WO2003041256A2

WO2003041256A2 - Converter and method for determination of a current sinor

Info

Publication number: WO2003041256A2
Application number: PCT/EP2002/011172
Authority: WO
Inventors: Harald Wolf; Ralph Mayer
Original assignee: Sew-Eurodrive Gmbh & Co
Priority date: 2001-11-08
Filing date: 2002-10-04
Publication date: 2003-05-15
Also published as: WO2003041256A3; CN1575539A; WO2003041258A2; DE10248375C2; DE10248375A1; WO2003041258A3; CN100361381C

Abstract

The invention relates to a converter and method for determination of a current sinor for a converter operated with pulse width modulation, comprising a signal electronic circuit and a power end-stage with a power circuit arranged as three half-bridges each divided into a lower and an upper branch. Means for determining the instantaneous current are arranged in either all three of the lower or all three of the upper branches of the half-bridges, at least one corresponding measured current value is recorded by two of the three means of determining current within a timespan and a current sinor is generated, or the current values in the output branches are generated from said current measured values, measured by the pair of current determination means, in particular for use in a control and/or regulation method. Said pair of means are variously selected depending on the output voltage sinor.

Description

Inverter and method for determining a current space vector

Description:

The invention relates to a converter and a method for determining a current space vector.

In the case of a converter with digital control operated on the output voltage side with pulse width modulation, in order to achieve sufficient control quality, it is necessary for a pulse width modulation period to determine a value for the current space pointer which corresponds to the mean value of the actually existing current space pointer formed over this pulse width modulation period. The pulse-width-modulated output voltage results in what is known as a current ripple in the case of inductive loads, in particular in the case of an electric motor. The determination of the current space vector must be unaffected by this current ripple. In some converters of the prior art, the current sensors are arranged in the feed lines to the motor, as a result of which continuous current signals occur which are freed from the current ripple by analog filtering and can then be used for the regulation.

From the book 'Practice of field-oriented three-phase drive control systems' by Nguyen Phung Quang and Jörg-Andreas Dittrich, second edition, from 1999, pulse-width modulated operable inverters are known from pages 110-113, in which current measurements are made with the help of three (Fig. 5.1) or only two (Fig.5.3, 5.5, 5.6, 5.9, 5.11) potential-isolating current sensors are provided in the output branches. A current space vector with two degrees of freedom is determined from the current measurements (Fig. 5.3, 5.5, 5.6, 5.9, 5.11). On pages 76 to 77 it is also stated that in order to record the fundamental vibration of the motor currents without harmonics, i.e. without current ripple, the current measurement values must be recorded at suitable times. The deviation from the fundamental oscillation is called the residual current space vector. The current measurement values are then acquired at the zero crossing of the residual current space vector. A disadvantage of these known converters is that at least two expensive, potential-isolating current measuring means are necessary. DE 196 81 189 T1 discloses a converter which can be operated with pulse width modulation and has an intermediate circuit capacitor (FIG. 1, reference number 3) and an output stage connected to it, which comprises only one current sensor for detecting the intermediate circuit current. Depending on the angle of the output voltage pointer, the current detection therefore contains information about a single phase current or sums of phase currents. The disadvantage here is that the determination of a current space vector cannot be carried out satisfactorily at certain angles of the output voltage vector.

A converter is known from the writing by Francesco Parasiliti, "Low cost current Sensing in DSP Based Drives", Industrial Electronics, 1999, ISIE'99, Proceedings of the IEEE, Volume 3, 1999, International Symposium, Volume 3, 1999, in which shunt resistors are arranged in all lower branches of the half bridges as a means for detecting the respective currents All three current measured values are recorded, the measurement impulses being in the middle of the time period within a pulse width modulation period in which the discrete switching state (000), ie the Zero vector v ₀ is present (page 1287, left column and Figure 7.) In the specific embodiment there is a small time offset to this center due to consideration of dead times, signal propagation times and the like. The current space pointer is then determined from the associated measured values however, that in the shaded areas of the hexagon (page 1287, left column, last ter section and Figure 8) no or only an inaccurate determination of the current space vector is possible, since either no zero vector is used in these shaded areas or only very briefly. If the zero vector is not present or is only present for a very short time, there is automatically no or only a very short current measurement signal in one phase, which is therefore for one

Evaluation is not available or only inaccurately. The result of this is that in the case of an output voltage space vector averaged over a pulse width modulation period from the shaded area (page 1287, FIG. 8), at least one current measurement value is falsified.

A converter is known from US Pat. No. 5,815,391, in which means for detecting the respective currents are arranged in all lower branches of the half bridges (FIG. 2A). The corresponding table (Figure 2B) explains that the current in phase A can be measured depending on the switching states (example 1st line: measurable;) or can be calculated from the two other current measured values (5th line: calculable;) , If However, if the lower switch of phase A and another lower switch are open, the measured current value of phase A cannot be determined. Such switching states are identified in this document as "unknown" in phase A.

