DE102018213757B4 - Method and system for generating a reference characteristic of a network component and a system for determining an operating state of the network component using such a reference characteristic - Google Patents

Abstract

A method for generating a reference characteristic of a network component in an energy supply network, the network component having a known operating state and the reference characteristic (6, 7) characterizing a behavior of the network component in this operating state and enabling the operating state to be determined by means of a PLC node, comprising the steps: Measuring a reaction of the network component on the power supply network by means of a measuring device, the measuring device being designed for frequency- and time-resolved measurement of at least one variable in the energy supply network, generating a primary characteristic from the measured reaction (1) of the network component and generating a reference characteristic (6, 7) by combining the primary characteristic (4, 5) with a reception characteristic (H1 (f), H2 (f)) of the PLC node, the reception characteristic (H1 (f), H2 (f)) influencing a PLC signal (S ) by the PLC node upon receipt g of the PLC signal (S).

Description

The invention relates to a method for generating a reference characteristic of a network component in a power supply network, the network component having a known operating state and the reference characteristic characterizing a behavior of the network component in this operating state and enabling the operating state to be determined by means of a PLC node. The invention also relates to a corresponding system.

In addition, the invention relates to a system for determining an operating state of a network component in a power supply network using a powerline communication system - PLC system - with a plurality of PLC nodes that are distributed over a geographical area.

With Powerline Communication (PLC), energy supply networks are used as a communication medium for the transmission of data. The data to be transmitted are modulated on frequencies in the low megahertz range (often 1 MHz to 30 MHz) and the PLC signals formed in this way are transmitted on the power supply cables. The condition of the power supply cables and the condition of the systems connected to the power supply network influence the transmission properties of the powerline communication. This means that when a PLC signal is transmitted from a transmitter to a receiver, it is influenced by the power supply network and the systems connected to it. This influence usually consists of the interspersion of additional signal components, but can also be expressed in the form of distortions, reflections or frequency-dependent attenuation.

Various systems and methods are known with which the state of an energy supply network and / or of systems connected to the energy supply network can be assessed using PLC technology. Exemplary be on EP 2 608 417 A2 , EP 2 863 551 A1 and EP 2 866 041 A1 referenced. A measurement node that can be used in such systems is in the EP 2 863 553 A1 disclosed. The challenge with such methods is to recognize characteristic changes in the transmission properties and to assign them to the respective states or changes in state of the energy supply network or of systems connected to the energy supply network. The one from the EP 2 863 551 A1 known methods, reference characteristics (also referred to as fingerprints) are used, which represent a characteristic reaction of the energy supply network or a system connected to the energy supply network (collectively referred to as network components below). These reference characteristics are compared with signal components of a PLC signal that were coupled into the PLC signal during transmission. The more similar these signal components are to a reference characteristic, the more likely a network component is in the state represented by the reference characteristic.

1. 1. Der Zustand der Netzkomponente wird bestimmt bzw. eine Zustandsänderung wird detektiert.
2. 2. Die charakteristische Beeinflussung des Zustandes/der Zustandsänderung auf die PLC-Übertragungseigenschaften wird mittels

eines PLC-Knotens aufgenommen und die Referenzcharakteristik erstellt. Diese Referenzcharakteristik wird dann mit Kommunikationseigenschaften anderer PLC-Verbindungen zur Detektion des zugehörigen Zustandes verglichen. Wenn die Referenzcharakteristik möglichst gut auf ein empfangenes PLC-Signal passt, kann auf den zugehörigen Zustand geschlossen werden.The following procedure is used to generate such reference characteristics:

1. 1. The state of the network component is determined or a change in state is detected.
2. 2. The characteristic influencing of the state / change of state on the PLC transmission properties is achieved by means of

of a PLC node and created the reference characteristic. This reference characteristic is then compared with the communication properties of other PLC connections to detect the associated status. If the reference characteristic matches a received PLC signal as well as possible, conclusions can be drawn about the associated status.

The advantage of this type of generation of reference characteristics is that the determined reference characteristics can be used directly when determining the status by a PLC node without the need for complex and computationally intensive transformations of the extracted signal components of the PLC signal. The disadvantage here, however, is that the quality of the reference characteristics is considerably limited by the quality of the measurement at the PLC node.

DE 10 2009 058 877 A1 discloses a method by which electrical components in a power grid can be tested. A test unit sends a test command, which is answered by an electrical component with a test response. The test response can contain characteristic information about the electrical component, for example an identification code, a serial number, nominal voltage, trigger characteristic, power consumption or the like. The test command and the test response can be transmitted via the PLC. It is therefore not possible to evaluate the energy supply network between the test unit and the electrical component.

The post-published EP 3 382 901 A1 refers to the recognition of High frequency interference in a power supply network based on characteristic signatures.

The present invention is based on the object of designing and developing a method and system of the type mentioned in such a way that the most universal possible generation of high-quality reference characteristics and the most reliable possible determination of an operating state of a network component is possible.

• Messen einer Rückwirkung der Netzkomponente auf das Energieversorgungsnetzwerk mittels eines Messgeräts, wobei das Messgerät zum frequenz- und zeitaufgelösten Messen mindestens einer Größe in dem Energieversorgungsnetzwerk ausgebildet ist,
• Erzeugen einer Primärcharakteristik aus der gemessenen Rückwirkung der Netzkomponente und
• Erzeugen einer Referenzcharakteristik durch Kombinieren der Primärcharakteristik mit einer Empfangscharakteristik des PLC-Knotens, wobei die Empfangscharakteristik die Beeinflussung eines PLC-Signals durch den PLC-Knoten bei Empfang des PLC-Signals wiederspiegelt.

According to the invention, the above object is achieved by the features of claim 1. Thereafter, the procedure in question comprises the steps:

• Measuring a reaction of the network component on the energy supply network by means of a measuring device, the measuring device being designed for frequency- and time-resolved measurement of at least one variable in the energy supply network,
• Generation of a primary characteristic from the measured reaction of the network component and
• Generating a reference characteristic by combining the primary characteristic with a reception characteristic of the PLC node, the reception characteristic reflecting the influence on a PLC signal by the PLC node when the PLC signal is received.

