DE112004002642T5 - Plausibility check of a three-phase electrical system - Google Patents

Abstract

Method for verifying the operation of a three-phase electrical system (1) for plausibility, wherein the system (1) comprises a power converter (11) and a load (13), in particular a drive motor of a railway towing vehicle, and wherein
- The load (13) via an AC line with three phases (5a, 5b, 5c) is connected to the power converter (11),
The load (13) is symmetrical to the three phases (5a, 5b, 5c),
Measuring at least two phase currents (5a, 5b) of the three phases (5a, 5b, 5c),
- the power converter (11) is controlled by a control device (4),
The control device (4) uses values of the measured phase currents for operating the system (1),
The control device (4) repeatedly generates identical or similar sequences of control signals for controlling the power converter (11),
wherein absolute values are calculated for operating times corresponding to the sequences, the absolute values being related to a characteristic of a quantity, the size being a ...

Description

The The present invention relates to a method for checking a Operation of a three-phase electrical system for plausibility. The system comprises a power converter and a consumer, the consumer having a AC line with three phases connected to the converter is. The consumer is symmetrical to the three phases and the Power converter is controlled by a control device, the measured phase current values the AC line used to operate the system. Farther The present invention relates to a corresponding arrangement. The power converter can, for example, via a DC link be connected to a second power converter inverter. In particular, the present invention relates to the field of electrical High performance applications such as the delivery of electrical Energy to a drive motor of a railway towing vehicle.

at many applications a power converter has a DC side and an AC side with three alternating current phases. The DC side is with a DC link connected. The electrical consumer is connected to the three alternating current phases on the AC side. The operation of the power converter is controlled by a control device controlled, the phase currents of the three alternating current phases used as input variables for the control method become. For example, it is possible to use a control device, the PWM signals (pulse width modulation) can generate. PWM signals corresponding switching signals are used for Power converter transmitted and cause the switching of electronic valves of the power converter. For controlling the power converter can however, other procedures are performed become.

During the Operation of the system can be several Errors and malfunctions occur: one or more of the three Phases of AC connection could be interrupted or the power converter itself could fail. One or more of the three phases could be connected to ground. Furthermore could a sensor used to detect a phase current of one of the phases is defective be and / or a power supply to operate the sensor could fail. A signal connection from the detector to the control device could be interrupted or (possibly due to a loose connector) may not be realized. Additionally or as an alternative could any device (such as an amplifier) used for Processing the signal used by the detector will be faulty or not be set correctly.

If when measuring the phase currents a malfunction occurs (including a malfunction of the Processing of the measuring signal), the controller will use a wrong input value and As a result, the controller will attempt to measure the phase current adapt. In other words, the control device becomes control signals to the converter to compensate for the wrong input value. If the measured phase current is permanently too low (e.g., zero due to a detection failure), the power converter may be damaged or even destroyed become. Furthermore, if the consumer has a machine (e.g. asynchronous machine), mechanical vibrations are in the range of the machine, which can damage the system.

at In many applications, the electrical load is among the three AC phases symmetrical. In other words, the consumer is balanced. Of the Expression "the consumer is symmetric to the three AC phases "means that the RMS values - root mean square) of the electric currents conducted by the three phases at normal Operation are the same. Examples are a star connected consumer and a delta connected consumer. In the latter case, the consumer includes the same resistance and the same inductance between every phase pair. It could but there is a small deviation from the symmetrical state for example, in the case of a delta-connected consumer. These Deviation is acceptable if a specified tolerance limit is not exceeded becomes. In the case of a star connected consumer, the sum is of the three phase currents the converter output is zero if there is no error such as a ground fault.

If the sum of the three phase currents is zero it should be possible (for example star connected consumer), it is possible all three phase currents to measure the sum of the three phase currents and the sum to compare with a threshold. When the threshold is exceeded can be concluded that there is a mistake and corresponding Actions can be taken such as turning off the power converter and thereby stopping the operation of the electromagnetic machine.

Furthermore, one of the three current sensors and other corresponding devices for processing the sensor signal have been omitted in order to save costs and space. As a result, the above-described method can not be applied. One way to verify the operation of the sensor signal processing in such an arrangement is to use an additional test winding for each sensor and Generating a signal current in the signal line by induction. The signal stream can then be processed and the function of the corresponding processing devices can be checked.

It An object of the present invention is a method for checking a To provide a three-phase electrical system for plausibility the one reliable Detection of errors during the operation of the system and that's fast enough to spot the mistake before the system or parts of the system are damaged become. Furthermore, it is desirable that the Effort for the plausibility check low is and the plausibility check is not the normal operation of the system interferes. Another object of the invention is to provide a corresponding arrangement.

To the Explain An important aspect of the invention becomes a rotating vector or pointer defined, wherein the pointer represents an operating state of the three-phase system. The pointer may, in particular, be a current-space pointer of the consumer current be. It can be used to calculate other quantities that use the Operating state describe, for example, a torque of an electromagnetic Machine that is part of the consumer or the consumer is. The pointer rotates at a frequency equal to the frequency of the guided by the phases AC is. The length of the pointer (i.e., the root of the square of the pointer) may become while change the normal operation, for example as a result of a change of the consumer or a change in demand. A change in length contains but also information about Input values (such as measured values of the phase currents of the system), of which the pointer depends is. As a result, absolute values relative to pointer length stand (e.g., the length, the length square and derived quantities such as the torque of the electromagnetic machine), from the measured Live Streaming of the three-phase system dependent.