The table 2B and the problem described relate to the switching states, that is to say the respective instantaneous state of the converter. The table 2B mentioned only relates to phase A. However, the associated tables for phases B and C can be drawn up. It is easy to see that with some switching states, namely (111), (110), (101), (011), no or only a current measurement value can be determined and therefore in no case a current space pointer for these instantaneous states. US Pat. No. 5,815,391 does not therefore show how to determine a current space pointer which corresponds to the mean value of the actually existing current space pointer formed over a pulse width modulation period. In particular, it is not shown how to proceed so that the current space pointer is not falsified by current ripples.

The duration of a "known" switching state within a pulse width modulation period depends on the following factors: carrier frequency of the pulse width modulation, type of pulse width modulation method. If this duration of a "known" switching state is too short, the current space pointer cannot be determined.

This document does not explain how the pulse width modulation method is carried out; however, it will be apparent to those skilled in the art that scripture teaches that these "unknown" conditions pose a problem. Column 1, lines 40-44 speaks of a 'back calculation', but this is not described in an executable manner. The document teaches the person skilled in the art that the problem of the 'darknown' states can be avoided by, according to FIGS. 5 and 6, measuring means, in particular potential-free measuring means such as Rogowski coils or the like, both in the upper and in the lower branches of the half bridges , are used. By clever

Combining the measurement signals, in particular adding the measurement signals of the lower and the upper measuring means, a continuous current measurement signal is achieved and the "unknown" states are no longer a problem. This teaching of US Pat. No. 5,815,391, to use measuring devices both in the upper and in the lower branches of the half-bridge, would also be applicable to the aforementioned IEEE document. In this way, the current measurement in the shaded areas there could be carried out easily and precisely. However, complex measuring means, in particular those for overcoming the potential barrier and thus expensive ones, would have to be provided. So the solutions would be very complex and expensive.

The invention is therefore based on the object of developing a converter in which as many expensive parts as possible can be dispensed with and in particular

Mass production an inexpensive manufacture can be achieved and nevertheless a method for determining a current space vector can be carried out.

According to the invention the object is achieved in a method according to the features specified in claim 1 or 16 and in a converter according to the features specified in claim 12.

Essential features of the invention in the method are that for determining a current space vector for a converter which can be operated with pulse width modulation, comprising signal electronics and a power output stage which comprises power switches arranged in three half bridges, each comprising a lower and an upper branch,

in which

means for detecting the respective currents are arranged either in all three lower or in all three upper branches of the half bridges,

at least one associated current sample value is determined with two of the three means for detecting the currents within a period of time that is a pulse width modulation period,

a first and a second associated current sample value are provided with a second means of the pair before and after the detection of a current sample value acquired with a first means of the pair, a current space pointer is formed from the current samples determined with this pair of means or the current values are formed in the output branches, in particular for use in a control and / or regulating method, and

this pair of means is selected differently depending on the mean output voltage space vector,

- The mean output voltage space vector is determined by the output potentials of the three averaged over a pulse width modulation period

Output phases.

To determine the current space vector, only the two most suitable of the three possible current measured values are advantageously selected for each pulse width modulation period. In particular, the invention is based on the surprising finding that the information about the current space vector is contained in the time course of the two most suitable current measurement signals. It is even permitted that even switching states may occasionally occur within this pulse width modulation period, which in themselves would not allow the current space vector to be determined, that is to say are “unknown” states in the sense of US Pat. No. 5,815,391!

Since in the variant with means for detecting the respective currents arranged in all three lower branches of the half bridges, that phase in which the corresponding pulse width modulation signal is shorter than a minimum duration in the LOW state is not always used, a reliable determination of the current space pointer is necessary guaranteed using a suitable pulse width modulation method, in particular a symmetrical pulse width modulation method. The lower circuit breaker is closed and the upper one is open in the respective LOW state. The reverse is the case with a HIGH state.

In addition, as many expensive parts as possible can be dispensed with, in particular potential-isolating current sensors, since the currents are not measured in the output branches but in the lower or upper branches of the half-bridges. In the case of mass production, inexpensive production can be achieved.

In particular, in the invention with the two of the three means for detecting the currents, the associated current samples are not recorded simultaneously within the pulse width modulation period and / or the associated measurement impulses are not carried out simultaneously. The advantage here is that a microcontroller with only a single analog-digital converter and no additional sample-hold circuit can be used with the converter, that is, no sample-hold circuit arranged externally from the microcontroller.

In particular, at least one of the two means detects a current sample more than once per pulse width modulation period and / or provides more than one measurement trigger within a pulse width modulation period. In particular, a current sample is detected exactly twice per pulse width modulation period with a first of the two means and the measurement impulse for the second of the two means lies in the middle of the pulse width modulation period. The two measurement impulses of the first of the two means are advantageously equidistant from the center of the pulse width modulation period, that is to say that the time interval from the first measurement impulse of the first means to the first measurement impulse of the second means is the same as the time interval from the measurement impulse of the second means to the second measurement impulse of the first mean and the last-mentioned measurement impulse of the second mean lies in the middle of the pulse width modulation period.

In the variant with means for detecting the respective currents arranged in all three lower branches of the half-bridges, the temporal center of the pulse width modulation period denotes the temporal center of the respective LOW states. During these states, current measurement signals are present on the means for detecting the respective currents.

