DE102011088242A1 - Method for operating induction machine e.g. permanent magnet synchronous machine (PMSM) of drive system, involves determining corrected default value for each phase for adjusting voltage vector at inverter based on preset error value - Google Patents

Abstract

The method involves setting a voltage vector of induction machine (4) through power switches (S1-S6) of inverter (1). The preset error value per phase (U,V,W) is determined corresponding to respective switch in order to compensate dead time period. The corrected default value for each phase is determined for adjusting the voltage vector at inverter based on determined preset error value per phase. The simulated phase current value is determined based on corrected default value and desired current requirements. Independent claims are included for the following: (1) device for operating induction machine; and (2) drive system.

Description

The present invention relates to a method for operating a rotating field machine according to the preamble of claim 1.

In an inverter, in particular a voltage source inverter, by a PWM modulation method, the desired voltages for a to be operated by means of the inverter induction machine, for. As a permanent-magnet synchronous machine (PMSM) can be set by means of generated PWM default values. The generated PWM default values from the controller or controller to produce a particular voltage vector will vary depending on the desired voltage frequency and amplitude.

As a semiconductor or current valves in the inverter MOSFETs, IGBTs, thyristors, GTOs, etc. are usually used. All have in common that the switching off is not sudden, but a certain time corresponding to a dead time t ₀ needed, ie until all charges are eliminated in the stopband of the semiconductor. This dead time depends on the semiconductors used. For example, this time is below 1μs for the MOSFET and between 1μs and 5μs for the IGBT, for example.

In the inverter, the two complementary semiconductors, ie high-side and low-side, must never be switched on at the same time in one phase. This can lead to a short circuit of the DC link and consequently to the destruction of the DC link capacitor and of the semiconductor. For this reason, a lock time or dead time t ₀ between switching off a semiconductor and switching on the complementary semiconductor in the same phase is generally taken into account.

When driving a machine, this dead time may result in a change in the intended voltage vector, resulting in distortions of the input voltages to the machine. In order to compensate for this dead time, it is proposed in the prior art, in particular in connection with control systems of rotary field machines, which measure the phase currents to establish a feedback in the control loop, to use the phase currents measured per phase. An accurate determination of the zero crossing is very important here, since the compensation otherwise leads to incorrect results.

In the context of drive systems which are like that according to 3 From a control structure for field-oriented control of the induction machine, in particular with adjustable dynamics make use, it is also known from a parallel application of the assignee of the present invention, to compensate for the dead time advantageously without a measurement of the phase currents. For this purpose, an expected performance of the induction machine is determined and set in relation to PWM default values, but some comparison operations are required.

Proceeding from this, the present invention has the object to provide an alternative method for operating a rotating field machine, which advantageously allows easy compensation of the dead time in the inverter also without a measurement of the phase currents and phase voltages.

This object is achieved by the features of claim 1. Advantageous embodiments and further developments are the subject of the dependent claims.

The invention proposes a method for operating a rotary field machine, in particular a permanent-magnet synchronous machine (PMSM) or an asynchronous machine, to an inverter, in particular a voltage source inverter. Such has a voltage intermediate circuit on the network side, which is preferably formed by means of a DC link capacitor. Furthermore, the inverter has power switch, z. As MOSFETs or IGBTs, which are arranged in half-bridges of an inverter bridge circuit as a high-side power switch or low-side power switch. Preferably, the inverter is formed by means of a B6 bridge circuit, which can be controlled by a driver stage, in particular of the inverter.

For operation of the induction machine - which in the context of the present invention, in particular a stator and a rotor - on the inverter, is provided on the induction machine voltage vector by means of the inverter by suitable switching positions of the circuit breaker of the same, in particular one voltage vector - preferably a space vector modulation Voltage vector - during each one switching period or PWM period. For this purpose, in the context of the method according to the invention, default values for each phase are determined, for. B. calculated. The default values For the purposes of the present invention, they may be voltage default values or preferably PWM default values, which express the latter duty cycles for selected switches of the half bridges, ie per phase, and for the inverter or its driver stage.

