 16 or the triggering protective component 18 returns to its initial state. The triggering protective component 18 can be designed as a (horn) spark gap or a gas discharge arrester ("GDT").

[0071] Furthermore, in the embodiment shown, the protective device 16 comprises, in addition to the triggering protective component 18, a voltage-limiting protective component 20 connected in series with the triggering protective component 18. For example, the voltage-limiting protective component 20 is designed in the form of a metal oxide varistor.

[0072] Furthermore, the protective device 16 is provided with an integrated measuring and / or sensor unit 22, which is arranged in the protective device 16. This means that both the triggering protective component 18, the optionally provided voltage-limiting protective component 20, and the optionally provided measuring and / or sensor unit 22 are enclosed by a housing 24 of the protective device 16. In other words, the housing 24 surrounds the aforementioned components 18 to 22, thereby protecting them from the environment.

[0073] In addition, the protective device 16 comprises a communication interface 26, with which the protective device 16 establishes a communication connection 28 with a separately formed evaluation unit 30, which also has a corresponding communication interface 32. The communication interface 26 of the protective device 16 is at least partially accommodated in the housing 24, thereby also protecting it from environmental influences.

[0074] The separately designed evaluation unit 30 is part of the monitoring system 10, wherein the separately designed evaluation unit 30 does not have to be provided at the same location as the protective device 16, but is arranged, for example, in another room or another building.

[0075] In particular, the evaluation unit 30 can be configured as a server-based evaluation unit, i.e., in a cloud. In this respect, the communication connection 28 between the protective device 16 and the evaluation unit 30 is established via an internet connection.

[0076] If the separately designed evaluation unit 30 is provided at the location of the protective device 16, the corresponding communication connection 28 can be formed by a Bluetooth connection or the like, in particular a wireless communication connection that has a short range.

[0077] In principle, the communication connection 28 between the protective device 16 and the separately designed evaluation unit 30 can be wired, for example via Ethernet cable, or wireless, for example via WLAN.

[0078] The communication interfaces 26, 32 of the protective device 16 and the evaluation unit 30 can each be designed as bidirectional communication interfaces, so that bidirectional communication between the protective device 16 and the evaluation unit 30 is possible. Alternatively, both the protective device 16 and the evaluation unit 30 can each have two communication interfaces that only allow a unidirectional communication connection, so that two unidirectional communication connections 28 are present between the protective device 16 and the evaluation unit 30. Overall, this then results in bidirectional communication between the protective device 16 and the evaluation unit 30. This embodiment with the separate communication interfaces is particularly important when the data transmission in one direction uses a different protocol orshould have a different security level than the data transmission in the other direction.

[0079] However, it can also be provided that the protective device 16 has only a unidirectional interface as a communication interface, via which the protective device 16 can only transmit data to the evaluation unit 30. In other words, the protective device 16 is then not designed to receive data from the evaluation unit 30. The communication connection 28 is then located between the one unidirectional interface and the at least one communication interface 32 of the evaluation unit 30.

[0080] In principle, it is possible for the protective device 16 to transmit a state value to the separately designed evaluation unit 30 via the communication connection 28, so that the evaluation unit 30, which has a processor unit 34, is configured to process the corresponding state value in order to obtain an evaluation result, for example to predict an output value or a state, which will be explained in more detail below.

[0081] The evaluation unit 30 can then transmit the evaluation result back to the protective device 16 via the communication connection 28, provided that the latter is designed to receive data, as shown in the embodiment according to Figure 1.

[0082] The protective device 16 can have a display 36 that is connected to the communication interface 26 for signal transmission, so that the evaluation result transmitted by the evaluation unit 30 can be displayed on the display 36 of the protective device 16. The display 36 can be a screen. The evaluation result itself, i.e., the output value or the predicted state, or a data representation thereof, can be displayed accordingly.

[0083] Alternatively, it can be provided that the evaluation unit 30 does not transmit the evaluation result directly to the protective device 16, but rather to a separately designed display device 38 which has a corresponding communication interface 40 with which the display device 38 communicates with the evaluation unit 30, in particular the communication interface 32 of the evaluation unit 30. This is particularly the case if the protective device 16 itself is not designed to receive data, since it only has a unidirectional communication interface, for example, which is only intended for sending data, i.e. to transmit the status value to the evaluation unit 30.