The current measured values of the phases belonging to the two means mentioned are sufficient to determine the current space vector. In the present invention, the total of three current measured values of the two means are always available because the invention uses an advantageous pulse width modulation method, such as a symmetrical pulse width modulation method, and the output voltage space vector is limited to the inner circle of the hexagon. This also means, for example, that no corner vectors of the hexagon occur permanently over an entire pulse width modulation period.

So that the current space pointer is free from the influence of the

Pulse ripples due to PULSE WIDTH MODULATION, the measurements must be initiated at suitable times. The points in time at which the two means started to measure would be advantageous simultaneously and in the middle of the

Pulse width modulation period. Then, however, either two A / D converters or external sample-hold circuits would be necessary, which would be expensive. According to the invention, two temporally staggered measurement impulses are carried out with the first means. The time at which the second agent starts to measure is in the middle of the pulse width modulation period. The times of the two measurement impulses of the first means each have the same time interval from the center of the pulse width modulation period. Thus, as a result of the averaging and because of the symmetry of the current ripple, which runs symmetrically to the center of the pulse width modulation period due to the symmetrical pulse width modulation method, a current measurement value is determined using the first average, which is equal to that current measurement value which is based on the first average Time of the middle of the pulse width modulation period would be measurable.

In particular, the current space vector is determined for each pulse width modulation period. The advantage here is that the most frequent possible determination of the current space vector can be provided even at a low pulse width modulation frequency in order to improve the control quality of the control method of the converter.

In preferred embodiments, the current measurement signals are formed at the lower intermediate circuit potential, current measurement values assigned to the respective half bridges being determined from current sampled values which are derived from the current measurement signals. In particular, the reference potential of the signal electronics, which comprises a control and regulating device of the converter, corresponds to the reference potential on which the current measurement signals are formed. The advantage here is that optocouplers for electrical isolation can be saved. The reference potential of the signal electronics also has the lower intermediate circuit potential. A significant advantage here is that the control signals of the lower circuit breakers in the half bridges can be generated by the signal electronics without a high voltage gap occurring, which would make complex electrical isolation necessary. Only the control signals of the upper one

Circuit breakers must be controlled via optocouplers or other potential-isolating devices. In particular in the case of a converter which does not have any connections for a sensor, such as a speed or position sensor, or for other devices to be isolated, an important step towards saving costs and reducing the number of parts can be achieved.

Power semiconductor switches, such as IGBT of the npn type, which can be controlled with control voltages which have the lower intermediate circuit potential as a reference potential, can advantageously be used as lower circuit breakers in the half-bridges. When using complementary power semiconductor switches, such as IGBT of the pnp type, the upper intermediate circuit potential must then be selected as the reference potential for the current measurement and for the signal electronics and the converter designed accordingly.

In a preferred embodiment, the pair is dependent on the angle of the

Output voltage space pointer and not selected different from the pulse width modulation pattern. The advantage here is that the technical implementation is particularly simple.

In another embodiment, a period of two or more pulse width modulation periods is used instead of the one pulse width modulation period. Of

The advantage here is that the current space vector can be determined even at a high switching frequency. There is a slight falsification, but this is low at a high switching frequency.

In a preferred embodiment according to the invention, the detection of the current sample value detected with the first current measuring means, ie means for detecting the currents, is carried out centrally in the pulse width modulation period. In particular, with a second of the two current measuring means, a first and a second associated current sample value are recorded symmetrically before and after the acquisition of a current sample value acquired with the first current measurement means. The advantage here is that current measurement values that are not falsified by current ripples due to pulse width modulation can be determined with only one analog-digital converter without external sample-hold circuits. The determination of the current measured values is therefore particularly simple and can be carried out with little computing effort.

If additional external sample-hold circuits are used, the associated current sample values could be acquired simultaneously and / or the associated measurement impulses could be carried out simultaneously using the two of the three means for acquiring the currents. In particular, the measurement impulses would be carried out centrally in the pulse width modulation period. The advantage here would be that the determination of the current measured values is particularly simple and yet there are no falsifications caused by current ripples.

In another preferred embodiment of the invention, with the first of the two means for detecting the currents, an associated current sample is acquired at least one pulse width modulation period after the acquisition of a current sample associated with the second means for detecting the currents. In particular, the measurement impulses associated with the two means for detecting the currents lie in different pulse width modulation periods. The advantage here is that a current space pointer can be determined even at high switching frequencies using only a single analog-digital converter and without additional sample-hold circuits, in particular without falsifications caused by current ripples.

In a preferred embodiment, the first current sample value of the pair is recorded more than once per pulse width modulation period and a value interpolated and / or mean value is formed from the recorded values according to the times of the respective acquisition. In particular, a first current sample of the pair associated with a first half bridge is acquired before and after the second current sample of the pair associated with a second half bridge. The advantage here is that a fictitious current measured value can be determined with good accuracy by interpolation or averaging, which is available at the same time as the other recorded current sample.

In a preferred embodiment, the detected current space vector corresponds to the mean value of the current space vector over a pulse width modulation period. The advantage here is that the current ripple as a result of the pulse width modulation does not falsify the result and thus also the control method.