The inverter is generally operable by default values for the respective switching periods, which already implement the compensation of the dead time, z. B. in the context of PWM default values, which are generated by a regulation with current measurement.

Furthermore, the inverter and insofar an induction machine in the context of the present invention by means of default values for the respective switching periods can be operated, which - as largely idealized default values, eg. B. provided as part of a field-oriented control - still require a correction for dead time compensation and such can be provided below as corrected default values for the inverter. The corrected default values may here be corrected voltage default values or preferably corrected PWM default values. This will be explained in more detail below.

In order to compensate for a dead time of the semiconductor switches of the inverter, in particular in the context of a field-oriented control, the invention provides that corresponding to a respective switching period of the inverter, a default value error per phase is determined by sign and magnitude, which with the default value of the phase to provide a corrected default value the phase for setting the voltage vector is related. The corrected default value can then be provided to the inverter.

In the method according to the invention, the default value error per phase is determined based on a phase current value of the phase simulated by means of setpoint current specifications or current setpoint values, in particular by means of filtering of setpoint current specifications. By use by means of setpoint current specifications or current setpoints of simulated phase currents, d. H. of phase current values, which are simulated or correspond to those actual phase current values on / in the induction machine and which can be obtained in particular as a result of filtering the desired current specifications, can advantageously account for a current measurement of the phase currents on the induction machine to compensate for the dead time. This is extremely advantageous, in particular for field-oriented control.

In this case, it is provided to correct the default values as a function of their respective type for dead time compensation. Ie. In particular, voltage default values are related solely to voltage bias errors to determine corrected voltage bias values, PWM command values having PWM bias errors to determine corrected PWM bias values.

ermittelt werden. Der Spannungs-Vorgabewertfehler ∆U_1,2,3 gemäß der Gleichung 2:

It is further provided that the sign of the default value error per phase is determined from the sign of the replicated phase current value of the phase. Furthermore, it is provided that the amount of the default value error, the respective phase, in particular PWM default value error concerning based on the dead time, ie the dead time of the semiconductor switches of the inverter, and a period is determined, in the case of a voltage default value error, the intermediate circuit voltage can additionally be used become. The period corresponds in particular to a switching period of the inverter. The PWM default value error ΔPWM _1,2,3 per phase can be determined in particular according to Equation 1:

be determined. The voltage default value error ΔU _1,2,3 according to Equation 2:

In equations 1) and 2), sign (I _s1,2,3Ref ) _denotes the sign of the respective simulated phase _{current value} I _s1Ref , I _s2Ref , I _s3Ref , t ₀ the dead time and T _S the period _duration . U _DC designates the intermediate circuit voltage at the inverter.

After determining the respective default value error ΔPWM _1,2,3 or ΔU _s1,2,3 , this can be related to the corresponding default value for this phase to determine the corresponding corrected default value, by means of which the dead time compensation is implemented. in particular be added. Subsequently, the output of the corrected default value per phase can be output for or to the inverter for setting a voltage vector, wherein the dead time is now compensated.

In the context of the present invention, it is provided that a simulated phase current value is always provided by means of filtering the nominal current specifications with a low-pass filter, in particular a digital filter. In this case, the setpoint current specifications are provided, in particular, as discrete-time variables, for the filtering of which the low-pass filter is preferably suitable in the context of the present invention. The low-pass filter is designed, in particular, to act on the setpoint current specifications with a time delay, insofar preferably acts as a time-discrete PT1 element. The low-pass filter can be formed in a simple manner computerized.

By means of the low-pass filter, the dynamics of a field-oriented control or regulation, in particular a field-oriented control with adjustable dynamics can be emulated by any suitable choice of the time constant of the low-pass filter. The filtered nominal current specifications or the simulated phase current values obtained therefrom correspond in their course to the actual phase current values of the phases of the induction machine. Since in this respect the sign of a respective simulated phase current value corresponds to that of the actual machine current value or phase current, such an invention can advantageously be used for dead time compensation.