[0084] The display device 38 also has a display 42, via which the corresponding evaluation result, in particular a data representation thereof, can be displayed in order to inform a user or operator of the monitoring system 10 accordingly. The display 43 of the display device 38 can also be designed as a screen.

[0085] In principle, the data representation can also include recommendations for action or similar, so that the user or operator is instructed to take certain actions that are intended, for example, to counteract a predicted state of the voltage network 12 and / or the protective device 16.

[0086] The state value transmitted by the protective device 16 can be a measured value detected by the protective device 16 or a value representative of a tripping state of the tripping protective component 18 of the protective device 16.

[0087] The respective measured value can be a measured value that has been recorded by the measuring and / or sensor unit 22. In particular, the measured value is then a value of a characteristic parameter of the voltage network 12. The characteristic parameter of the voltage network 12 can be a voltage, a current, a power, a frequency, a distortion ("Total Harmonic Distortion" - THD), a harmonic (up to the 50th harmonic), a reactive power, and / or an energy value.

[0088] The value representative of the tripping state of the tripping protection component 18 can be binary, i.e. “0” for a non-tripped state or “1” for a tripped state.

[0089] As a result, the protective device 16 transmits corresponding information and data to the evaluation unit 30, which can be used by the evaluation unit 30 to determine an aging state of the protective device 16 in general or a state of the voltage network 12, in particular to predict a corresponding future state of the protective device 16 or of the voltage network 12.

[0090] In principle, the evaluation unit 30 can also communicate with at least one further device, for example a database 44, a measuring device 46 which is provided separately from the voltage network 12, i.e. does not record any data or information of the voltage network 12, a measuring device 48 which is designed separately from the protective device 16 but records information or data of the voltage network 12, and / or a further protective device 16 which is assigned to a different voltage network.

[0091] The evaluation unit 30 thus receives further input data which can be taken into account for the evaluation of the status value of the protective device 16, so that a joint evaluation of different data takes place, whereby, for example, correlations between the different data can be detected.

[0092] In particular, the evaluation unit 30, i.e. the processor unit 34, has an artificial intelligence 50 which is configured to predict a (future) state of the voltage network 12 and / or a (future) state of the protective device 16 at least based on the state value received from the protective device 16.

[0093] This allows a failure prediction of the protective device 16 or the voltage network 12 to be made. Likewise, an aging state of the protective device 16 and / or other events in the voltage network 12 can be predicted, for example, the future power quality of the voltage network 12. This is done based on the input data processed by the artificial intelligence 50.

[0094] Using artificial intelligence 50, large amounts of data, especially data of different types and / or origins, can be processed together and correlations between the different data can be identified, enabling more in-depth analyses that cannot easily be carried out on-site due to the required computing power.

[0095] In principle, the processor unit 34 of the evaluation unit 30 can be configured to execute a computer program with program code means in order to monitor the voltage network 12.

[0096] The computer program can be installed on the processor unit 34 from a computer-readable data carrier on which the computer program is stored, so that the monitoring system 10, in particular the evaluation unit 30 having the processor unit 34, is able to carry out a corresponding method for monitoring the voltage network 12.

[0097] The evaluation unit 30 accordingly has a corresponding interface via which the data carrier can be coupled to the evaluation unit 30 in order to install the computer program.

[0098] The computer program may also have been transmitted via a data transmission device configured, for example, as a communication interface, for example, for wireless communication. It may also be a LAN interface through which communication is possible.

[0099] The method carried out by the computer program is explained below using Figure 2.

[0100] In a first step S1, at least one state value is recorded at a first point in time by the protective device 16. This can be done via the triggering protective component 18, the voltage-limiting protective component 20, and / or the measuring and / or sensor unit 22. In any case, it is ensured that the protective device 16 records the state value. In a second step S2, which is optional, the state value is recorded by the protective device 16 at a second point in time that is different from the first point in time.

[0101] In principle, the state value can be recorded at multiple points in time. This means that the state value is recorded several times consecutively, especially periodically, so that a time series / sequence of the state value is obtained.