It is a basic idea of the invention to evaluate measurement information, the in proportion stand to this pointer or in other words the properties of the Describe pointer.

During the Operation of a power converter can be any, the operating state of the Three-phase system descriptive size oscillations and / or fluctuations subject. So the pointer length can change in a way that it makes it difficult, useful To get information. To overcome this problem, it is proposed certain operating conditions and / or times to select where the fluctuations are less significant.

Different said the following procedure can be performed: An error in Measuring phase currents of the system is evaluated by evaluating at least one absolute value recognized, the pointer to the ratio stands, wherein the pointer rotates at a frequency that coincides with the Frequency of the guided by the phases AC is identical. A control device generates repeatedly identical or similar Consequences of control signals for controlling the power converter. Based on the timing of these sequences becomes the time of determining (e.g., measuring) of the at least one absolute value selected or corresponding information will be for later evaluation secured. That the time corresponds to a rotation position pointer.

• – der Verbraucher über eine Wechselstromleitung mit drei Phasen mit dem Stromrichter verbunden ist,
• – der Verbraucher zu den drei Phasen symmetrisch ist,
• – von mindestens zwei der drei Phasen geführte Phasenströme gemessen werden,
• – der Stromrichter durch eine Steuervorrichtung gesteuert wird,
• – die Steuervorrichtung Werte der gemessenen Phasenströme zum Betreiben des Systems benutzt,
• – die Steuervorrichtung wiederholt identische oder ähnliche Folgen von Steuersignalen zum Steuern des Stromrichters erzeugt,

wobei Absolutwerte für Betriebszeiten berechnet werden, die den Folgen entsprechen, wobei die Absolutwerte mit einem Merkmal einer Größe im Verhältnis stehen, wobei die Größe einen Betriebszustand des Systems beschreibt, wobei die Größe ein rotierender Zeiger ist, der mit einer Frequenz rotiert, die einer Frequenz der von den Phasen geführten Phasenströme gleich ist, wobei eine Länge des rotierenden Zeigers von den Phasenströmen abhängig ist, wobei die Absolutwerte von der Länge des rotierenden Zeigers abhängig sind, und wobei mindestens einer der Absolutwerte zum Detektieren eines Fehlers beim Messen mindestens eines der Phasenströme benutzt wird.In particular, the following is proposed: a method for checking the operation of a three-phase electrical system for plausibility, the system comprising a power converter and a load, in particular a drive motor of a railway towing vehicle, and wherein:

• The consumer is connected to the converter via an AC line with three phases,
• - the consumer is symmetrical to the three phases,
• Measuring phase currents conducted by at least two of the three phases,
• The converter is controlled by a control device,
• The controller uses values of the measured phase currents to operate the system,
• The control device repeatedly generates identical or similar sequences of control signals for controlling the power converter,

wherein absolute values are calculated for operating times corresponding to the sequences, wherein the absolute values are related to a feature of a magnitude, the magnitude describing an operating condition of the system, the magnitude being a rotating pointer rotating at a frequency corresponding to a frequency is equal to the phase currents carried by the phases, wherein a length of the rotating pointer depends on the phase currents, the absolute values being dependent on the length of the rotating pointer, and at least one of the absolute values for detecting an error in measuring at least one of the phase currents becomes.

In particular, the power converter can be an inverter with electronic valves, wherein the power converter is connected to a motor via the three phases. The electronic valves are repeatedly turned on and off by the control device to generate the alternating current, and thereby a rotating magnetic field is he witnesses that causes the rotation of the rotor of the motor. Desired or predetermined figures of the magnetic flux of the motor as a function of the angle of rotation can be realized by outputting corresponding control signals to the inverter. For example, the desired flow figure may be circular or a polygon with six or eighteen corners. However, the flow figures obtained in practice differ from the desired flow figures due to the switching of the electronic valves. In 8th and 11 In the accompanying drawings, which will be described in more detail later, examples of flow figures are shown which can be realized in practice. Especially at the corners of the river figures, ie at certain rotational positions of the magnetic field generated by the stator identical or similar sequences of the control signals are generated in each rotation cycle. For example, one particular valve will be turned on and off during the sequence while the other valves remain off.

In the practice can the switching operations corresponding values in a data group for an entire Rotation cycle or part of a rotation cycle are stored. The entire control process may be software and / or hardware carried out become. In this case, the calculation of the absolute values, the of the length of the rotating pointer (e.g., the current space hand), in particular triggered by the consequences become. As an alternative or in addition can the absolute values are continuously calculated and certain values, the same or similar Follow or corresponding rotational positions of the magnetic field can correspond be evaluated and / or secured. For example, if the rotating pointer is a current space pointer, readings of at least two measured phase currents secured and there will be an evaluation process to check the plausibility regularly like for example, once at the end or after the end of each rotation cycle carried out of the magnetic field.

As already mentioned For example, the rotating pointer may be a current space pointer. It can however, other pointers are defined that have the desired information about contain the behavior of the system, in particular pointers to the Current space pointers are equivalent or using the same be calculated.

Especially For example, the absolute values are preferably measured using the measured values phase currents calculated when the rotating pointer is a Stromraumzeiger, the Phase shifts and amplitudes of the three phase currents corresponds.

Specifically, the absolute value (eg, the torque of the electromagnetic machine) by the controller using the measured phase currents can be calculated when the absolute value of the rotating pointer and the magnetic flux of an electromagnetic machine is calculated (eg, by calculating a product of the pointer and the flow). For example, the controller may model the performance of the consumer using a software implemented model. The measured phase currents, other input variables (such as an electrical voltage and the rotational speed of the machine) and parameters of the machine are used by the model. For example, in DE 195 31 771 A1 (Inventor: Depenbrock) describes a method and an apparatus for determining a rotational speed of a rotating field machine. It is disclosed in the document that signal processing comprises a complete machine model with a power converter whose AC side is connected to the machine. In particular, such a model can be used to calculate the rotating pointer and / or the absolute values.