Essential features of the invention in the converter, comprising a signal electronics and a power output stage, which in three, each a lower and an upper branch comprehensive circuit breakers arranged half bridges, wherein the converter can be operated pulse width modulated, are that

- Current measurement signals detected by the three means for detecting the currents are fed and / or can be fed to only a single analog-digital converter via a multiplexer.

The advantages here are that expensive potential-isolating means can be saved and the construction volume of the converter can be reduced.

In a preferred embodiment, the means for current detection comprise resistors, in particular shunt resistors. The advantage here is that the current acquisition is extremely inexpensive.

In a preferred embodiment, the means for current detection are arranged in the half bridges in such a way that they are connected either to the upper or to the lower intermediate circuit potential. The advantage here is that funds for

Electrical isolation can be saved. In a further preferred embodiment, the signal electronics have a reference potential, which is also the reference potential for the means for current detection. If this reference potential is Uz-, the means for potential isolation for the control signals of the lower circuit breakers of the half bridges can be saved. If this reference potential is Uz +, then the means for

Electrical isolation for the control signals of the upper circuit breakers of the half bridges can be saved.

In a preferred embodiment, the signal electronics have a reference potential, which is also the reference potential for the means for current detection. The advantage here is that costly potential-isolating means can be saved. Because if the means for current detection were provided in the supply lines to the motor, expensive potential-isolating means would be necessary. In a preferred embodiment, each means for current detection can be assigned to a single analog-digital converter by means of a multiplexer and / or switch. The advantage here is that the converter can be carried out inexpensively, in particular by saving additional analog-digital converters.

In principle, it would also be possible to assign more than one analog-digital converter to the means for current detection. It would be advantageous that when using appropriately arranged multiplexers, the respectively selected pair of means for detecting the currents can be assigned to the two analog-digital converters, for example, and additional sample-hold circuits can be saved. In particular, an analog-digital converter could be assigned to each means for current detection. The advantage here is that no additional sample-hold circuits are necessary and a synchronous acquisition of the current samples can be guaranteed. A sample-hold circuit would then preferably also be arranged between each means for current detection and the assignable analog-digital converter. The advantage here would be that the current space pointer without

Falsification by current ripple can be determined and a single analog-digital converter is sufficient. The signal electronics could further preferably have a microcontroller which has a single analog-digital converter and a sample hold circuit which is not included in the microcontroller, that is to say an additional external one. The advantage here would be that an inexpensive microcontroller can be used, which comprises only a single analog-to-digital converter.

In a preferred embodiment, the signal electronics comprise means for generating pulse-width-modulated control signals for the circuit breakers, and the signal electronics have a reference potential, which is also the reference potential for the means for current detection. The advantage here is that means for electrical isolation can be saved.

In an advantageous embodiment of the invention, a respective means for current detection can be assigned to a single analog-digital converter by means of a multiplexer and / or switch. Costly further analog-digital converters can then be dispensed with.

In another advantageous embodiment of the method according to the invention, the essential features are that means for detecting the respective currents are arranged either in all three lower or in all three upper branches of the half bridges,

the pulse width modulation frequency is greater than a minimum frequency,

within a period of time that is two pulse width modulation periods, at least one associated current sample value is determined with two of the three means for detecting the currents, and the same average output voltage space vector is output in both pulse width modulation periods,

- A first current sample in the middle of the first of the two pulse width modulation periods is detected with a first average of the pair and an associated current sample in the middle of the second, ie immediately following, with a second average of the pair

Pulse width modulation period is recorded,

a current space pointer is formed from the current samples determined with this pair of means or the current values are formed in the output branches, in particular for use in a control and / or regulating method, and

- And wherein the mean output voltage space vector is determined by the differences between the output potentials of the three output phases, which are averaged over a pulse width modulation period.

The advantage here is that even at high pulse width modulation frequencies, a single analog-to-digital converter is sufficient and no external sample-hold circuits are necessary. If the frequency falls below the minimum frequency, which is, for example, 10 kHz, the method according to the invention described above can be switched over. Thus, the advantages are available even at all pulse width modulation frequencies. Further advantageous embodiments result from the subclaims. LIST OF REFERENCE NUMBERS

1 control and regulating device

21 Sample-hold circuit

22 microcontrollers

23 multiplexers

The invention will now be explained in more detail with reference to figures:

A power output stage for a three-phase converter is schematically outlined in FIG. The circuit breakers with associated freewheeling diodes are identified by six switch _symbols S _Ro , S _So , S _To , S _Rll , S _Su and S _Tu . In the lower branches of the

Half bridges are shunt resistors R _R , R _S and as means for current detection

R _τ arranged. These are thus connected to the lower reference potential U _z , which is also the reference potential of the signal electronics of the converter comprising a control and regulating device 1. The motor-side output potentials of the converter are U _R , U _s and U _τ ; the motor currents are labeled I _R , I _s and I _τ .