With regard to a system for generating reference characteristics, the above object is achieved by the features of claim 9. According to this, the system in question comprises a database system, a query unit and a conversion unit,
wherein the database system comprises at least one database that stores a plurality of primary characteristics and several reception characteristics, each of the primary characteristics representing a characteristic reaction of a network component in a known operating state on a power supply network and each reception characteristic the influence of a PLC signal by a PLC node when the PLC signal is received,
wherein the query unit is designed to load at least one primary characteristic and / or one receiving characteristic from the database system based on a received query and to transfer it to the conversion unit, and
wherein the conversion unit is designed to form a reference characteristic in each case by combining one of the at least one primary characteristic with the receiving characteristic.

• ein erfindungsgemäßes System zum Bereitstellen von Referenzcharakteristika,
• eine Auswerteeinheit, die zum Empfangen und Auswerten eines PLC-Signals aus dem PLC-System ausgebildet ist,
• wobei die Auswerteeinheit zum Extrahieren eines Signalanteils aus dem PLC-Signal, der durch eine Rückwirkung der Netzkomponente auf das Energieversorgungsnetzwerk entstanden und in das PLC-Signal eingekoppelt ist, zum Erzeugen einer Anfrage an die Abfrageeinheit, zum Empfangen mindestens einer Referenzcharakteristik in Reaktion auf die Anfrage an die Abfrageeinheit, zum Vergleichen des Signalanteils mit einer oder mehreren der mindestens einen Referenzcharakteristik und zum Bestimmen des Betriebszustands der Netzkomponente basierend auf der Referenzcharakteristik, die dem extrahierten Signalanteil oder einer aus dem Signalteil gebildeten Charakteristik möglichst ähnlich ist, ausgebildet ist.

With regard to a system for determining an operating state of a network component, the above object is achieved by the features of claim 11. According to this, the system in question comprises:

• a system according to the invention for providing reference characteristics,
• an evaluation unit designed to receive and evaluate a PLC signal from the PLC system,
• The evaluation unit for extracting a signal component from the PLC signal, which is caused by a reaction of the network component on the energy supply network and is coupled into the PLC signal, for generating a request to the interrogation unit, for receiving at least one reference characteristic in response to the request to the interrogation unit, for comparing the signal component with one or more of the at least one reference characteristic and for determining the operating state of the network component based on the reference characteristic which is as similar as possible to the extracted signal component or a characteristic formed from the signal component.

1. 1. Beschränkte Auflösung über die Frequenz: PLC-Knoten ermitteln beispielsweise das SNR (Signal-zu-Rausch-Verhältnis) oder die verwendeten Modulationen trägerscharf, sodass diese Größen lediglich über die Breite eines Trägers und nicht in Bezug auf einzelne Frequenzen oder schmale Frequenzbänder bekannt sind. In einzelnen Fällen entspricht dies einer Messbandbreite von 24 kHz. Die Frequenzbreite eines Trägers definiert dann die Frequenzauflösung der Referenzcharakteristik. Eine verlässliche Nutzung der Referenzcharakteristik für andere PLC-Knoten mit geringerer Trägerbreite scheidet aus.
2. 2. Beschränkte Auflösung der eigentlichen Messgröße: PLC-Knoten erfassen relevante Kenngrößen (beispielsweise SNR oder Modulation) in diskreten Stufen (Quantisierung). Die Stufenhöhe/-anzahl definiert die Genauigkeit der Referenzcharakteristik-Erzeugung.
3. 3. Ungenauigkeit des Messverfahrens im PLC-Knoten: Das eingesetzte Messverfahren im PLC-Knoten ist von der Genauigkeit her mit hochpräziser Messtechnik nicht zu vergleichen, da hier zum Beispiel durch die verringerte Dynamik der Eingangswandler (Analog-zu-Digital-Wandler) stärkere Abstufungen aber auch größeres Rauschen in den Messungen vorkommen werden.
4. 4. Verfälschung durch modemspezifischen Eingangskanal: Da PLC-Knoten primär zur Datenübertragung konstruiert und aufgebaut sind, wird der primäre Fokus auf die Übertragungseigenschaften gelegt. Im Umkehrschluss führt dies dazu, dass bei der direkten Aufnahme von Referenzcharakteristika, modemtypspezifischen Verfälschung mit in das Ergebnis einfließen, die im Nachhinein nicht mehr von der eigentlichen Referenzcharakteristik zu unterschieden sind. Dies führt dazu, dass ein und dieselbe Referenzcharakteristik, die von unterschiedlichen Modems aufgenommen wird, unterschiedlich aussieht, was sich auf die Qualität der Referenzcharakteristik negativ auswirkt.

In the manner according to the invention, it was first recognized that the advantages of reference characteristics that are matched to certain PLC nodes need not be dispensed with if the quality of the generation of the reference characteristic and the usability of the reference characteristics are to be improved. Rather, it is possible to record measured values about the reactions of a network component "neutrally" with a suitable measuring device, then to transform these measured reactions to certain PLC nodes and to use them in an already transformed form when determining the states of network components by means of a PLC node. It has been recognized that PLC signals are always influenced in a similar way when received by a PLC node, so that a reception characteristic can be determined for this PLC node and thus the influence of the received PLC signal by the PLC node can be simulated . These influences include, for example:

1. 1.Limited resolution based on the frequency: PLC nodes determine, for example, the SNR (signal-to-noise ratio) or the modulations used in precise carrier-specific fashion, so that these variables are only known across the width of a carrier and not in relation to individual frequencies or narrow frequency bands are. In individual cases this corresponds to a Measurement bandwidth of 24 kHz. The frequency width of a carrier then defines the frequency resolution of the reference characteristic. Reliable use of the reference characteristic for other PLC nodes with a smaller carrier width is ruled out.
2. 2. Limited resolution of the actual measured variable: PLC nodes record relevant parameters (e.g. SNR or modulation) in discrete steps (quantization). The height / number of steps defines the accuracy of the reference characteristic generation.
3. 3. Inaccuracy of the measurement method in the PLC node: The accuracy of the measurement method used in the PLC node cannot be compared with high-precision measurement technology, because here, for example, due to the reduced dynamics of the input transducers (analog-to-digital converters), stronger gradations but there will also be greater noise in the measurements.
4. 4. Falsification by modem-specific input channel: Since PLC nodes are primarily designed and built for data transmission, the primary focus is on the transmission properties. Conversely, this leads to the fact that when reference characteristics are recorded directly, modem-type-specific falsification is included in the result, which can no longer be differentiated from the actual reference characteristic afterwards. This means that one and the same reference characteristic, which is recorded by different modems, looks different, which has a negative effect on the quality of the reference characteristic.