The setpoint current specifications for filtering are preferably provided in a d, q two-size system, in particular as time-discrete variables. As a result, an unobtrusive further processing is possible. After filtering, the now-filtered target current specifications can be transformed into a three-digit system for providing the simulated phase current values per phase. It is envisaged here that based on the setpoint current specifications or current setpoint values, the default values per phase are also determined.

Provided is within the scope of the present invention preferred that the induction machine is operated in a controlled manner during the process according to a field-oriented control, in particular with adjustable dynamics. The time constant of a filter can be selected corresponding to the dynamics, in particular depending on the set dynamics. In addition, the use of the method is also conceivable in controller structures which do not make use of a current measurement.

The method can be carried out by means of a device which is designed to carry out the method, in particular includes an inverter, and provides the corrected default values for operating the inverter. Such a device can be formed largely computerized, z. B. by means of at least one μControllers or processor, in particular as a control unit. Depending on the corrected default values, the inverter can drive a rotating field machine connected to this power side, which is preferably part of a drive system that can be formed by means of the device.

The device preferably has a control structure for field-oriented control, in particular for field-oriented control with adjustable dynamics, wherein the device provides corrected default values in the course of the field-oriented control. It is also conceivable to implement the control structure in a controller structure for field-oriented control, which is optionally switchable to the control structure.

It is envisaged in each case that the device has a filter as described above or its functionality for filtering the setpoint current specifications or for simulating the phase currents. The time constant of the filter is preferably set to model the dynamics of the controller.

It is also proposed a drive system with a rotating field machine and an inverter, wherein the drive system is designed for operation of the induction machine according to the method described above.

The method, which allows the operation of the induction machine without phase current sensor, is advantageously applicable especially at high speeds. The method can be carried out computerized, in particular continuously, z. B. by means of a microcontroller, for. A control or regulation of a device according to the invention, in particular e.g. same control or regulation, which also calculates the default values.

The proposed method, the device or the drive system are preferably used in a motor vehicle, in particular in a drive train of the motor vehicle. The hereby operated rotary field machine is preferably used for actuating an actuator of the drive train, ie for carrying out a control function of the drive train, for example, to perform gearshifts in a multi-gear vehicle transmission, for coupling / uncoupling drive shafts, for performing a startup of an internal combustion engine, etc. Also preferred the induction machine serves as a traction drive (drive motor) in the vehicle drive train, ie for generating a driving torque for propulsion of the motor vehicle. In the latter case, the motor vehicle is in particular an electric vehicle or an electric motor hybrid vehicle hybrid. However, the induction machine can also be used preferably for another actuator of the motor vehicle. In particular, the induction machine can be used in a vehicle steering system for electromotive steering power assistance (electric power steering) or for adjusting a suspension component, such as an electrically adjustable Fahrzeugfahrwerksstabilisators or vehicle suspension arm. However, the proposed method, the device or the drive system are not limited thereto. Thus, for example, the control of an induction machine of an elevator drive, a cable car drive, a machine tool drive or adjustment system or a wind turbine control system can be done.

Further features and advantages of the invention will become apparent from the following description of embodiments of the invention, with reference to the figures of the drawings, which show details essential to the invention, and from the claims. The individual features can be realized individually for themselves or for several in any combination in a variant of the invention.

Preferred embodiments of the invention are explained below with reference to the accompanying drawings. Show it:

1 an example of a voltage source inverter with an operable by this rotary field machine;

2 an example of an equivalent circuit diagram for the arrangement according to 1 ;

3 exemplarily a drive system with a control structure according to the prior art;

4 Exemplary and schematically a drive system with an apparatus for performing the method according to a possible embodiment of the present invention.

In the following description of the figures, the same elements or functions are provided with the same reference numerals. Corresponding elements or functions in the prior art are provided with primed reference numerals.