The resistors are connected to amplifier circuits V _R , V _s and V _τ , which each generate a current measurement signal I _m , I _SM and I _{m on the} output side. From the three

Current measurement signals are derived by analog-digital conversion current samples, from which current measured values are formed. The current space vector is determined from these current measurement values, the half bridges belonging to the current measurement values being selected as a function of the angle of the voltage space vector. Since I _R + I _s + I _τ = 0 applies to the motor currents according to Kirchhoff's rule, this has

Current space pointer two degrees of freedom. It therefore has two independent parameters. It is therefore sufficient in principle that only current measurement values assigned to two branches are used to determine the current space vector. It should be noted, however, that the pulse width modulation signals of the converter do not have to be constant within a pulse width modulation period, i.e. the switching states change in such a way that even switching states can occur from time to time that, by themselves, would not allow the current space vector to be determined!

To achieve a sufficient control quality, it is sufficient for the control method of the converter to determine a current space pointer as a computing variable once per pulse width modulation period, which represents the mean value of the physical current space pointer formed over a pulse width modulation period. The pulse width modulation signals PWM _R (t), PWM _s (t) and PWM _τ (t) generated by the control and regulating device 1 determine the state of the circuit breakers

^S _R0 's _o > Sτ _o> S _RU ' ^S s_ ^{and s} τ _u ■ ^The pulse width modulation signals are provided as follows: If the respective pulse width modulation signal, for example PWM _R (t), is 1, the associated upper power switch, for example S _Ro , closed and the associated lower circuit breaker, for example S _Ru , opened. In this case, the associated output voltage potential is U _{z +} . Is that

Pulse width modulation signal 0, that is to say in the LOW state, the associated circuit breakers are in the respective other state and the associated output voltage potential is U _z _. The so-called dead time inserted in the practical implementation, during which the respective upper and lower switch is open, is not relevant for the basic function of the invention.

When a pulse width modulation signal belonging to a half bridge is in the LOW state, the associated motor current flows in the lower branch of the associated half bridge and thus via the respective shunt resistor. In order to record the current measured values precisely, the associated motor current must flow through the shunt resistor for longer than a minimum period. The minimum duration is dependent on the filter effect of the measuring amplifier circuits, including amplifier circuits V _R , V _s and V _τ . A filter effect is achieved by appropriate circuitry to suppress noise components or interference components in the current measurement values. For example, with a pulse width modulation frequency of 16 kHz, a filter time constant in the range from 0.5 μs to 2 μs is advantageous.

In the invention, the amplitude of the averaged

Output voltage space vector limited to? 7 _Z / / 3, so that a sinusoidal three-phase voltage system can always be generated. In the case of the invention, this limitation denotes the maximum modulation. Oversteer, i.e. leaving this work area defined by the limitation, is always avoided.

The invention therefore relates only to those converters which are operated in such a way that pulse width modulation means that in no pulse width modulation period is there such a high modulation that only a single active one Switching state occurs for a full pulse width modulation period. In such an active switching state, the pulse width modulation signals PWM _R (t), PWM _s (t) and PWM _τ (t) take the values (110), (101), (100), (001), (010) or ( 011), which are also referred to as discrete active output voltage space pointers. Active switching states therefore do not include switching states (111) and (000). The last two switching states mentioned are each also referred to as a discrete zero-voltage space vector or vector.

So-called edge vectors are those output voltage space pointers averaged over a pulse width modulation period which, when a single one of the active switching states is theoretically applied, over a whole

Pulse width modulation period would result. In the invention, the output voltage space pointer averaged over a pulse width modulation period never takes on the value of such an edge vector because of the above-mentioned limitation of the working range.

Spoken in the picture of the aforementioned IEEE script by Francesco Parasiliti according to FIG. 8 there on page 1287, the above-described limitation of the output voltage space vector averaged over a pulse width modulation period means that it lies within the maximum inner circle of the hexagon. The averaged output voltage space pointer never takes on the value of an edge vector, and therefore never comes to lie on the corners of the hexagon in the invention.

Since the output voltage space vector is limited, various discrete switching states alternate. A current space pointer can be determined for each pulse width modulation period by skillfully selecting the times for the measurement impulses within a pulse width modulation period. The alternating states include, among other things, states in which the determination of a current space vector would not be possible if they were applied continuously over a pulse width modulation period, such as state (101) according to the last line of Table 2B from US Pat. No. 5,815 391. In particular, for example, a measurement initiation is carried out at a first time in a first phase, a second measurement initiation at another second time in a second phase and a third at a third time in the first phase Measuring offense. By averaging the first and third measured values, it is achieved that the ascertainable current space pointer is not falsified by current ripples in the pulse width modulation method permissible for the invention, in particular symmetrical pulse width modulation methods in which the output voltage space pointer is limited to the inner circle of the hexagon. This means that only a single analog-to-digital converter is necessary, in particular without additional external sample-hold circuits.

FIG. 2 shows a feasible variant for part of the control and regulating device 1 from FIG. 1. This comprises a sample-hold circuit 21, to which the three current measurement signals are fed, and a microcontroller 22 with an integrated analog-to-digital converter. The microcontroller 22 uses the signals S / H1, S / H2 and S / H3 to control the sampling times, which are also referred to as measurement impulses, at which the sample-hold circuit 21 holds current measurement signals until analog-to-digital conversion. The microcontroller 22 then only has to comprise a single analog-digital converter together with an analog multiplexer or switch 23. The advantage here is that simultaneous acquisition of current samples is achieved with only a single analog-digital converter, although a sequential analog-digital conversion takes place. In the microcontroller 22, current measurement values or a current space pointer are then determined from the current samples. The microcontroller 22 controls the sample-hold circuit and thus the measurement impulses via the signals S / H1, S / H2 and S / H3 in accordance with the method described below.