[0102] In a third step S3, the state value detected at least once by the protective device 16 is transmitted to the evaluation unit 30, in particular the processor unit 34, so that the state value can be evaluated.

[0103] In addition, it can be provided that at least a second variable is recorded, for example a characteristic parameter of the voltage network 12, a value from the database 44, a value from a measuring device 46, 48, or a status value of another protective device 16. The second variable is different from the status value. The second variable can therefore be recorded independently of the voltage network 12, for example, read from the database 44, to which the evaluation unit 30 has access, in particular via the communication interface 32. Data can be obtained from the database 44 that is included in the analysis, for example environmental data such as weather data, time data (working day, weekend, holiday, day and / or night), or usage data from systems in the surrounding area, for example a train timetable or usage data from electrical charging systems in the surrounding area.This allows additional information to be obtained that explains any difference between the state values ​​at different points in time.

[0104] The second variable obtained from the database 44 is transmitted to the processor unit 34 of the evaluation unit 30 so that the processor unit 34 detects the second variable of the voltage network 12.

[0105] However, the second variable can also be a characteristic parameter of the voltage network 12, which is detected by one of the measuring devices 46, 48 or the other protective device 16, wherein the characteristic parameter is transmitted to the processor unit 34 of the evaluation unit 30, so that the processor unit 30 detects the characteristic parameter of the voltage network 12.

[0106] If the state value of the protective device 16 is also a characteristic parameter, two characteristic parameters can be determined which differ from each other.

[0107] In principle, the respective characteristic parameter of the voltage network 12 can be a voltage, a current, a power, a frequency, a distortion, a harmonic, a reactive power and / or an energy value of the voltage network 12, in particular of a phase of the multi-phase voltage network 12.

[0108] As already mentioned above, a time series of the state value can be recorded. This can also apply to the optionally recorded second variable, which is therefore recorded multiple times, especially periodically.

[0109] For example, the second quantity is always recorded when the state value is recorded, so that they are recorded in parallel, in particular at the same time.

[0110] This means in particular that when the state value is recorded at the different points in time, a corresponding data record is also read out from the database 44, which represents the second variable, for example the weather data available at the respective points in time.

[0111] In a fourth step S4, at least the state variable is fed into the processor unit 34, i.e. the artificial intelligence 50.

[0112] In particular, the state variable detected at the first time, the state variable detected at the second time, and the second variable are fed as input variables into the artificial intelligence 50, which processes the input variables jointly, so that, based on the state value detected at the at least two different times, a (future) state of the voltage network 12 and / or a (future) state of the protective device 16 is predicted by the processor unit 34, in particular the artificial intelligence 50. For the prediction, the processor unit 34 has an artificial intelligence 50, which receives at least the state value, in particular the state value detected at the at least two different times and the second variable, as input variables and outputs the future state of the voltage network 12 or the protective device 16 as output variables. The output variable therefore corresponds to the evaluation result of the evaluation unit 30.

[0113] The artificial intelligence 50 may comprise at least one artificial neural network, for example a convolutional neural network (CNN) or an artificial recurrent neural network (RNN), such as a Long Short-Term Memory (LSTM) network or a Gated Recurrent Unit (GRU).

[0114] Accordingly, the artificial intelligence 50 is capable of predicting a future state of the voltage network 12 and / or the protective device 16 based at least on the received time series, i.e., the time sequence of the state value. To this end, the artificial intelligence 50 processes at least the state value, which has been recorded in particular at different times, and optionally the second variable, for example, the value from the database 44, the measured value of the measuring device 46, 48, and / or the value of the other protective device 16.

[0115] The artificial intelligence 50 may have been previously trained using a method in which the artificial intelligence 50 was trained to predict the (future) state of the voltage network 12 and / or the protective device 16 based on the state value, which was recorded, in particular, at least two different points in time. Consequently, the artificial intelligence 50 is a trained artificial intelligence 50.