1 shows an example and schematically a pulse inverter 1 with a voltage intermediate circuit 2 , which by means of an intermediate circuit capacitor 3 is formed. The inverter 1 is formed in B6 bridge circuit with three half-bridges, each having two power switches S1, S2 and S3, S4, and S5, S6 in the form of MOSFETs. Three circuit breakers S1, S3, S5 of the inverter 1 are arranged as a high-side circuit breaker, three switches S2, S4, S6 as a low-side circuit breaker. The center taps of the half bridges are each with a phase U or V or W of a rotary field machine 4 in the form of a permanent magnet synchronous machine (PMSM) electrically connected. The induction machine 4 , which is intended for use in a motor vehicle, for. B. in a hybridized or electric drive train, alternatively z. As for a power steering, has a stator with a stator winding, in particular with the three winding strands U, V, W, which are preferably symmetrically wound and arranged offset by 120 degrees, and further in particular a rotor equipped with permanent magnets.

By means of the switches S1 ... S6 can via a driver stage, not shown, of the inverter 1 the usual in a space vector modulation voltage vectors on the induction machine 4 be set over the eight possible switching states of the circuit breaker S1 to S6 of the inverter 1 ,

A turn-on state of a half-bridge for a phase U, V, W can generally be a PWM default value, which in the context of 2 to 4 with PWM1 and PWM2 and PWM3, respectively, are controlled corresponding to a duty cycle per PWM period. 2 This illustrates an equivalent circuit diagram, in each of which two switches S1, S2 or S3, S4 or S5, S6 each a half-bridge as a Switch S _U , S _V , S _W according to the functionality of the inverter 1 are shown. Depending on a respective PWM default value for one of the phases U, V, W, each switch S _U , S _V , S _W between high-side and low-side potential for a certain duration per PWM period from -1 to +1 are switched (or vice versa), wherein the high-side potential is illustrated by the switch position +1, low-side potential by the position -1.

The inventive method can with a device 5 as well as a drive system 6 especially depending on 4 be performed. Visible here are compared to the prior art, sz B. 3 , advantageously only minor modifications required.

A preferred embodiment of the invention is described below with reference to 4 explained in more detail.

4 shows a control structure 7 , which by means of a device 5 of the drive system 6 is implemented. The device 5 or the drive system 6 point to the formation of the tax structure 7 suitable functional units, which are formed in particular by means of at least one μControllers, in particular integrated or alternatively z. B. separated from each other, which will be explained in more detail below. In addition, have device 5 or drive system 6 an inverter as described above 1 on, the drive system 6 furthermore a rotary field machine as described above 4 ,

i.e. basierend auf dt der Winkelinformation eines Positionsgebers 11, z.B. eines Inkrementalgebers oder eines Resolvers, als auch in Abhängigkeit der Polpaarzahl Zp auf bekannte Weise ermittelt, wozu eine entsprechend ausgebildete Funktionseinheit 12 vorgesehen ist. Der Eingang 10 ist z. B. mit einem Ausgang der zweiten Funktionseinheit 12 verbunden, z. B. mittels einer elektrischen Verbindungsleitung 13. The tax structure 7 the device 5 or the drive system 6 includes a functional unit 8th at the entrance nine Current setpoint values or setpoint current specifications I _sdRef , I _sqRef , which are provided in the d, q two-variable system as time-discrete variables. The functional unit 8th has another entrance 10 on, as an additional input variable, the actual electrical angular velocity ω _{el of} the rotor of the induction machine 4 is made available. The electrical angular velocity ω _el is based on the determined mechanical angle θ _mech, for example, by considering the temporal change of the mechanical angle θ _mech accordingly

ie based on dt the angle information of a position sensor 11 , For example, an incremental encoder or a resolver, as well as in dependence on the number of pole pairs Zp determined in a known manner, including a suitably trained functional unit 12 is provided. The entrance 10 is z. B. with an output of the second functional unit 12 connected, z. B. by means of an electrical connection line 13 ,

The functional unit 8th is intended, in particular taking into account the actual, determined electrical angular velocity ω _{el of} the rotor and the desired current specifications I _sdRef , I _{sqRef setpoint} voltage _components , ie a (stator) longitudinal voltage component or d-voltage component and a (stator) transverse voltage component or q- Determine the voltage component in the d, q-two-size system by means of respective FOS algorithms and provide them. The d, q two-size system denotes a coordinate system which rotates in a known manner with the rotor.