In another exemplary embodiment according to the invention, the sample hold circuit 21 can be saved in accordance with FIG. 3, in contrast to the variant in accordance with FIG. However, it is necessary to record the current samples one after the other in accordance with the method described below.

FIG. 4 shows exemplary time profiles of the motor-side output potentials U _R , U _s and averaged over a respective pulse width modulation period

U _{τ of} the converter is shown over an output voltage period, the motor-side output potentials U _R , U _s and U _τ are shown in a standardized manner and the output voltage angle extends over the range from 0 to 2π. According to the standardization, the potential value U _z _ corresponds to the value -1 and U _{z + to} the value +1. It can be seen in FIG. 4 that the mean values of the output potentials contain a third harmonic.

Furthermore, FIG. 4 shows angular ranges 1 to 6 of the output voltage vector, in which various means for current detection are used according to the method according to the invention.

With an output voltage angle of α = π / 6, the mean value of the output potential U _{s is} zero. The mean value of the output potential U _R is in the vicinity of the positive maximum value 1. The mean value of the output _potential U _τ is in the vicinity of the minimum value -1. With this output voltage angle and with the amplitude chosen as an example, the mean values of the output potentials almost reach the maximum drive of the converter.

In FIG. 5, for the output potentials from FIG. 4 there are again associated time profiles of the pulse width modulation signals PWM _R (t), PWM _s (t) at the output voltage angle α = π / 6.

PWM _τ (i) outlined. The control signals for the circuit breakers S _Ro , S _So , S _To , S _Ru , S _Su and S _r "are derived from these. If the respective pulse width modulation signal, for example PWM _R (t), is 1, the associated upper power _switch , for example S _Ro , is closed and the associated lower power _switch , for example S _Ru , is opened. In this case, the associated output voltage potential is U _R = U _{z +} . If the pulse width modulation signal is 0, the associated circuit breakers change into the respective other state and the associated output voltage potential is U _z _. The so-called dead time inserted in the practical implementation, during which the respective upper and lower switch is open, is in principle not relevant for the function of the invention. 5 shows two pulse width modulation periods in the abscissa direction.

The output voltage potential U _s = 0 averaged over a pulse width modulation period is achieved according to FIG. 5 by a PWM _s (t) signal which is 50% of the pulse width modulation period 1 and 50% of the pulse width modulation period 0. The Pulse width modulation ratios of the other pulse width modulation signals PWM _R (t) and

distributed according to their output voltage potential averaged over a pulse width modulation period.

FIG. 6 shows the exemplary motor currents I _R (t), I _s (t) and I _τ (t) associated with the two pulse width modulation periods mentioned in the output branches. In addition, the current ripple resulting from the pulse width modulated output potentials is indicated.

FIG. 7 shows the idealized temporal profile of the current measurement signals I _m (f), I _SM (Ϊ) ^{and <} ^ I _JM Q) associated with the two pulse width modulation periods without the filtering effect of the amplifier circuits V _R , V _s and V _τ , A non-zero

Current sample value always only results at those time periods in which the pulse width modulation signal of the associated half-bridge is zero; that is, the lower switch is closed. If the upper switch is closed, the motor current does not flow through the shunt resistor of the half-bridge, so that during this time the current measurement signal is zero regardless of the actual current. The current measurement signals therefore do not continuously represent the motor currents, but only represent them in those time periods in which the pulse width modulation signals PWM _R (t), PWM _s (t) nά PWM _τ (t) are zero. In the course shown, the current measurement signal I _m (t) differs from zero only for a very short period of time. With full control of the output voltage, this period would even completely disappear. It can therefore be seen from FIG. 7 that the determination of a current space pointer fails in such operating states if the current space pointer is formed using current measurement signals which differ only briefly from zero, as in FIG. 7, for example, I _m (t) that they the filter effect at the time of detection, i.e. when initiating the measurement, are significantly falsified. However, a minimum filter effect is necessary to suppress measurement noise and interference signals.

In the present invention, therefore, for each pulse width modulation period, the “optimal” pair that has the respectively wider pulse width modulation signals is selected from the three half bridges, ie only current measurement signals of these two special half bridges are used. The selection is made in such a way that the current measurement signal is not is used, the lower switch within a respective Pulse width modulation period has a shorter time the closed state than the other two lower switches.

FIG. 7 shows an operating state for two pulse width modulation periods in which the selected pair is I _SM (t) and I _m (t). For these, the times at which the current samples are recorded and which are also referred to as measurement impulses are indicated by jagged arrows.

Two basic types of methods can be used according to the invention.

For a first method, the lugs are drawn with solid, not jagged arrows. The current samples for the selected pair I _SM (t) and I _TM (t) are recorded simultaneously. Here are different

Available versions. In a first embodiment, each current sample is formed by a separate analog-to-digital converter. In a second embodiment, the three current measurement signals according to FIG. 2 are each fed to a sample-hold amplifier of the sample-hold circuit 21, the sample-hold amplifiers changing to the hold state at the time of the sampling. Thus, the analog-digital converter can convert the fixed current measurement signals one after the other by means of the multiplexer or switch. In a third embodiment, two analog-digital converters and corresponding changeover switches are used. In other versions, mixed forms are also possible.