As already mentioned There are different possible reasons for one Failure or malfunction of the phase current measurement. If, for example a malfunction of a for When the current measurement used power supply occurs, this usually becomes one Immediate reaction by observing the current space pointer can be detected. In another case, the properties may be a shunt resistor (measuring a phase current to convert a current signal is used in a voltage signal) gradually change with time. In this case could the control device that uses the measured phase currents, try to compensate for the corresponding measurement error. It but it is still possible the measurement error to recognize, especially if as already mentioned a model of the control device is used.

Of the Absolute value is preferably a square value that is the square of the Length of the represents rotating pointer. An advantage of using a square value lies in the fact that a Length of one Pointer easily by using squared values of components of the pointer can be calculated, for example, from measured Live Streaming in the case of a current space pointer. It saves measurement time and reduces costs yourself.

It is further preferred that the absolute values be calculated for a plurality of rotational positions of the pointer, each of which corresponds to one of the same or similar sequences. A difference between the absolute values of at least two of the rotational positions or deviations of at least two of the rotational positions and ei A comparison value is used to determine a fault type when measuring at least one of the phase currents. In particular, this method can be used to distinguish amplification and offset errors. These two different types of errors show individual characteristic behavior with respect to variations in pointer length during rotation, which will be described in more detail later. It is therefore possible to detect whether the measurement error is a gain error and / or an offset error.

One Way to detect an error while measuring at least one of the phase currents consists in the implementation of a comparison: the absolute value comes with a previous one Absolute value compared at a same rotational position of the pointer, or is one of a plurality of preceding absolute values derived comparative value. The comparison result will be used to detect the error.

If the plausibility check is carried out It can be a special procedure for deciding whether a failure and / or there is a malfunction. This particular Method is referred to in the present specification as "fuzzy evaluation". It leads to special reliable Decision results.

If the plausibility check is carried out It is proposed to repeatedly provide a first plausibility value determine, wherein the first plausibility value is an instant measure of a Degree of plausibility and / or implausibility is. Furthermore, a large number of the first plausibility values will be used a second plausibility value each of the first plausibility values corresponding to the second plausibility value his degree of plausibility and / or implausibility can influence. An appropriate action is taken when the second plausibility value fulfills a predetermined criterion.

at the fuzzy evaluation reflects the second plausibility value plausibility a plurality of first plausibility values. It is therefore less likely that one incorrect first plausibility value For example, results in an interruption of the converter operation. The effect of fuzzy evaluation can be compared to the effect of a smart filter. Furthermore, she lets herself easy to implement in hardware and / or software, for example by using a counter, where the counter value the second plausibility value is. The term "counter" is not on the Counting from limited to integers. In the example, a first plausibility value results with a lesser degree of plausibility a greater increase in the number of counters as a first plausibility value with a higher one Degree of plausibility. Conversely, the counter number be reduced if the plausibility level of the first plausibility value is high.

It There are several features that are independent of each other in this basic approach or in combination of at least some of the features can. First, the plausibility level the first plausibility value are weighted and the second plausibility value is determined according to the weighted first plausibility value affected. In the simplest case, a weighting factor of 1 can be used for all plausibility levels the first plausibility value to be used. However, it is also possible the weight of the first Plausibility values accordingly an increase in the degree of plausibility to reduce.

Secondly Is it possible, to decide whether the plausibility level of the first plausibility value one lower threshold and / or less than the threshold is. If this is the case, the second plausibility value is affected to prevent the reduced plausibility show. If this is not the case, the second plausibility value does not become amended or modified, to an increased total plausibility to show (e.g., the counter number ) Is reduced.

thirdly Is it possible, that only the effect of a limited number of the youngest initial plausibility values the second plausibility value affected. That means first Plausibility values the to incidents belong, that took place a long time ago, not the second plausibility value influence. If, as an alternative, a counter for the second plausibility value is used is, is both increase as well as decreasing the counter number possible dependent on from the plausibility level the first plausibility value. As a result, the counter number be reduced to zero and in this case, incidents that influence in the past, no longer the second plausibility value.

• – Kleine und/oder seltene Abweichungen von einem normalen, erwarteten oder gewünschten Zustand ergeben nicht unbedingt eine Unterbrechung des Stromrichterbetriebs.
• – Große und/oder viele Abweichungen von einem normalen, erwarteten oder gewünschten Zustand ergeben eine schnelle und zuverlässige Erkennung des Ausfalls und/oder der Fehlfunktion.
• – Das Verfahren läßt sich leicht implementieren. Die Parameter des Verfahrens wie beispielsweise Gewicht und Schwellenwerte können leicht angepaßt werden.
• – Ein Verfahren, das bei einer geringen Wiederholungsfrequenz funktionieren könnte aber Teil einer Steuerroutine auf höherer Ebene sein könnte, kann das Ergebnis der Fuzzy-Auswertung zuverlässig überprüfen, insbesondere die Zählerzahl überprüfen.