und Gleichung B) für die (Stator-)Sollquerspannungskomponente

As FOS algorithms, which z. B. in a permanent memory of the functional unit 8th are deposited or z. B. are loaded as needed in a volatile memory on which the functional unit 8th equations A) and B) can be used, which are solved in the respective current or to be performed sampling step, ie equation A) for the (stator) Solllängsspannungskomponente:

and equation B) for the (stator) desired lateral stress component

In the equations A) and B), the indexing _{_k} denotes the magnitudes of the current and _{_k-1} the magnitudes of the preceding scanning step of the duration T and _s the stator reference, ie a quantity relating to the stator of the induction machine 4 , U _{sd_k} and U _{sd_k-1} denote the desired longitudinal stress component, U _{sq_k} or U _{sq_k-1,} the target transverse stress component, ie each set for z. B. further functional units of the drive system 6 or for controlling the induction machine 4 ,

T _Ed denotes a time constant resulting from L _sd / R and L _sd the longitudinal component of the inductance of the stator winding, T _Eq a time constant resulting from L _sq / R and L _sq the _{transverse component of} the inductance of the stator winding, I _{sdRef_k} or I _{sdRef_k-1 is} the nominal longitudinal current specification, I _{sq.ref_k} or I _{sq.ref_k-1 is} the set cross current specification, R is the ohmic winding resistance of the stator winding, ω _{el is} the electrical angular velocity, and Ψω _{el is} the pole rotor or rotor voltage assumed to be constant for a particularly very short or short sampling step duration T, where Ψ represents the pole wheel flux. The pole wheel voltage is determined, for example, by means of stored characteristic curves.

The winding resistance R is z. B. deposited or in individual cases according to the respective induction machine 4 predetermined, z. B. also in a memory for the functional unit 8th deposited. The Polradspannung Ψω _el is assumed to be constant, since normally the mechanical time constant of the induction machine 4 is greater than the electrical time constant. The speed can be considered constant during the rapid change of the current.

The d, q current setpoints I _{sdRef_k-1} and I _{sqRef_k-1 of} the respective preceding sampling step required to solve the equations A) and B) in a current sampling step _{_k} are e.g. B. also in a memory for the functional unit 8th deposited. The deposit and access to the stored values are in particular by the functional unit 8th causes. The memory is z. B. a volatile or a RAM memory.

Using the FOS algorithms according to Equations A) and B), the field-oriented control (FOS) dynamics determined by the control structure 7 can be realized or made possible to be set optionally on the time constants T ₁ and T ₂ , ie by transmission elements. The time constants T _Ed and T _{Eq of} the (control) path are compensated here.

The determined _desired voltage _components U _{sd_k} , U _{sq_k} are at the output 14 the functional unit 8th provided, cf. 4 ,

The functional unit 8th in signal flow direction to the inverter 1 subordinate has the tax structure 7 furthermore a functional unit 15 which is intended to that of the functional unit 8th provided d, q-set voltage _components U _{sd_k} , U _{sq_k} in particular by means of an inverse Park and Clarke transformation, in target voltage values U _s1 , U _s2 , U _s3 a multi-phase three-phase system, in particular a three-phase three-phase system to convert, z. B. also within the current sampling step.

The functional unit 15 has an entrance 16 on, which with the exit 14 the functional unit 8th is in communication, for. B. via electrical connection lines. The functional unit 8th downstream functional unit 15 is designed in particular as a signal converter or as a transformation unit.

In order to determine the setpoint voltage values U _s1 , U _s2 , U _s3 in the multi-variable three-phase system, the functional unit is used 15 Also provided an angle information available, ie the actual electrical angle θ _{el of} the rotor resulting from the mechanical angle as a function of the number of pole pairs Z _p communicated. This size comes from another functional unit 17 , alternatively z. B. from the functional unit 12 , z. B. based on the angle information of the position sensor 11 ascertained in a known manner and z. B. via another input 18 the functional unit 15 provided with an output of the functional unit 17 communicates.