In order to avoid such influences, which are listed by way of example, from a PLC node, a measuring device is therefore used according to the invention which is designed for frequency and time-resolved measurements. Such a measuring device is matched to the measurement of physical quantities within the energy supply network and can therefore be optimized for measurement tasks. In this way, high input dynamics and / or a higher frequency resolution can be implemented in the measuring device than is possible with a PLC node. Furthermore, the quantization of the measured values can be selected to be smaller, as a result of which the quantization noise can be reduced. Overall, this leads to a more precise recording of the reference characteristics and thus contributes to an improvement in the use of the reference characteristics.

When generating a reference characteristic according to the invention, a reaction of a network component is measured in a first step, the network component being in a known operating state. Such a reaction of a network component on the energy supply network arises from the fact that the network component is supplied with energy from the energy supply network or feeds energy into the energy supply network. The reaction of the network component results essentially from the interconnection and control of the individual components from which the network component is composed. In particular, active components, such as transistors, IGBTs (Insulated-Gate Bipolar Transistor), thyristors, triacs or other controllable semiconductor elements, produce particularly useful signal components. Inverter systems that convert, for example, a direct voltage from a photovoltaic module into an alternating voltage for feeding into the energy supply network, couple signal components into the energy supply network that have a characteristic time and frequency curve. These depend on the structure of the inverter bridge of the inverter, the semiconductor elements used in the inverter bridge, the PWM signal (pulse width modulated signal) used to control the inverter bridge, the wiring of the components, the input voltage applied to the inverter and many other influences. This exemplary enumeration applies correspondingly to many other network components, as a person skilled in the art will recognize. Depending on the structure of the network components, this creates a characteristic reaction on the energy supply network, which acts like a kind of "electrical fingerprint" of the network component in the energy supply network. The reaction is also dependent on the tolerances of the individual components of the network component, but a type-dependent characteristic is retained. In addition, the reaction depends on the operating status of the network component. A faultlessly working inverter will have a different effect on the energy supply network than, for example, an inverter in which a semiconductor element in the inverter bridge has failed. When measuring the reaction of the network component, it is advisable if the measurements are carried out largely independently of other interfering influences. This means that in the part of the energy supply network that is used to supply the network components during the measurement of the reaction, preferably other influences, such as PLC signals, interference from other network components or radio signals that are interspersed, have been completely or at least significantly filtered out. In this way it can be ensured that the measured reaction actually originates from the network component.

By measuring the reaction of the network component, the measuring device generates a multitude of measured values that are used as the measured reaction of the network component when generating a primary characteristic. These measured feedback effects can include a frequency spectrum from a lower limit frequency to an upper limit frequency at a defined point in time. This defined point in time can be a specific point in time during a period of the supply voltage of the energy supply network, such as, for example, a zero crossing or a peak of the supply voltage. The defined point in time can, however, also be defined by other events, for example the start of a particularly characteristic disturbance event or the occurrence of a voltage at a specific frequency. Additionally or alternatively, the measured feedback effects can represent a time curve that is recorded over a predefined period of time. This predefined time span can, for example, be a few or several milliseconds long or - if special characteristics of a network component can be deduced from it - even include several periods of the voltage of the energy supply network.

In a further step, a primary characteristic of the network component is generated from the measured feedback. This primary characteristic reflects a characteristic behavior of the network component in the current operating state of the network component and is at the same time a "neutral" image of the reaction of the network component without any significant influence from the measuring device. In particular, the primary characteristic is not influenced by the reception properties of a PLC node.

wobei y(t) eine Ausgangsgröße im Zeitbereich, h(t) die Übertragungsfunktion im Zeitbereich und x(t) die Eingangsgröße ebenfalls im Zeitbereich ist. Insbesondere wenn die Primärcharakteristik und die Empfangscharakteristik als spektralzerlegte Größe, beispielsweise nach einer Fourier-Transformation vorliegt, vereinfacht sich diese Faltung zu einer Multiplikation: $$\text{Y}\left(\text{f}\right)\text{=H}\left(\text{f}\right)\cdot \text{X}\left(\text{f}\right),$$

wobei Y(f), H(f) und X(f) jeweils die Fouriertransformierten der entsprechend bezeichneten Größe im Zeitbereich darstellt. Hierbei repräsentiert X(f) die Primärcharakteristik, Y(f) die Referenzcharakteristik und H(f) die Empfangscharakteristik.So that the determined characteristic can nevertheless be used directly for a state recognition by means of PLC nodes, the primary characteristic is converted according to the invention into a reference characteristic in a further step. This is done by combining the primary characteristic with a receiving characteristic of a PLC node with which the determination of an operating state is to be carried out later. A reception characteristic reflects the influence on a PLC signal by the PLC node when the PLC signal is received. This reception characteristic represents, for example, frequency quantization in carriers, which are used for the transmission of PLC signals by the PLC node, or amplitude quantization, with which variables such as voltages, SNR or modulation levels are recorded. This reception characteristic can be described by a transfer function H, which usually has the frequency as a variable. A combination of the primary characteristic with this reception characteristic can then include a convolution of the primary characteristic with the transfer function H in the time domain: $$y\left(t\right)=H\left(t\right)*x\left(t\right),$$

where y (t) is an output variable in the time domain, h (t) is the transfer function in the time domain and x (t) is the input variable also in the time domain. In particular, if the primary characteristic and the receiving characteristic are available as a spectrally decomposed quantity, for example after a Fourier transformation, this convolution is simplified to a multiplication: $$\text{Y}\left(\text{f}\right)\text{= H}\left(\text{f}\right)\cdot \text{X}\left(\text{f}\right),$$

where Y (f), H (f) and X (f) each represent the Fourier transform of the correspondingly designated variable in the time domain. Here, X (f) represents the primary characteristic, Y (f) the reference characteristic and H (f) the reception characteristic.