In the first method, a symmetrical pulse width modulation method is advantageously used, the switching frequency not being selected to be higher than 20 kHz and the maximum modulation described above not being exceeded. Thus, there is always a pulse pattern within a pulse width modulation period, in which such a pulse width modulation signal occurs in a maximum of one output branch, the LOW state of which is shorter than the minimum duration. This means that both pulse width modulation signals are never LOW in two phases at the same time for less than the minimum duration. In the event that the LOW state of one of the pulse width modulation signals, for example PWM _R (t), falls below the minimum duration, it is nevertheless ensured that the current space pointer is correctly determined. Because the other two pulse width modulation signals, in the example PWM _s (t) and PWM _τ (t), do not fall below the minimum duration. Thus, correct determination is always in the present invention of the current space vector, provided that the respective type of pulse width modulation method ensures that only a maximum of one pulse width modulation signal stays in the LOW state for less than the minimum duration.

In a second method according to the invention, that in the middle is the

Pulse width modulation period measurement impulse for I _{m of} the first method according to the invention replaced by two measurement impulses, each with a time offset Δt. First there is the first measurement initiation for I _m , then that for I _SM and then the second measurement initiation for I _m . It is advantageous to place the measurement trigger for I _SM in the middle of the pulse width modulation period. The time offset .DELTA.t is as small as possible, but larger than the analog-digital converter time. The mean value is formed from the first and second current sample values for I _m and is used to determine the current space vector. This mean value formed in this way corresponds to that fictitious current sample value for I _m , which would be detectable at the same time for detecting the current sample value I _SM . The current space pointer determined in this way is thus free from the influence of the current ripple. This makes it possible to determine the current space vector with only a single analog-digital converter without being distorted by current ripples. Additional sample-hold circuits are not necessary. Additional sample-hold circuits are not to be understood as those sample-hold circuits that are an integral part of conventional analog-digital converters.

A symmetrical pulse width modulation method is also advantageously used in the second method, the switching frequency not being selected to be higher than 20 kHz and the maximum modulation described above not being exceeded. Thus, in turn, there is always a pulse pattern within a pulse width modulation period in which such a pulse width modulation signal occurs in a maximum of one output branch, the LOW state of which is shorter than the minimum duration. In the pulse width modulation method used in the second method according to the invention, there is always at least one pulse width modulation signal whose LOW state lasts longer than the sum of the minimum duration and twice the time offset, that is to say 2 * Δt. A correct determination of the current space vector is thus always ensured in the present invention, provided that the respective type of pulse width modulation method ensures that the pulse width modulation signals have a sufficient pulse width. Although flat-top methods are generally symmetrical pulse width modulation methods, they cannot be used advantageously for small output voltages.

In a further exemplary embodiment according to the invention, the current samples are not determined in the same pulse width modulation period, but the first current sample is recorded in a first pulse width modulation period and the second current sample in one of the subsequent pulse width modulation periods. Again, only a single analog-to-digital converter can be provided without an additional sample-hold circuit. This is particularly advantageous with a high switching frequency and thus short pulse width modulation periods. There is no distortion caused by current ripples. Only the change in the angle of the current space vector occurring during this time can cause falsifications. The pair is selected in accordance with FIG. 8.

In Figure 4, angular ranges 1 to 6 are shown, which correspond to different angular ranges of the output voltage pointer. In Figure 8 this is for everyone

Angular ranges the pair to be used indicated by jagged arrows. For example, only S and T, ie I _SM and I _m , should be used in angular range 1. In the angular range 2 only R and T, i.e. I _m and I _m , should be used

As described above, in the second method according to the invention, the measurement impulse symmetrically in the middle of the pulse width modulation period for a current sample value of the first method according to the invention is replaced by two measurement impulses which each have a time offset Δt. FIG. 9 shows how the measurement impulses are to be carried out as a function of the angular ranges of the output voltage space vector. In particular, it is shown in which half-bridge two measurement impulses and in which half-bridge the individual measurement imposition in the middle are to be carried out. In the angular range 1 there is therefore a first measurement initiation for I _m , then a single, central measurement initiation for I _SM and finally a second measurement initiation for I _m . The measurement impulses have a time interval of Δt.

In the angular range 3 there is therefore a first measurement impulse for I _m , then a single, central measurement impulse for I _m and finally a second measurement impulse for I _m . In other exemplary embodiments according to the invention, the two current samples of the current measurement signal I _{TM are} not recorded symmetrically around the measurement initiation for recording the current sample value of the current measurement signal I _SM , that is to say not with a time interval Δt before and after the detection of I _SM , but with different time intervals. Then, instead of the mean value, an interpolated value is formed, which takes into account the corresponding time intervals, whereby, however, variables that characterize the motor and also the type and duration of the switching states are to be taken into account when interpolating.

In other exemplary embodiments according to the invention, instead of the two current samples mentioned, more current samples are acquired. In addition, each current sample can in principle be replaced by several current samples. A further reduction in measurement noise can thus be provided.