[0116] In a first training step, a training data set is provided for the artificial intelligence 50, which includes at least the state value at a first point in time, the state value at a second point in time, and an actual state of the voltage network 12 or of the protective device 16 at a third point in time. The third point in time is later than the first point in time and the second point in time, so that it is a future point in time to be forecast based on the first point in time and the second point in time. In addition, a second variable can be included in the training data set, which is different from the state value, so that the training data set includes at least two different variables. The second variable can be a characteristic parameter of the voltage network 12, which is different from the state value.However, the second quantity can also correspond to a measured value from a measuring device 46, 48 or a value from a database 44.

[0117] In particular, the training data set may comprise a time series or time sequence of the state value, so that the state value was measured or recorded at different times.

[0118] The optionally provided second variable can also be included in the training data set as a time series or time sequence. The training data set can therefore include data from at least two different variables for a specific period of time, as well as information on the state of the voltage network 12 or the protective device 16 obtained at a time later than the specific period.

[0119] However, it is also possible that the training data set includes corresponding information of more than just two different sizes, which means that more information or data is made available overall, making the training more comprehensive and the informative value of the correspondingly trained artificial intelligence 50 higher.

[0120] In a second training step, at least the state value acquired at the first time and the state value acquired at the second time, in particular the time series or time sequence of the state value, are fed into the processor unit 34, which has the artificial intelligence 50 to be trained. The processor unit 34, which has the artificial intelligence 50, jointly processes the state value acquired at the different times, in particular the time series or time sequence, and outputs a predicted future state of the voltage network 12 or the protective device 16 at the third time, at which the training data set includes the actual state of the voltage network 12 or the protective device 16.Accordingly, during training, the artificial intelligence 50 learns corresponding relationships between the state value recorded at the first time and the state value recorded at the second time, as well as the effect(s) on the later state of the voltage network 12 or the protective device 16, so that the artificial intelligence 50 is trained to predict the future state of the voltage network 12 or the protective device 16 based on the past or current data.

[0121] In addition, the second variable can optionally be included, whereby correlations between the different variables can be detected during training, i.e. correlations between the state value that was recorded at different times and the second variable that was recorded once or also at different times.

[0122] In a third training step, the predicted future state of the voltage network 12 or the protective device 16 at the third point in time is compared with the actual state of the voltage network 12 or the protective device 16 at the third point in time, the latter being included in the training data set. This comparison determines any deviation between the predicted future state of the voltage network 12 or the protective device 16 and the actual state of the voltage network 12 or the protective device 16. Thus, during training, it is determined how accurate the prediction of the artificial intelligence 50 already is, i.e., how well the prediction matches the actual state.

[0123] In a fourth training step, the detected deviation between the predicted future state of the voltage grid 12 or the protective device 16 and the actual state of the voltage grid 12 or the protective device 16 is fed back into the artificial intelligence 50 to be trained in order to adjust weighting factors of the artificial intelligence 50 to be trained if the deviation lies outside a tolerance range. The tolerance range can be specified and / or set by a user.

[0124] Subsequently, at least the third training step is repeated, with the deviation determined during the comparison in the third training step being progressively reduced. After a certain number of repetitions (iterations), the deviation is so small that it lies within the tolerance range, so that the deviation is no longer fed back. At this point, the artificial intelligence 50 has at least reached a (pre-)trained state for the training dataset, so that it can be used.

[0125] The artificial intelligence 50 can then be further trained with the same training steps with corresponding iterations, so that the artificial intelligence 50 is trained, for example, on further variables, in particular characteristic parameters of the voltage network 12 and / or different pairings of variables.

[0126] In particular, the training of the artificial intelligence 50 may also involve feeding in more than two different variables, for example, up to eight different variables or more. The corresponding training set used for this purpose therefore includes more data that is provided and fed in.

[0127] Typically, the training steps are repeated for several different actual states of the voltage network 12 or the protective device 16 and / or several different data sets of the variables in order to train the artificial intelligence 50. In the final training step, as already described, the weighting factors of the artificial intelligence 50 to be trained are adjusted such that the respective predicted future state of the voltage network 12 or the protective device 16 always lies within the tolerance range.

[0128] The artificial intelligence 50 used in the method for monitoring the voltage network 12 has been trained according to the aforementioned training method, so that it is a trained artificial intelligence 50 which determines or predicts the (future) state of the voltage network 12 or the protection device 16 based on at least the state value detected by the protection device 16.