In particular, the first plausibility value is a difference between a current counter length (or the length square) and a comparison value, eg the average value of the length values of one or more rotation cycles of the pointer. Additionally or alternatively, a greater deviation from a normal, expected, or desired result has a lower degree of plausibility, and vice versa. If the difference is the first plausibility value, the plausibility level for higher differences is lower. For example, the predetermined criterion may be that a threshold value of the second plausibility value is reached and / or exceeded. In general, the fuzzy evaluation or one of its embodiments has the following advantages:

• - Small and / or rare deviations from a normal, expected or desired condition do not necessarily result in an interruption of the converter operation.
• - Large and / or many deviations from a normal, expected or desired state result in a fast and reliable detection of the failure and / or malfunction.
• The method is easy to implement. The parameters of the method such as weight and thresholds can be easily adjusted.
• A method that could operate at a low repetition frequency but could be part of a higher level control routine can reliably verify the result of the fuzzy evaluation, in particular checking the count.

Especially can if an error is detected while measuring at least one of the phase currents has been stopped, the operation of the power converter and again can be started and it can be a detailed analysis of a reason for the error after stopping and before or during restarting the converter. For example can detailed test routines be carried out at initialization of the converter operation.

Farther It is proposed to provide an arrangement that the method according to at least can perform any of the embodiments described above. The arrangement may be part of the power converter, in particular part of a Be control device of the power converter. It can by hardware and / or software are realized.

• – eine mit Computer ladbare Datenstruktur, die das Verfahren gemäß einer der in der vorliegenden Beschreibung beschriebenen Ausführungsformen bei Ausführung auf einem Computer durchführen kann,
• – ein Computerprogramm, wobei das Computerprogramm das Verfahren gemäß einer der in der vorliegenden Beschreibung beschriebenen Ausführungsformen bei Ausführung auf einem Computer durchführen kann,
• – ein Computerprogramm mit Programmitteln zum Durchführen des Verfahrens gemäß einer der in der vorliegenden Beschreibung beschriebenen Ausführungsformen bei Ausführung des Computerprogramms auf einem Computer oder einem Computernetz,
• – ein Computerprogramm mit Programmitteln gemäß dem vorhergehenden Punkt, wobei die Programmittel auf einem von einem Computer lesbaren Speichermedium gespeichert sind,
• – ein Speichermedium, wobei eine Datenstruktur auf dem Speichermedium gespeichert ist und wobei die Datenstruktur das Verfahren gemäß einer der in der vorliegenden Beschreibung beschriebenen Ausführungsformen durchführen kann, nachdem es in einen Haupt- und/oder Arbeitsspeicher eines Computers oder eines Computernetzes eingeladen worden ist, und
• – ein Computerprogrammprodukt mit Programmcodemitteln, wobei die Programmcodemittel auf einem Speichermedium gespeichert werden können oder gespeichert sind, um das Verfahren gemäß einer der in der vorliegenden Beschreibung beschriebenen Ausführungsformen bei Ausführung der Programmcodemittel auf einem Computer oder einem Computernetz durchzuführen.

Furthermore, the present invention includes the following:

• A computer-loadable data structure that can perform the method according to any of the embodiments described in the present specification when executed on a computer,
• A computer program, the computer program being able to carry out the method according to one of the embodiments described in the present description when executed on a computer,
• A computer program with program means for carrying out the method according to one of the embodiments described in the present description when the computer program is executed on a computer or a computer network,
• A computer program with program means according to the preceding item, the program means being stored on a computer-readable storage medium,
• A storage medium, wherein a data structure is stored on the storage medium, and wherein the data structure can perform the method according to any of the embodiments described in the present specification after it has been loaded into a main and / or working memory of a computer or a computer network, and
• - A computer program product with program code means, wherein the program code means may be stored on a storage medium or stored to perform the method according to one of the embodiments described in the present specification when executing the program code means on a computer or a computer network.

Especially may be the computer or part of the computer mentioned in any of the above paragraphs Computer network by using a processing unit, especially realized a central processing unit of the control device become. It can at least be the processing of the absolute value and error detection by the computer or computer network. The computer program product is to be understood as a product that can be sold or traded as a good.

in the The following will be specific embodiments of the present invention by way of example and with reference described on the accompanying drawing. The figures of the drawing show schematically:

1 an arrangement with a power converter, the AC side connected to an electrical load, in particular an asynchronous motor;

2 Devices and units used to process a sensor signal of an in 1 shown sensor used;

3 a flow chart of a collection of measurements that are used to perform a plausibility check;

4 a flowchart of an evaluation of measured values, in particular in the method according to the 3 collected measurements;

5 a flow chart of the fuzzy evaluation;

6 Diagrams showing the effect of an offset error;

7 Diagrams showing the effect of gain error;

8th a polygonal magnetic flux figure;

9 a graphical representation of the change in length of a rotating pointer during four consecutive rotation cycles in the case of the magnetic flux of 8th and in the case of a gain error;

10 a graphical representation of the change in length of a rotating pointer during four consecutive rotation cycles in the case of the magnetic flux of 8th and in the case of an offset error;

11 a circular magnetic flux figure;

12 a graphical representation of the change in length of a rotating pointer during four consecutive rotation cycles in the case of the magnetic flux of 11 and in the case of an offset error;

13 a graphical representation of the change in length of a rotating pointer during four consecutive rotation cycles in the case of the magnetic flux of 11 and in the case of a gain error; and

14 an array of units and devices used to perform a plausibility check.

1 shows a system 1 with a power converter 11 and a machine 13 , The power converter 11 can be constructed as known from the prior art. For example, the power converter may be a DC-AC converter (inverter) and may include three parallel paths connecting first and second DC connection lines of a DC link (not shown in the figure). Each of the paths may include a bridge with two electronic valves (eg, grid-thyristors or Insulated Gate Bipolar Transistors (IGBTs)) connected in series with one another. Every phase 5a . 5b . 5c an AC connection with a machine 13 may be connected to a connection point between the two electronic valves of a path.