How such a specific reference characteristic can be used can be, for example, the EP 2 863 551 A1 can be removed. In the method and system disclosed there, reference characteristics are used to determine the operating status of a network component. Reference is hereby expressly made to the content of this document.

In a further development of the method according to the invention, primary characteristics of the network component are generated for several known operating states. If necessary, reference characteristics can be generated in addition to the primary characteristics. In both cases, the multiple operating states preferably include at least one normal operating state and at least one fault state of the network component. A normal operating state can include various states that can be referred to as error-free. For example, in the case of an inverter, a normal operating state can be comprised of operation of the inverter without a connected load and another normal operating state of operation of the inverter with a load dimensioned within a nominal range. Faulty operation can exist if, in this example, a diode in a rectifier stage of the inverter is defective, a filter element of the input filter is no longer working correctly or a semiconductor element in the inverter bridge is defective. Depending on the operating state, the network component will then generate a reaction that is characteristic of the operating state, which in turn is reflected in the primary characteristics or the reference characteristic.

In a preferred embodiment of the step of measuring a reaction, at least one variable is measured that results from the charging of the network components with energy from the energy supply network or from feeding energy into the energy supply network. This at least one variable can be formed by the most varied of physical variables which can cause or indicate a characteristic reaction of a network component on the energy supply network. These variables are preferably formed by a voltage and / or a current intensity at one frequency or a plurality of frequencies. Detecting the physical variable at a single frequency can be useful if it is known that the network component at a dedicated frequency has a characteristic effect on the energy supply network. In many cases, however, a frequency range will have to be considered which extends from a lower to an upper limit frequency. It is also conceivable that several non-contiguous frequency ranges are considered.

In a further development, several measurements of the reaction can be carried out in the step of measuring a reaction. These multiple measurements are made at different times and preferably over a predefined frequency range. To ensure that the measured feedback remains comparable, the operating state of the network component should not change over the several measurements. The different times can be selected differently. The multiple points in time can thus have a fixed time interval from one another, for example the length of one or more periods of the supply voltage of the energy supply network. However, since defined relative phase positions to the supply voltage are not always guaranteed when the reference characteristics are used later, the multiple measurements at different times can also be selected and triggered largely as desired.

The specific design of the measuring device with which the reaction of the network component is measured in the step of measuring the reaction is not of decisive importance. However, the measuring device should be able to deliver significantly better resolutions both in the frequency and in the time domain as well as in the amplitude quantization. The measuring device is preferably designed to generate high-resolution measured values of the reaction. The measured values are preferably highly resolved both in terms of their amplitude and in terms of their spectrum. What a high resolution means in a specific case will depend on the size measured in each case. Preferably, however, the measured values are represented with at least 16 bits, particularly preferably with numbers that are at least 32 bits large. With an n-bit number, this results in 2 ⁿ amplitude or frequency levels. The measuring device is very particularly preferably formed by a spectrum analyzer.

The step of measuring feedback is preferably carried out in an energy supply network or a part of an energy supply network, in which feedback from other network components other than the network component to be measured is practically non-existent. This can be achieved by filtering out interference signals received via the energy supply network by means of filters or other protective measures.

When generating the primary characteristic, measured values from several measurements are preferably combined with one another. This combination of measured values is preferably carried out by averaging or determining a range covered by the measured values. The measured values that correspond to one another are preferably always combined with one another. Usually these will be the measured values that come from the same frequency or the same frequency range. In the case of an averaging, all corresponding measured values would be included in an averaging, for example by adding up all of the measured values and dividing the sum by the number of measured values. Instead of this arithmetic mean, other mean value formations would in principle also be conceivable. Since a primary characteristic will usually represent a frequency-dependent characteristic curve, the averaging would be repeated over the entire spectrum of the primary characteristics. The same applies to a range formation in which the lowest measured value of a specific frequency or a frequency range represents a respective lower range limit and the highest measured value of this frequency or this frequency range represents a respective upper range limit.

In a further development, a database can be provided in which the generated primary characteristics are stored together with the respective associated operating status. In addition, information would be helpful that makes it possible to assign the recorded primary characteristic to the associated network component. This information can be, for example, a model number or a serial number. It is also possible that this information refers to a further database in which further information of the network component is stored, for example the manufacturer, the model, the year of production, any special features, etc. The database for the primary characteristics can additionally store reception characteristics that can be used together with the primary characteristics to calculate the reference characteristics. Alternatively, the reception characteristics can also be stored in a further database.

A system according to the invention for generating reference characteristics of a network component which can be used for determining an operating state of the network components is described with the independent claim. This system can be used in particular to carry out a method according to the invention. The system comprises a database system, a query unit and a conversion unit. The database system comprises at least one database which stores a large number of primary characteristics and several reception characteristics. A separate database can be provided for the primary characteristics and the reception characteristics. However, a database can also store both the primary characteristics and the reception characteristics as well as possibly further information.

Each of the primary characteristics represents a characteristic reaction of a network component in a known operating state in response to the energy supply network. This means that several reference characteristics are usually required for a network component in order to map different operating states of the network component. Each reception characteristic reflects the influence of a PLC node on a PLC signal when the PLC signal is received.

The query unit is designed to load at least one primary characteristic and / or one receiving characteristic from the database system based on a received query and to transfer it to the conversion unit. The conversion unit is designed to form a reference characteristic in each case by combining the at least one primary characteristic with the receiving characteristic. In a preferred embodiment, the database system and / or the query unit and / or the conversion unit can be operated on a server or a service system.

In addition, the system can have a measuring device which is designed to measure the effects of a network component in a known operating state. This measuring device generates measured values that represent a measured reaction. A processing unit, which can be an additional component of the system, can generate a primary characteristic from the measured feedback. This primary characteristic can then be stored in the database system. Additional information on the network component and the known operating status could be stored here.

The system according to the invention for generating reference characteristics can be part of a system according to the invention for determining an operating state of a network component in an energy supply network using a PLC system. Such a system is described in a further independent claim. The basics of determining the operating state of a network component based on reference characteristics are already addressed in FIG EP 2 863 551 A1 disclosed, the content of which is hereby expressly referred to again. In addition to the system according to the invention for generating reference characteristics, the system according to the invention for determining an operating state includes an evaluation unit. The evaluation unit is designed to receive and evaluate a PLC signal from the PLC system.