In further exemplary embodiments according to the invention, the time behavior of the

Amplifier circuits V _R , V _s and ^ are taken into account in that all measurement impulses are delayed by the filter time constant of the amplifier circuits. The filter time constant is less than half the minimum duration.

A so-called symmetrical pulse width modulation method is advantageously used in the present invention. In the case of such symmetrical pulse width modulation methods, the temporal mean value, which is formed from the value of the time of a first switching state change of the pulse width modulation signal in a first phase and the value of the time of the subsequent switching state change associated in the same phase, has the same value as the correspondingly formed one Averages of the other two phases. Therefore, the switching state changes from HIGH to LOW and back in all three phases are symmetrical to the middle of the pulse width modulation period.

Claims

claims:

1. Method for determining a current space vector

for a converter which can be operated with pulse width modulation, comprising signal electronics and a power output stage which comprises circuit breakers arranged in three half bridges, each comprising a lower and an upper branch,

in which

a first and a second associated current sample value are provided with a second means of the pair before and after the acquisition of a current sample value acquired with a first means of the pair,

2. The method according to at least one of the preceding claims, characterized in that the pair is selected differently depending on the angle of the output voltage space vector.

3. The method according to at least one of the preceding claims, characterized in that the current samples are determined by means which are connected to the lower or to the upper intermediate circuit potential.

4. The method according to at least one of the preceding claims, characterized in that a reference potential of the signal electronics corresponds to a reference potential on which the current measurement signals lie and / or the current samples are recorded.

5. The method according to at least one of the preceding claims, characterized in that the detection of a current sample takes place centrally to that state of the pulse width modulation period in which the circuit breaker arranged in the same branch as the associated current measuring means is conductive.

6. The method according to at least one of claims 1 to 4, characterized in that the detection of a current sample is offset by the filter time constant to the middle of that state of the pulse width modulation period in which the circuit breaker arranged in the same branch as the associated current measuring means is conductive.

7. The method according to at least one of the preceding claims, characterized in that the second means of the pair is the means at which the longest conducting state of the three lower power switches occurs during the respective pulse width modulation period, provided the current measuring means in the three lower Branches are arranged and that the first average of the pair is the average at which the second longest conducting state of the three lower circuit breakers occurs during the respective pulse width modulation period.

8. The method according to at least one of claims 1 to 6, characterized in that the second means of the pair is the means at which the longest conducting state of the three upper power switches occurs during the respective pulse width modulation period, provided that the current measuring means in the three upper branches are arranged, and that the first average of the pair is the average at which the second longest conductive state occurs during the respective pulse width modulation period of the three upper circuit breakers.

9. The method according to at least one of the preceding claims, characterized in that the measurement impulses for detecting the current sampling values associated with an average within a pulse width modulation period each have the same time interval to the center of the

Pulse width modulation period are.

10. The method according to at least one of the preceding claims, characterized in that a current sample is detected more than once per pulse width modulation period with both means and / or more than one measurement trigger is provided within a pulse width modulation period.

11. The method according to at least one of the preceding claims, characterized in that a value interpolated according to the times of the respective detection and / or mean value is formed from the detected current samples as a current measured value for determining the current space vector.

12. converter, comprising signal electronics and a power output stage, which comprises power switches arranged in three half bridges, each comprising a lower branch and an upper branch,

wherein the converter can be operated with pulse width modulation,

characterized in that

- Means for detecting the respective currents are arranged either in all three lower or in all three upper branches of the half bridges and

13. Converter according to at least one of the preceding claims, characterized in that the means for current detection include resistors, in particular shunt resistors, and / or the means for current detection are arranged in the half-bridges such that they are either with the upper or with the lower DC link potential are connected.

14. Converter according to at least one of the preceding claims, characterized in that the signal electronics comprise means for generating pulse-width-modulated control signals for the circuit breakers.

15. Converter according to at least one of the preceding claims, characterized in that the signal electronics has a reference potential, which is also the reference potential for the means for current detection.

16. Method for determining a current space vector

in which

the pulse width modulation frequency is greater than a minimum frequency,

- Within a period of time, which is two pulse width modulation periods, with at least one of two of the three means for detecting the currents

Current sample value is determined and the same mean output voltage space vector is output in both pulse width modulation periods,

a first current sample in the middle of the first of the two pulse width modulation periods is acquired with a first mean of the pair and an associated current sample value is acquired in the middle of the second, ie immediately following pulse width modulation period with a second means of the pair,

- from the current samples determined with this pair of means

Current space pointer is formed or the current values are formed in the output branches, in particular for use in a control and / or regulating method, and

and the mean output voltage space vector is determined by the differences between the output potentials of the three output phases, which are averaged over a pulse width modulation period.

17. The method according to claim 16, characterized in that at least one of the methods according to claims 1 to 11 is carried out when the pulse width modulation frequency is selected to be lower than the minimum frequency and the method according to claim 16 is carried out when the pulse width modulation frequency is selected to be greater than the minimum frequency ,

18. The method according to at least one of the preceding claims, characterized in that the minimum frequency has a value between 8 and 16 kHz.