[0129] The processor unit 34, which includes the trained artificial intelligence 50, accordingly outputs the evaluation result in a fifth step S5, for example, the predicted future state of the voltage network 12 or the protective device 16. The evaluation result can be transmitted to the protective device 16, which receives the data via the communication interface 26 and forwards it to the display 36, so that the evaluation result or a data representation thereof is output, in particular displayed, on the protective device 16 itself. In addition to the evaluation result, a recommended action can also be included.

[0130] As an alternative to the protective device 16, the evaluation result can also be transmitted to the separately designed display device 38, so that the evaluation result and any recommended action are output on the display device 38, in particular on the display 42 of the display device 38.

[0131] Since the evaluation unit 30 is designed separately from the protection device 16, the evaluation unit 30 can be designed as a high-performance computer. In any case, it is possible for the evaluation unit 30 to perform computationally intensive evaluations and then transmit the evaluation results to the protection device 16 itself or to the separately designed display device 38.

Claims

Patent claims 1. Monitoring system for monitoring a voltage network (12), wherein the monitoring system (10) comprises a protective device (16) to be integrated into the voltage network (12), which includes at least one tripping protective component (18), wherein the protective device (16) has at least one communication interface (26) which is configured to transmit a status value of the protective device (16) to an evaluation unit (30) designed separately from the protective device (16), and wherein the evaluation unit (30) is configured to perform an evaluation, so that an outsourced evaluation is provided.

2. Monitoring system according to claim 1, characterized in that the state value of the protective device (16) is a measured value detected by the protective device (16) or a value representative of a tripping state of the triggering protective component (18) of the protective device (16).

3. Monitoring system according to claim 1 or 2, characterized in that the protective device (16) is an overvoltage protection device.

4. Monitoring system according to one of the preceding claims, characterized in that the monitoring system (10) comprises an evaluation unit (30) separately designed from the protection device (16), which has at least one communication interface (32) configured to communicate with the communication interface (26) of the protection device (16), in particular wherein the communication interface (26, 32) is designed as a bidirectional communication interface.

5. Monitoring system according to one of the preceding claims, characterized in that the evaluation unit (30) is configured to receive input data from at least one database (44) or a measuring device (46, 48).

6. Monitoring system according to one of the preceding claims, characterized in that the evaluation unit (30) has at least one processor unit (34) which is configured to process at least the state value obtained from the protection device (16).

7. Monitoring system according to claim 6, characterized in that the processor unit (34) comprises an artificial intelligence (50) which is configured to predict a state of the voltage network (12) and / or a state of the protection device (16) based on the state value obtained from the protection device (16).

8. Monitoring system according to one of the preceding claims, characterized in that the separately designed evaluation unit (30) is configured to determine an evaluation result based on the state value of the protection device (16), in particular an output value and / or a state.

9. Monitoring system according to claim 8, characterized in that the separately designed evaluation unit (30) is configured to transmit the determined evaluation result to the protection device (16) or a display device (38).

10. Monitoring system according to one of the preceding claims, characterized in that the protection device (16) has at least one display (36) which is configured to display an evaluation result of the evaluation unit (30), in particular wherein the display (36) is connected to the communication interface (26) of the protection device (16) in a signal-transmitting manner.

11. Monitoring system according to claim 10, characterized in that there is no direct signal transmission between the at least one triggering protection component (18) and the display (36).

12. Monitoring system according to one of the preceding claims, characterized in that the protective device (16) has a housing (24) in which at least the triggering protective component (18) and the communication interface (26) are at least partially housed.

13. Method for monitoring a voltage network (12) using a monitoring system (10), comprising the following steps: Acquiring a status value by means of a protective device (16) which is integrated into the voltage network (12) to be monitored, Transmitting the recorded status value to an evaluation unit (30) designed separately for the protection device (16), and Processing the transmitted state value by means of the evaluation unit (30) so that a separate evaluation takes place.

14. Method according to claim 13, characterized in that the The evaluation unit (30) determines an evaluation result, wherein the evaluation unit (30) transmits the evaluation result back to the protection device (16) or a display device (38), and wherein the evaluation result is output.

Images