Input signal lines of the electronic valves for receiving switching signals may be connected to a control signal input 10 be connected to the power converter. The control signal input 10 is with a control signal of a control device 4 over a connection 9 the in 1 connected arrangement shown connected. On the DC side, the power converter can be connected to a second power converter, which can output a direct current to the power converter. On the input side, the second power converter can be connected to a power supply network, for example a single-phase AC network of a railway system. However, other configurations and operations are possible, such as return of electrical energy from the power converter to the second power converter.

After the presentation in 1 connect the three phases 5a . 5b . 5c the AC connection the converter 11 with an electrical consumer 13 , in the present case with an asynchronous motor, which may be the drive motor of a railway towing vehicle.

A first 5a and a second 5b the phases 5a . 5b . 5c In any case, with only one current sensor 3a . 3b combined to measure the phase current. As the electrical consumer 13 in the three phases 5a . 5b . 5c is symmetrical, becomes the third phase 5c not combined with a current sensor. The current sensors 3a . 3b are via a sensor signal connection 7a . 7b for each sensor with the control device 4 connected.

For high power applications such as railway towing vehicles, the current sensor function may be based on the principle of detecting a current by evaluating the magnetic field generated by the current. An example of such a current sensor 3 and a corresponding arrangement for processing the sensor signal is in 2 shown. The current sensor 3 is in a phase 5 connected and generates a current signal that through the phase 5 guided current corresponds. A signal line for outputting the sensor signal is provided with a current / voltage converter 41 (eg, a shunt resistor) for converting the current signal into a voltage signal. The current / voltage converter 41 is optionally with a filter 43 connected to filter the voltage signal to eliminate signal parts that are the result of transient noise and to eliminate noise. An output of the filter 43 is with an input of an amplifier 45 connected to amplify the voltage signal. If the filter 43 is not provided, the output of the current / voltage converter 41 directly with the amplifier 45 be connected.

Between the current / voltage converter 41 and the amplifier 45 , for example between the filter 43 and the amplifier 45 , a comparator can be provided. The comparator can be used to compare the measured phase current with a phase current measured by a redundant current sensor.

An output of the amplifier 45 is with egg a signal converter / digital converter 47 (A / D) connected to digitize the voltage signal. The digitized signal can be used for digital data processing, in particular from a computer of the type described in US Pat 1 shown control device 4 is carried out. For example, the devices 41 . 43 . 45 . 47 the in 2 shown arrangement between the current sensor 3a or 3b and a signal input of the control device 4 be arranged. As an alternative, at least some of the devices 41 . 43 . 45 . 47 Part of the control device 4 be.

During operation of the power converter 11 uses the control device 4 for controlling the power converter 11 Phase current values, which are due to the current sensors 3a . 3b carried out measurement. For example, from a central processing unit of the control device 4 Model calculations to model the operation of the electrical load 13 and the plausibility check are performed.

After the in 3 The flowchart shown may include a control device (such as the control device 4 of the 1 ) operate a power converter as follows: in step S10, the control device reads a switching state of control valves and initiates corresponding switching actions in step S11. For example, the switching state or corresponding electrical potentials of the three phases may be represented using a set of three binary numbers saved in a group of consecutive switching operations. For example, a set of numbers 1, 0, 0 means that a first phase is to be switched to a high potential, a second phase to a low potential, and a third phase to a low potential. The high and low potentials may be the instantaneous potentials of a DC link connected to the power converter.

in the subsequent step S12, phase currents of at least two of Phases which correspond to the time of the instantaneous switching state of the valves, sampled. Typically, an asynchronous motor rotates as Drive motor of a railway towing vehicle is used, with one Rotor frequency from 0 to 150 Hz or higher, the switching frequency of the Power converter is in the range of 250 Hz to 800 Hz and the sampling frequency the sampling of the measured phase current values is even higher, namely in Range of several kHz, e.g. 2 to 4 kHz.

in the subsequent step S13, it is decided whether the sampled Phase current values are to be evaluated in a plausibility check. If For example, a specific sequence of successive switching states detected (e.g., the switching states 0, 0, 0 and 0, 0, 1 alternately repeated at least four times), is decided that the system is in a predetermined operating condition and that the sampled Phase current values are evaluated. As a result, that drives The method continues with step S14, wherein the sampled phase current values be secured so that the corresponding operating state of the system can be determined later. For example the values are in an additional Data field backed up to the appropriate set of numbers current switching action is assigned. It is also possible that phase current values for others Switching states, in particular for each Switching be secured, and that a marker information saved for marking the status and / or the values for evaluation becomes. Also can the phase current values are evaluated immediately.

To the step S14, and if it is decided in the step S13 that the sampled Phase current values are not evaluated in the plausibility check, returns the process returns to step S10.

As a result of the method, at least one quantity of the measured phase current values, preferably several quantities, is / are available for the evaluation. In a particular embodiment, six values of each measured phase current are used for evaluation in each rotation cycle of the magnetic stator flux vector of the machines, the six values corresponding to constantly spaced angles of rotation of the flux vector, eg corners of a polygonal flux figure (see, for example, Figs 8th shown river figure). Since the rotational speed of the machine is almost constant during the same rotation cycle, the six values also correspond to six constantly spaced points in time.