In a preferred mode of operation of this system according to the invention, the evaluation unit extracts a signal component from a received PLC signal that was created by a reaction of the network component on the energy supply network and that was coupled into the PLC signal. In addition, the evaluation unit is designed to generate a query to the query unit. In response to the received request, the query unit loads at least one primary characteristic and, if necessary, a reception characteristic from the database system and transfers the loaded characteristics to the conversion unit. The conversion unit generates at least one reference characteristic from the received characteristics and returns this at least one reference characteristic - possibly via the interrogation unit - back to the evaluation unit. The evaluation unit receives the at least one reference characteristic that has been generated in response to the query. The evaluation unit can store the received at least one reference characteristic in a memory (and hold it permanently or at least for a longer period of time) or use it only once for evaluating the received PLC signal. In a next step of this preferred operation, the evaluation unit compares the extracted signal component with one or more of the at least one reference characteristic that has been generated in response to its own query. Based on this comparison, the reference characteristic is determined which is as similar as possible to the extracted signal component or a characteristic formed from the signal component. Since each reference characteristic corresponds to a known operating state of a network component, the specific Reference characteristic of the operating state of the network component can be determined.

In a preferred development, the evaluation unit is designed to send an initialization request to the query unit during an initialization step. This initialization step can be carried out during commissioning or when switching on a PLC node, which can be part of the evaluation unit. In this initialization step, the evaluation unit can collect all of the essential information that is required during operation of the evaluation unit and that is unlikely to change or will only change at large time intervals. This information, which is collected during the initialization step, can include the required reference characteristics. Since a PLC node will have largely unchangeable reception characteristics and the primary characteristics should only rarely be changed or supplemented, it will be sufficient in practice if this information is provided once or at relatively long time intervals, for example several days, a few weeks or even a few months, loaded and stored in a preferably non-volatile memory.

In response to this initialization query, the query unit can be designed to load a plurality of primary characteristics from the database system and to keep them ready as loaded primary characteristics.

In a further development of this system, the initialization request can include information about the PLC node. This information can be, for example, a type identifier, a serial number, a MAC-ID (Medium Access Control Identifier) and / or some other piece of information that can be uniquely assigned to a device. The query unit can be designed to load or query a reception characteristic based on the information about the PLC node and to hold it ready as a loaded reception characteristic.

The interrogation unit can transfer the loaded primary characteristics and the loaded receiving characteristic to a conversion unit which generates a plurality of reference characteristics by combining the loaded primary characteristics with the loaded receiving characteristic. In this way, reference characteristics are created which are tailored to the PLC node whose identification information was contained in the initialization request and which can easily be used to evaluate received and extracted signal components. These reference characteristics can then be transmitted to the evaluation unit, preferably via the interrogation unit.

In another development of this system, the initialization request does not contain any information about the PLC node. In this development, the evaluation unit can transmit the loaded primary characteristics to the evaluation unit. The evaluation unit then loads or queries a reception characteristic. This can be done, for example, in that the evaluation unit has access to a memory with a reception characteristic. Since the reception characteristics of a PLC node will usually not change, the reception characteristics can be permanently programmed in the PLC node and made available to the evaluation unit. However, the evaluation unit can also send a dedicated query to a database, for example a database of the database system used according to the invention, and query the reception characteristics itself. Regardless of the way in which the reception characteristic is queried, the evaluation unit receives a charged reception characteristic. The evaluation unit transfers the charged primary characteristics and the charged reception characteristic to the conversion unit, which generates several reference characteristics by combining the charged primary characteristics with the charged reception characteristic and returns the several reference characteristics to the evaluation unit.

In a particularly preferred embodiment, the evaluation unit and / or the conversion unit is part of a PLC node or is assigned to it. Each PLC node of the PLC system according to the invention, which is designed to determine an operating state based on reference characteristics, would contain an evaluation unit and / or a conversion unit, or an evaluation unit and / or a conversion unit would be assigned to it.

• 1 ein Diagramm mit einem Messwertverlauf einer Rückwirkung sowie einer mittels PLC-Knoten ermittelter Referenzcharakteristik gemäß Stand der Technik,
• 2 das Diagramm nach 1 mit einigen Ergänzungen zur Verdeutlichung von Ungenauigkeiten bei der Bestimmung von Referenzcharakteristika mittels PLC-Knoten,
• 3 ein Diagramm mit mehreren Messwertverläufen einer Rückwirkung eines Netzknotens,
• 4 ein Diagramm mit einer aus den Messwerten gemäß 3 gebildeten Primärcharakteristik durch Mittelwertbildung (4A) bzw. durch Bereichsbildung (4B),
• 5 Diagramme zur Veranschaulichung der Bildung einer Referenzcharakteristik aus einer Primärcharakteristik unter Verwendung einer ersten Empfangscharakteristik,
• 6 Diagramme zur Veranschaulichung der Bildung einer Referenzcharakteristik aus einer Primärcharakteristika unter Verwendung einer zweiten Empfangscharakteristik,
• 7 ein Blockdiagramm eines ersten Ausführungsbeispiels eines erfindungsgemäßen Systems zum Bestimmen eines Betriebszustands einer Netzkomponente und
• 8 ein Blockdiagramm eines zweiten Ausführungsbeispiels eines erfindungsgemäßen Systems zum Bestimmen eines Betriebszustands einer Netzkomponente.

There are now various possibilities for designing and developing the teaching of the present invention in an advantageous manner. For this purpose, reference is made, on the one hand, to the claims subordinate to the independent claims and, on the other hand, to the following explanation of preferred exemplary embodiments of the invention with reference to the drawing. In connection with the explanation of the preferred exemplary embodiments of the invention with reference to the drawing, generally preferred configurations and developments of the teaching are also explained. Show in the drawing

• 1 a diagram with a measured value curve of a reaction as well as a reference characteristic determined by means of a PLC node according to the state of the art,
• 2 the diagram according to 1 with some additions to clarify Inaccuracies in the determination of reference characteristics using PLC nodes,
• 3 a diagram with several measured value curves of a reaction of a network node,
• 4th a diagram with one of the measured values according to 3 primary characteristic formed by averaging ( 4A) or through area formation ( 4B) ,
• 5 Diagrams to illustrate the formation of a reference characteristic from a primary characteristic using a first reception characteristic,
• 6th Diagrams to illustrate the formation of a reference characteristic from a primary characteristic using a second reception characteristic,
• 7th a block diagram of a first embodiment of a system according to the invention for determining an operating state of a network component and
• 8th a block diagram of a second embodiment of a system according to the invention for determining an operating state of a network component.