Based on that of the functional unit 15 at an exit 19 output nominal voltages U _s1 , U _s2 , U _s3 are then PWM default values PWM1, PWM2, PWM3 generated, which correspond to values with which the induction machine 4 would be operable in an idealized case of operation, ie idealized values that the dead time t ₀ in contrast to the z. B. generated in the course of a control PWM default values PWM1, PWM2, PWM3, however, do not take into account.

The conversion of the nominal voltage values U _s1 , U _s2 , U _s3 to PWM default values PWM1, PWM2, PWM3 is performed by a functional unit in the form of the PWM generator 20 the tax structure 7 , which is formed as a vector modulator or space vector pulse width modulator and which also has the DC voltage U _{dc of} the intermediate _circuit of the inverter 1 is supplied as a further input variable, ie to a entrance 21 , The nominal voltage values U _s1 , U _s2 , U _s3 are present at an input 22 the functional unit 20 On, the PWM default values PWM1, PWM2, PWM3 are at an output 23 to disposal.

With the exit 23 the functional unit 20 connected is the entrance 24 a functional unit 25 the tax structure 7 , which is provided for dead time compensation and based on the PWM default values PWM1, PWM2, PWM3 generates corrected PWM default values PWM1n, PWM2n, PWM3n for each phase U, V, W, by means of which the dead time is compensated. The corrected default values PWM1n, PWM2n, PWM3n, which respectively express duty cycles, are output 26 for controlling the inverter 1 provided, based on which the induction machine 4 operates suitably, ie adjusts the intended voltage vector via its power switches S1 to S6.

In order to be able to provide the corrected PWM default values PWM1n, PWM2n, PWM3n appropriately, the control structure has 7 a filter 27 on, in particular a digital filter, which is designed to apply the time-discrete target current specifications I _sdRef , I _sqRef in the course of filtering with a delay or to filter them. For this purpose, the filter acts as a low-pass filter, in particular as a PT1 element. The filter 27 are the set current specifications I _sdRef , I _sqRef , which at the input nine the functional unit 8th be tapped, at an entrance 28 provided.

The filter 27 has a time constant which is suitable, a respective dynamics of the controller or the device 5 or the drive system 6 reproduce, the time constant is set in particular according to the nachzubildenden dynamics. This makes it possible to _generate phase _{current values} I _s1Ref , I _s2Ref , I _s3Ref per phase U, V, W based on the setpoint current specifications I _sdRef , I _sqRef , which with regard to their magnitude and their sign reflect the actual machine currents or values per phase U, V, W correspond.

The filtered nominal current specifications I _sdRefF , I _sqRefF are at one output 29 provided as sizes in the d, q two-size system. To provide the phase currents of simulated phase current values per phase U, V, W, these are subsequently transformed back into the three-digit system. For this purpose, the tax structure 7 a functional unit 30 as it were the functional unit 15 for transformation on, at the entrance 31 the filtered nominal current specifications I _sdRefF , I _{sqRefF are} applied and which angle information required for the transformation, ie the electrical angle θ _el , via another, with an output 32 the transformation unit 15 connected input 33 receives.

At an exit 34 the functional unit 30 After transformation, the simulated phase _{current values} I _s1Ref , I _s2Ref , I _s3Ref per phase U, V, W are now provided, which at a further input 35 the functional unit 25 for determining the corrected PWM default values PWM1n, PWM2n, PWM3n.