After the in 4 In the flow chart shown, the evaluation by the control device can be carried out as follows: In step S20, the control device reads the measured phase current values for at least one rotational position of the flux vector and calculates the length of the rotating pointer, in particular the current space pointer. For example, the measured phase current values of two of the phases are squared and the length square is calculated, taking into account the 120 degree angle between the two phases and the rotational position of the pointer.

In step S21, the length is compared with an average value (eg, the average of the pointer length values of the previous three or four (or other numbers) of rotation cycles of the flux vector or rotating pointer at the same rotational position.) Since a length change caused by a change in the consumer current ver caused by three or four consecutive rotation cycles is relatively low - if the consumer is an electric machine - the comparison will detect the effect of a measurement error.

In step S22, it is decided whether the difference between the length and the average exceeds a threshold. If the threshold is exceeded, the process proceeds to step S23. If the threshold is not exceeded, the process returns to step S20. In step S23, an optional error type analysis is performed, as described in more detail later with reference to FIG 6 and 7 is described. In another embodiment, step S23 is omitted. In step S24, a similar action is taken, such as interrupting or stopping the operation of the power converter. For example, the threshold may be 10 to 15 percent of the instantaneous length of the rotating pointer. It is possible to adjust the threshold to the operating state of the system and / or to define the threshold as a value related to the mean or instantaneous length.

The method, with reference to 4 may be modified. In particular, it may not be appropriate to interrupt power converter operation if the threshold is exceeded only once. It can therefore follow the procedure 5 be applied.

An exemplary manner of performing the fuzzy evaluation will be described below with reference to FIG 5 described. In step S1, the length of the rotating pointer is calculated. In step S1a, a corresponding comparison value of the length is obtained and / or processed. The calculated value and the comparison value are transmitted to a comparator comparing the two values in step S2. In particular, the comparator calculates the difference between the two values and optionally generates the absolute value of the difference. The difference or its absolute value is compared with a threshold value in step S3, for example by the threshold comparison means. If the threshold is exceeded, the method continues in step S4, wherein a counter value is increased by the amount by which the difference exceeds the threshold. If the threshold is not exceeded, the process proceeds to step S5, wherein the counter value is decreased by a constant amount. After the step S5, the process returns to the beginning and proceeds to the step S1.

To moves to step S4 the method proceeds to step S6, wherein a decision is made will, whether the counter value corresponds to a second threshold or to the second threshold equals or exceeds him. If this is the case, it is decided that a failure and / or a Malfunction has occurred, and in step S7 is a corresponding Act undertaken. If the second threshold is not exceeded, the process returns to the beginning and proceeds to step S1.

The The method described above may be by software and / or hardware be implemented. Furthermore, it is possible to modify the method. For example can be the counter value be reduced in step S5 by an amount that depends on the operating state depending on the system or the consumer.

6 and 7 show left figures of a rotating pointer. In each case, two circular or nearly circular lines and a starting point M are shown. This representation is simplified because the pointer length is usually such as in 12 is subject to fluctuations. The three straight lines with arrow passing through the starting point M enclose an angle of 120 degrees between each pair of lines and define a non-orthogonal coordinate system. They are not necessarily aligned with the directions of the three phase currents when the rotating pointer is a current space pointer of the three phases.

Six intersections of each of the figures in one of the circular or nearly circular lines with the arrow lines are marked with small circles and are designated t1 to t6. The distances between the starting point M and the intersections correspond to the lengths of the rotating pointer at the corresponding rotational position. Right in the 6 and 7 a corresponding diagram of the length is shown as a function of the rotational position or as a function of time. When the rotational speed of the pointer is constant, both interpretations are equivalent to the horizontal dimension and t1 to t6 can be construed as temporally constant spaced points.

In the 6 For example, a horizontal line representing the average of the length of a rotation cycle is shown as a dashed line. The mean value is the same for both circular lines left in the 6 , In the 7 a dashed line represents the mean value of the circular line. The mean value of the oval line is left in the 7 is as a thin dotted line right in the 7 drawn.

6 shows the effect of an offset error in measuring phase currents, wherein the rotating pointer of the current space pointer or, for example, a torque vector of an electromagnetic machine can be. The offset error gives a shift in the circular line (generally the pointer length figure), but does not affect the average of the length over a rotation cycle. As a result, there are two intersections between the length as a function of the rotational position or time, as shown on the right in the 6 seen. As a result, an offset error can be detected by a corresponding method. For example, if the pointer length is evaluated at six constantly spaced rotational positions (corresponding to t1 to t6), it is possible to decide that an offset error exists if the differences between the average and the length at the first rotational position or time t1 and at the fourth t4 have opposite signs, that is, when the difference is positive at one of the positions (t1) and positive at the other position (t4). Of course, an offset error could occur in measuring another of the phase currents or offset errors in measuring more than one of the phase currents, so that the circular line is shifted in a different direction. As a result, it is preferable to evaluate the difference between the average value and the pointer length at pairs of rotational positions or times separated by half a rotation cycle or half a rotation cycle period. However, other methods for detecting the characteristic influence of an offset error as shown in FIG 6 be performed. More broadly, the offset error results in oscillation of the pointer length with the fundamental frequency about the average pointer length value. Alternatively or additionally, this vibration could be detected by another method.

7 shows the effect of a gain error in measuring one of the phase currents. The length of the pointer has three points of intersection with the average of the length of a rotation cycle as shown on the right in FIG 7 shown on. As a result, the gain error can be detected by the same method as described in the previous section, but the differences between the average value and the length have the same sign for the pair of rotational positions or times (positive or negative). More broadly, the gain error results in oscillation of the pointer length at twice the fundamental frequency about the average pointer length value. Alternatively or additionally, this vibration could be detected by another method.