Based on 1 and 2 the inaccuracies in the determination of reference characteristics with a PLC node should be clarified again. 1 shows a diagram with a measured value curve 1 a reaction of a network component as well as a reference characteristic determined by means of a PLC node 2 . The frequency f is plotted on the abscissa. In the example shown, let the measured value curve 1 formed by a voltage that can be extracted from a PLC signal and does not result from the PLC signal or the supply voltage of the power supply network. Correspondingly, a voltage U is plotted on the ordinate of the diagram. For a fundamental understanding of the characteristic curves, it is irrelevant in which frequency or voltage range the curves actually lie. Therefore, in the 1 and 2 no zero point and no scaling shown. The course of the measured values 1 is shown as a dashed line, the reference characteristic 2 with a solid line.

In 2 A rectangular grid is also drawn in, which illustrates the quantization levels of the PLC node. In the abscissa direction, the quantization is defined by the width Δf of the carriers of the PLC signal. Since the electronics of the PLC node only evaluate the PLC signals received at its input in a carrier-specific manner, each PLC signal is resolved in the frequency direction to a maximum of the width Δf of a carrier. In the ordinate direction, the quantization is defined by the size of the amplitude steps ΔU, which can be resolved on the input side by the PLC node. Arrows 3 illustrate how far the actual values of the retroactive effect can deviate from the values determined for the reference characteristic. This shows that there is potential for improvement in the determination of reference characteristics compared to this state of the art.

Based on 3 until 6th it should be made clear how this has happened in the method according to the invention. In a first step, the effects of a network component in a known operating state are measured. For this purpose, a high-resolution measuring device is used that is designed for frequency and time-resolved measurements and allows a "neutral" measurement of the feedback over the frequency. In the present case, this measuring device is formed by a spectrum analyzer. In 3 Several curves are drawn in that have been measured at different times as repercussions of the network components in the known operating state. The family of curves results in particular from the fact that individual parameters of the network component can change despite the unchanged operating state. Furthermore, noise, for example quantization noise, amplification noise or interference on the energy supply network, can cause the measured values to deviate from a previous measurement.

When generating a primary characteristic, these deviations must be taken into account accordingly. Based on 4th should this be examined in more detail. 4A shows an embodiment in which the primary characteristic is generated by averaging the various measured values of several measurements. An arithmetic mean is used. The resulting primary characteristic is given a reference symbol 4th marked. In 4B an area is formed in which a band is determined within which the measured values of the individual measurements are located. The resulting primary characteristic is given a reference symbol 5 shown.

Both 5 and 6th it is assumed in each case that the primary characteristic has been generated by averaging. On the left of the 5 and 6th primary characteristics are therefore shown as a relatively thin line. These primary characteristics 5 , 6th are then multiplied by a transfer function H ₁ (f) or H ₂ (f). These transfer functions H ₁ (f), H ₂ (f) each represent a receiving characteristic of a PLC node of different types, which in the present case are referred to as “type 1” and “type 2” for the sake of simplicity. When multiplying a primary characteristic by a Transfer function, the primary characteristic is combined with a reception characteristic. This results in a reference characteristic in each case 6th or. 7th . It can be seen that different reference characteristics arise here, even if the primary characteristic, which is assumed in each case, is identical in both cases. In the case of the 5 are the carriers of the PLC node type 1 relatively wide and the amplitude levels relatively coarse. This creates a strongly quantized reference characteristic. In the case of the 6th are the carriers of the PLC node type 2 significantly narrower and the amplitude levels smaller. This creates a significantly less quantized reference characteristic. This makes it clear that the method according to the invention can produce a reference characteristic individually tailored to a PLC node by forming a primary characteristic from high-resolution measured values and by combining this primary characteristic with a receiving characteristic.

7th shows a first embodiment of a system according to the invention for determining an operating state of a network component. The system 8th consists of an evaluation unit 9 , a memory 10 , an interrogation unit 11th , a conversion unit 12th as well as a database system 13th that is made up of three individual databases 14th , 15th , 16 consists. The evaluation unit 9 and the memory 10 are implemented in a PLC node, while the other components shown here run on a server or a server system. The evaluation unit sends in an initialization step 9 an initialization request to the interrogation unit 11th . The initialization request contains identification information for the PLC node, which it sends to the interrogation unit 11th allows a reception characteristic of the PLC node to be identified. In response to the initialization request, the interrogation unit loads 11th various information from the database system 13th . The interrogation unit uses this for this 11th first the identification information and accesses a database 14th with detailed information on the PLC nodes. From the database 14th such information is then loaded as it is for the interrogation unit 11th allow correct reception characteristics to be loaded. In a further step, the interrogation unit loads 11th different primary characteristics from a database 15th of the database system 13th and a reception characteristic from a database 16 of the database system 13th . The loaded primary characteristics and the loaded reception characteristics are then used by the interrogation unit 11th the conversion unit 12th passed, which forms reference characteristics therefrom, and to the interrogation unit 11th returns. The query unit 11th then sends the reference characteristics to the evaluation unit 9 back which the obtained reference characteristics in memory 10 filed.

The evaluation unit 9 is designed to receive and evaluate PLC signals S. For this purpose, the evaluation unit extracts 9 the signal components from the PLC signal that are caused by a reaction of a network component on the energy supply network and that are coupled into the PLC signal. The signal components extracted in this way are then compared with reference characteristics and the reference characteristic is determined which is as similar as possible to the extracted signal components. Since each reference characteristic is representative of a defined operating state of a network component, the operating state and the network component can be determined from it.

In particular, if there are a large number of reference characteristics, it can be useful if not all of the reference characteristics in question are loaded in advance. Rather, the evaluation unit can be designed to query only a part of relevant reference characteristics. For this purpose it is conceivable that the evaluation unit 9 Determines parameters from the extracted signal components and specifically requests reference characteristics with these parameters. Suitable parameters can, for example, be particularly pronounced signal peaks or contained frequency components. In this embodiment, the interrogation unit 11th then additionally able to determine matching reference characteristics for the parameters and only send these to the evaluation unit 9 to send back.