As already mentioned shows the 8th a hexagonal polygonal magnetic flux figure of an electromagnetic machine, especially an asynchronous motor. The corners correspond to six constantly spaced rotational positions or times t1 to t6. Since the sequences of control signals generated in the area of the corners at each corner of the rotation cycle of the machine during static operation are similar or the same, these sequences or the information about these sequences can be used to determine the times t1 to t6. Similarly, the 11 an almost circular magnetic flux figure. Corresponding centers or starting points are designated by the reference letter M.

The The aim of each control method for controlling a power converter is a magnetic flux, a circle as close as possible comes. In practice, frequency control methods are chosen, the approximations the ideal circular magnetic flux figure generate (e.g., a hexagonal or eighteen-cornered river figure). The number and position of the sample points is determined by the number and Position of the phases / phase windings, for example for the typical System with three phases determined. For the typical case of three Phases with 120 degrees phase shift (and the corresponding angular position) can be like already described the six sampling points with 60 degrees intervals between be used to them. It is preferred that the control system be connected to a the sampling points to one of the phases switches. Switch all non-circular control methods typically to the next Phase around when the flow pointer crosses the angular phase position, and thus create a "corner" (in the river as well in the stream, where the current may be out of phase under load).

In the 9 . 10 . 12 and 13 Illustrated graphics of a rotating pointer belong to a current space pointer. From the graph in 9 It can be seen that there are significant variations in pointer length in successive rotation cycles. However, it can also be seen that the variations at the six corners of the graph are relatively small. As a result, it is possible to reduce the gain error that causes the graph in FIG. deformed by applying the above with reference to 6 and 7 to recognize the method described. In particular, can be out 9 recognize that the distance between the corners at rotation angles or times t3 and t6 is smaller than the distances between the other pairs of opposite corners, indicating a gain error.

In the 10 For example, the graph is shifted down (relative to the starting point M), indicating an offset error.

In the 12 and 13 shown graphs of a current space pointer correspond to the Flußfigur the 11 , In the 12 is a shift of the graphical representation to the left (with respect to the starting point M) recognizable, which indicates an offset error. In the 13 The graph shown is similar to the oval graph of FIG 7 deformed, indicating a gain error.

Around erroneous detection (e.g., due to variations in the electrical Load) the ones mentioned above Deviations (which are the result of vibrations) for several Periods of the fundamental frequency of the phase currents are detected. The detection does not have to be fast be because the deviations are typically small. It can to at the same time other methods of detecting more serious disturbances have been applied become. As a result, it is safe, reliable and reliable at the same time sufficient, the above mentioned Apply fuzzy evaluation. In addition, procedures could to compensate for load deviations (e.g. Use of torque instead of current pointer length). Thereby becomes the reliability the detection increases.

In the 14 The arrangement shown comprises units 51 to 59 that part of a device 50 could be. The device 50 can be realized by software and / or hardware and can be part of the control device 4 of the 1 be. An A / D converter 47 (in particular the converter 47 of the 2 ) is with a measurement processing device 51 for processing measurements of phase currents conducted through at least two of the three phases. The measurement processing device 51 is with a calculation device 53 for calculating absolute values dependent on a length of a rotating pointer, the rotating pointer rotating at a frequency equal to a fundamental AC frequency of the phase currents. A selection device 55 for selecting sequences of control signals is provided, the sequences being either identical or similar sequences of control signals. These are repeatedly generated by the control device for controlling the power converter. With the selection device 55 the calculation of the absolute values for operating times of the selected sequences can be effected. In the present particular embodiment, the selection device is 55 with a control unit 57 connected to generate control signals for controlling the operation of the power converter. The control unit 57 may include means for reading the sequences and generating corresponding control signals such as PWM signals. The selection device 55 is with the calculation device 53 connected and the computing device 53 is with a recognition device 59 for detecting an error in measuring at least one of the phase currents. In an alternative embodiment, the calculation device may also be connected only indirectly to the selection device.

During operation of the system, measured values of the phase currents in digital form are obtained from the A / D converter 47 to the measurement processing device 51 transferred, the measured values to the computing device 53 transmits, namely at least the measured values, the one by the selection device 57 the selection procedure. A corresponding connection between the selection device 55 and the measurement processing device 51 can be realized in a modified arrangement. Furthermore, the measurement processing device 51 process the readings for other purposes, such as the use of the readings in controlling the operation of the system.

In the by 14 Example shown by the selection device 55 detects when the identical or similar sequences of control signals occur during operation of the system. It is triggered and / or caused by the calculation of the corresponding absolute values, so that the detection device 59 can detect an error in measuring the phase currents.

Summary

Plausibility check of a electric three-phase system

One Error when measuring phase currents of the Systems is detected by absolute values for certain operating times be evaluated. The absolute values are of the length of one depending on the rotating pointer, which rotates at a frequency equal to a frequency of the phase currents, being the length of the rotating pointer is dependent on the phase currents. The absolute values are used to detect an error when measuring at least one the phase currents used. A control device repeatedly generates identical or similar Consequences of control signals for controlling the power converter (steps S10 and S11). Based on the information about these consequences, the corresponding absolute value selected or a corresponding measurement information stored for later evaluation (Steps S12, S13 and S14).