A second embodiment of a system according to the invention for determining an operating state that is shown in 8th is shown is in its function relatively similar to the first embodiment. However, more components are implemented in the PLC node. So is next to the evaluation unit 9 and the memory 10 nor the conversion unit 12th arranged in PLC nodes. In addition, the system of the second embodiment includes a memory 17th , in which a reception characteristic is stored. Even if databases 14th , 16 for the PLC node data and for reception characteristics would not be required in this embodiment, they are still part of the database system 13th shown. In this exemplary embodiment, the evaluation unit can send an (initialization) request to the interrogation unit without any identification information from the PLC node 11th send. The query unit 11th can interpret this in such a way that only primary characteristics should be returned. In response to the request, the interrogation unit asks 11th the (relevant) primary characteristics and sends the obtained primary characteristics to the evaluation unit 9 back. The evaluation unit 9 transfers the received primary characteristics to the conversion unit 12th that run out of memory 17th loads a reception characteristic and combines each with the obtained primary characteristics. The resulting reference characteristics are sent to the evaluation unit 9 returned this to memory 10 drops.

1. 1. Schritt: Hochauflösende Erfassung Mittels bewährter Messtechnik (z.B. Spektrum Analyzer) wird die relevante Messgröße (z.B. leitungsgeführte Störung) hochauflösend und quasi neutral erfasst. Da Zustände in Abhängigkeit ihrer Ausprägung leicht unterschiedliche Messergebnisse liefern, sieht das neue Verfahren vor, dass mehrere unterschiedliche Ausprägungen eines Zustandes (z.B. unterschiedliche ausgeprägte Schirmfehler an einem Mittelspannungskabel oder unterschiedliche Anlagen mit gleichem Fehlertyp) aufgenommen werden.
2. 2. Schritt: Erzeugung eines Primärabdrucks In einem zweiten Schritt wird aus den Einzelaufnahmen eine Primärcharakteristik erzeugt. Dabei bieten sich unterschiedliche Ansätze zur Aggregation der Einzelaufnahmen an. Unter anderem können dies sein: Mittelwertbildung, Bereichsbildung zwischen Minimum und Maximum, ...
3. 3. Schritt: Transformation in eine gerätespezifische Referenzcharakteristik In einem dritten Schritt erfolgt nun die Transformation der Primärcharakteristik in eine gerätespezifische Referenzcharakteristik. Im Transformationsprozess wird die Primärcharakteristik mit den erfassbaren Eigenschaften des jeweiligen Modemtyps verrechnet. Diese Eigenschaften H(f) - hier als Empfangscharakteristik bezeichnet - sind für jeden Gerätetyp definierbar. Hierbei werden die horizontale Auflösung der Primärcharakteristik an die Auflösung des Modems angepasst (Frequenzschrittweite). Zusätzlich kann die Primärcharakteristik mit anderen modemspezifische Charakteristika, wie Eingangsbeschaltung (Signalfilterung und Schutzbeschaltung), die bei der Übertragung einen bekannten Bestandteil des kompletten Übertragungskanals bilden, und Auflösung des Analog-Digital-Wandlers, der die Dynamik und das systemimmanente Quantisierungsrauschen bedingt, angepasst werden, um den Fingerprint, den das Modem misst, zu erhalten. Diese Rechenoperation lässt sich, ausgehend von der Primärcharakteristik, für beliebige Modemtypen (bzw. andere Messgeräte) durchführen, was zu einer einfachen Umsetzung und verbesserten Datenhaltung von Referenzcharakteristika führt. Beim Einsatz einer neuen Modemtechnologie kann so auf bestehende Daten zurückgegriffen werden, um die Messung der Fingerprints ab dem ersten Tag zu ermöglichen. Die rechnerische Anpassung der Primärcharakteristika kann so universell und über Herstellergrenzen hinweg zum Einsatz kommen.

Overall, the generation of the reference characteristics can be divided into three steps:

1. 1st step: High-resolution acquisition Using tried and tested measurement technology (e.g. spectrum analyzer), the relevant measured variable (e.g. conducted interference) is recorded in a high-resolution and quasi-neutral manner. Since states provide slightly different measurement results depending on their characteristics, the new method provides that several different characteristics of a state (e.g. different pronounced shielding faults on a medium-voltage cable or different systems with the same fault type) are recorded.
2. 2nd step: Generating a primary impression In a second step, a primary characteristic is generated from the individual images. There are different approaches to aggregating the individual images. Among other things, this can be: averaging, area formation between minimum and maximum, ...
3. 3rd step: Transformation into a device-specific reference characteristic In a third step, the primary characteristic is transformed into a device-specific reference characteristic. In the transformation process, the primary characteristic is offset against the detectable properties of the respective modem type. These properties H (f) - referred to here as reception characteristics - can be defined for each device type. The horizontal resolution of the primary characteristic is adapted to the resolution of the modem (frequency step size). In addition, the primary characteristic can be adapted with other modem-specific characteristics, such as input circuitry (signal filtering and protective circuitry), which form a known component of the complete transmission channel during transmission, and the resolution of the analog-digital converter, which determines the dynamics and the system-inherent quantization noise. to get the fingerprint that the modem is measuring. Starting from the primary characteristic, this arithmetic operation can be carried out for any type of modem (or other measuring devices), which leads to a simple implementation and improved data management of reference characteristics. When using a new modem technology, existing data can be used to enable the measurement of fingerprints from day one. The mathematical adjustment of the primary characteristics can thus be used universally and across manufacturer boundaries.

The transformation process is able to generate modem-type-specific reference characteristics from primary characteristics that have been generated, so that the status detection can be used regardless of the modem type or the built-in PLC chip.

With regard to further advantageous embodiments of the method according to the invention and the systems according to the invention, reference is made to the general part of the description and to the appended claims in order to avoid repetition.

Finally, it should be expressly pointed out that the exemplary embodiments described above only serve to explain the teaching claimed, but do not restrict them to the exemplary embodiments.