Claims (18)

Method for verifying operation of a three-phase electrical system ( 1 ) on plausibility, whereby the system ( 1 ) a power converter ( 11 ) and a load ( 13 ), in particular a drive motor of a railway towing vehicle, and wherein - the load ( 13 ) via an AC line with three phases ( 5a . 5b . 5c ) with the power converter ( 11 ) connected - the load ( 13 ) to the three phases ( 5a . 5b . 5c ) is symmetrical, - of at least two ( 5a . 5b ) of the three phases ( 5a . 5b . 5c ) guided phase currents are measured, - the power converter ( 11 ) by a control device ( 4 ), - the control device ( 4 ) Values of the measured phase currents for operating the system ( 1 ), - the control device ( 4 ) repeats identical or similar sequences of control signals for controlling the power converter ( 11 ), wherein absolute values are calculated for operating times corresponding to the sequences, the absolute values being related to a characteristic of a quantity, the quantity describing an operating condition of the system, the quantity being a rotating pointer rotating at a frequency which a frequency of the phase currents carried by the phases is equal, wherein a length of the rotating pointer depends on the phase currents, the absolute values being dependent on the length of the rotating pointer, and at least one of the absolute values for detecting an error in measuring at least one of the Phase currents is used. Method according to claim 1, wherein the absolute values calculated using the measured phase currents. The method of claim 1 or 2, wherein the rotating Pointer is a rotating current space pointer, the phase shifts and amplitudes of the three phase currents. Method according to claim 1 or 2, wherein the rotating pointer is activated by the control device ( 4 ) using the measured phase currents is calculated pointer. The method of claim 4, wherein the load is a rotating one Three-phase machine includes in particular an asynchronous motor, and wherein the rotating pointer corresponds to a torque of the machine. Method according to one of claims 1 to 5, wherein the absolute value a square the length of the Rotating pointer is representing square value. Method according to one of claims 1 to 6, wherein the absolute values for one Variety of rotational positions of the pointer are calculated, of which each one of identical or similar Corresponds to consequences, and where A difference between the Absolute values for at least two of the rotational positions or - deviations at least two of the rotational positions of a comparison value used to determine a fault type when measuring at least one of the phase currents become. Method according to one of claims 1 to 7, wherein a comparison by comparing the absolute value with a previous absolute value at a same rotational position of the pointer or by comparison the same with one of a plurality of preceding absolute values derived comparison value, and wherein the comparison result is used to detect the error in measuring at least one of the phase currents. Method according to one of claims 1 to 8, wherein in carrying out the Plausibility check a first plausibility value is determined repeatedly, wherein the first plausibility value a Momentary measure of one Degree of plausibility and / or implausibility is a second plausibility value of a plurality of first plausibility values each of the first plausibility values the second plausibility value according to its degree of plausibility and / or implausibility and wherein an action is taken when the second plausibility value is on predetermined criterion fulfilled. Method according to one of claims 1 to 9, wherein upon detection of an error in measuring at least one of the phase currents of the operation of the power converter ( 11 ) and a detailed analysis of a reason for the error after stopping and before or during a restart of the power converter ( 11 ) is carried out. Computer-loadable data structure configured to perform the method according to one of the preceding method claims while the data structure is executed on a computer, in particular on a computer of a control device for controlling a mode of operation of the power converter ( 11 ). A computer program, the computer program being designed to carry out the method according to one of the preceding method claims while the computer program is executed on a computer, in particular on a computer of a control device for controlling a mode of operation of the power converter ( 11 ). Computer program with program means for carrying out the method according to one of the preceding method claims when the computer program is executed on a computer, in particular on a computer of a control device device for controlling the functioning of the converter ( 11 ), or on a computer network. Computer program with programs after the previous one Claim, wherein the program means are readable on a computer readable Storage medium to be stored. A storage medium in which a data structure is stored on the storage medium and wherein the data structure can perform the method according to one of the preceding method claims after it has been loaded into a main and / or main memory of a computer, and in particular on a computer of a control device for controlling a Operation of the power converter ( 11 ), or a computer network. Computer program product comprising program code means, the program code means being or able to be stored on a storage medium for carrying out the method of one of the preceding method claims when the program code means are executed on a computer, in particular on a computer of a control device for controlling a functioning of the power converter ( 11 ), or on a computer network. Arrangement for checking the operation of a three-phase electrical system ( 1 ) on plausibility, whereby the system ( 1 ) a power converter ( 11 ) and a load ( 13 ), in particular a drive motor of a railway towing vehicle, wherein the load ( 13 ) via an AC line with three phases ( 5a . 5b . 5c ) with the power converter ( 11 ), the load ( 13 ) to the three phases ( 5a . 5b . 5c ) is symmetrical, the power converter ( 11 ) by a control device ( 4 ) and wherein the arrangement comprises: a measurement processing device ( 51 ) for processing measurements of at least two ( 5a . 5b ) of the three phases ( 5a . 5b . 5c ) guided phase currents, - a calculation device ( 53 ) for calculating absolute values in response to a length of a rotating pointer, the rotating pointer rotating at a frequency equal to an AC frequency of the phase currents, and wherein the computing device ( 53 ) with the measurement processing device ( 51 ), - a selection device ( 55 ) for selecting sequences of control signals, the sequences having identical or similar sequences repeated by the control device ( 4 ) generated control signals for controlling the power converter ( 11 ) and wherein the selection device ( 55 ) can cause the calculation of the absolute values for operating times of the selected sequences, - a recognition device ( 59 ) for detecting an error in measuring at least one of the phase currents, wherein the recognition device ( 59 ) with the computing device ( 53 ) and can detect the error using at least one of the absolute values. Power converter ( 11 ) with the arrangement of the preceding claim.