List of reference symbols

1

Measured value curve

2

Reference characteristic

3

Deviations

4th

Primary characteristic (mean)

5

Primary characteristic (area)

6th

Reference characteristic

7th

Reference characteristic

8th

system

9

Evaluation unit

10

Storage

11th

Interrogation unit

12th

Conversion unit

13th

Database system

14th

Database (for PLC node data)

15th

Database (for primary characteristics)

16

Database (for reception characteristics)

17th

Storage

Claims (17)

Method for generating a reference characteristic of a network component in an energy supply network, the network component having a known operating state and the reference characteristic (6, 7) showing a behavior of the Characterized network components in this operating state and enables the operating state to be determined by means of a PLC node, comprising the steps of: measuring a reaction of the network component on the energy supply network by means of a measuring device, the measuring device being designed for frequency- and time-resolved measurement of at least one variable in the energy supply network , Generating a primary characteristic from the measured feedback (1) of the network component and generating a reference characteristic (6, 7) by combining the primary characteristic (4, 5) with a receiving characteristic (H ₁ (f), H ₂ (f)) of the PLC Node, the reception characteristics (H ₁ (f), H ₂ (f)) reflecting the influence of a PLC signal (S) by the PLC node when the PLC signal (S) is received. Procedure according to Claim 1 , characterized in that the primary characteristics (4, 5) and / or the reference characteristics (6, 7) of the network component are generated for several different known operating states, the different operating states preferably including at least one normal operating state and at least one fault state of the network component. Procedure according to Claim 1 or 2 , characterized in that during the step of measuring a reaction at least one variable is measured which results from charging the network component with energy from the energy supply network or feeding energy into the energy supply network, the at least variable preferably being a voltage and / or a Includes current strength at one frequency or multiple frequencies. Method according to one of the Claims 1 until 3 , characterized in that, in the step of measuring a reaction, several measurements are made at different times and preferably over a predefined frequency range. Method according to one of the Claims 1 until 4th , characterized in that in the step of measuring a reaction, high-resolution measured values of the reaction are generated, the measured values preferably being highly resolved both in their amplitude and in their spectrum. Method according to one of the Claims 1 until 5 , characterized in that the step of measuring a reaction is carried out in an energy supply network in which there are practically no repercussions from other network components unlike the network component to be measured. Method according to one of the Claims 1 until 6th , characterized in that, in the step of generating the primary characteristic (4, 5), measured values from several measurements are combined with one another, for which purpose a mean value is preferably formed or a signal range covered by the measurements is determined. Method according to one of the Claims 1 until 7th , characterized in that a database (15) with generated primary characteristics is provided and / or reception characteristics are stored in the database or in a further database (16). System for generating reference characteristics of a network component for determining an operating state of the network component, in particular for carrying out a method according to one of the Claims 1 until 8th , comprising a database system (13), a query unit (11) and a conversion unit (12), the database system (13) comprising at least one database (14, 15, 16) which has a plurality of primary characteristics (4, 5) and several Receiving characteristics (H ₁ (f), H ₂ (f)) stores, each of the primary characteristics (4, 5) representing a characteristic reaction of a network component in a known operating state on a power supply network, and each receiving characteristic (H ₁ (f), H ₂ (f)) reflects the influencing of a PLC signal by a PLC node upon receipt of the PLC signal, the query unit (11) being designed to, based on a received request, at least one primary characteristic (4, 5) and / or to load a reception characteristic (H ₁ (f), H ₂ (f) from the database system (13) and to transfer it to the conversion unit (12), and the conversion unit (12) is designed to each by Combination of one of the at least one primary characteristic (4, 5) with the receiving characteristic (H ₁ (f), H ₂ (f)) to form a reference characteristic (6, 7). System according to Claim 9 , characterized by a server or a server system, the database system (13) and / or the query unit (11) and / or the conversion unit (12) being operated on the server / server system. System for determining an operating state of a network component in a power supply network using a powerline communication system - PLC system - with a plurality of PLC nodes that are distributed over a geographical area, comprising: a system for providing reference characteristics according to Claim 9 or 10 , an evaluation unit (9) which is designed to receive and evaluate a PLC signal (S) from the PLC system, the evaluation unit (9) for extracting a signal component from the PLC signal (S) which is caused by a reaction of the Network components have arisen on the energy supply network and are coupled into the PLC signal (S), for generating a request to the interrogation unit (11), for receiving at least one reference characteristic (6, 7) in response to the request to the interrogation unit (11), designed to compare the signal component with one or more of the at least one reference characteristic (6, 7) and to determine the operating state of the network component based on the reference characteristic (6, 7) which is as similar as possible to the extracted signal component or a characteristic formed from the signal component is. System according to Claim 11 , characterized in that the evaluation unit (9) is designed to send an initialization request to the query unit (11) during an initialization step. System according to Claim 12 , characterized in that the interrogation unit (11) is designed to load a plurality of primary characteristics (4, 5) in response to the initialization request and to keep them ready as loaded primary characteristics. System according to Claim 12 or 13th , characterized in that the initialization request includes information about the PLC node and that the query unit (11) is designed to assign a reception characteristic (H ₁ (f), H ₂ (f)) based on the information about the PLC node load or query and keep it ready as a charged reception characteristic. System according to Claim 13 and 14th , characterized in that the interrogation unit (11) transfers the loaded primary characteristics and the loaded reception characteristic to a conversion unit (12) which generates several reference characteristics (6, 7) by combining the loaded primary characteristics with the loaded reception characteristic and the several reference characteristics (6 , 7) preferably via the interrogation unit (11) to the evaluation unit (9). System according to Claim 13 , characterized in that the query unit (11) transmits the loaded primary characteristics to the evaluation unit, that the evaluation unit (9 _{) loads or queries a reception characteristic (H 1} (f), H ₂ (f)) and receives the loaded reception characteristic that the Evaluation unit (9) transfers the charged primary characteristics and the charged reception characteristics to the conversion unit (12), which by combining the charged primary characteristics with the charged reception characteristics generates several reference characteristics (6, 7) and the several reference characteristics (6, 7) to the evaluation unit (9) returns. System according to one of the Claims 11 until 16 , characterized in that the evaluation unit (9) and / or the conversion unit (12) is part of a PLC node or is assigned to it.

Images