Corresponding to a respective switching period of the inverter 1 determines the functional unit 25 for compensating the dead time t ₀ based on the phase _{current values} I _s1Ref , I _s2Ref , I _{s3Ref of} the phase U, V, W simulated by means of the set current specifications I _sRef , I _sqRef a default value error ΔPWM _1,2,3 per phase U, V, W after Sign and magnitude to relate to the respective default value PWM1, PWM2, PWM3 per phase U, V, W. Based on the determined PWM default value error ΔPWM _1,2,3 per phase and the default value PWM1, PWM2, PWM3 of the respective phase then the corrected default value PWM1n, PWM2n, PWM3n per phase to adjust the voltage vector at the inverter 1 determined.

wobei ∆PWM_1,2,3 den PWM-Vorgabewertfehler je Phase U, V, W, sign(I_s1,2,3Ref) das Vorzeichen des jeweiligen, insbesondere mittels der Filterung mit dem Filter 27 erhaltenen, nachgebildeten Phasenstromwerts I_s1Ref, I_s2Ref, I_s3Ref bezeichnet, und wobei t₀ die Totzeit und T_S die Schaltperiode des Wechselrichters 1 bezeichnet. To determine the respective default value error ΔPWM ₁ or ΔPWM ₂ or ΔPWM ₃ triggers the functional unit 25 for each phase U, V, W the equation 1):

where ΔPWM _1,2,3 the PWM default value error per phase U, V, W, sign (I _s1,2,3Ref ) the sign of each, in particular by means of filtering with the filter 27 The simulated phase _{current value} I _s1Ref , I _s2Ref , I _s3Ref is obtained, and t _{0 is} the dead time and T _{S is} the switching period of the inverter 1 designated.

By adding a respectively determined with respect to its magnitude and its sign PWM default value error ΔPWM _1,2,3 per phase with the corresponding PWM default value PWM1 or PWM2 or PWM3 is now for each phase U, V, W a new or corrected PWM default value PWM1n or PWM2n or PWM3n, which is for output to the inverter 1 for adjustment a voltage vector which takes into account the dead time t ₀ can be output or provided, ie at the output 26 , The corrected PWM default values PWM1n or PWM2n or PWM3n thus result PWM1n = PWM1 PWM + _1, PWM2n = PWM2 PWM + _2, and PWM3n = PWM3 + PWM. ₃

It should be noted that the induction machine 4 is operated controlled during the described method according to a field-oriented control, in particular with adjustable dynamics.

Within the scope of the present invention, an embodiment is also provided in which, based on the simulated phase _{current values} I _s1Ref , I _s2Ref , I _s3Ref, voltage default value _errors are determined in order to _correlate these with voltage default values for the output of corrected voltage default values. Based on the corrected voltage default values, PWM default values can then be generated which require no further correction for dead time compensation. For this purpose, instead of the functional unit 25 an alternative functional unit may be provided, which in signal flow direction to the inverter 1 in front of the functional unit 20 is arranged, ie upstream of the same. In this case, the voltage values Us1, Us2 and Us3 are used as voltage default values.

To output corrected voltage specification values Us1n, Us2n, Us3n by the alternative functional unit to the functional unit 20 taking account of the dead time t ₀ , the equation 2) for each phase U, V, W can be solved by the alternative functional unit, ie the equation:

Here ΔU _{s1,2,3 designates} the voltage default value _error , sign (I _s1,2,3Ref ) the sign of the respective, obtained in particular by means of the filtering, the simulated phase _{current value} I _s1Ref , I _s2Ref , I _s3Ref , and where t ₀ the Dead time, T _S, the period duration, which corresponds in particular with a switching period of the inverter, and U _DC denotes the _DC link voltage of the inverter, which is provided to the alternative functional unit suitably.

The determined voltage default value errors Us ₁ , Us ₂ , Us ₃ can in turn be related to the voltage default values Us1, Us2, Us3, so that the corrected voltage default values Us1n, Us2n, Us3n result in: Us1n Us1 = Us + _1, Us2n = Us2 + Us ₂ , and Us3n = Us3 + Us ₃ .

In the context of the present invention, a deadtime compensation is made possible without the use of current sensors, along with a reduced cost for sensors, A / D converter, board surface, z. B. for use in power steering systems. By eliminating current measurement, current measurement issues such as noise, offset, noise, and torque fluctuations can all be reduced. In particular, at low speeds, a dead time compensation in the inverter without current measurement can be performed along with increased accuracy. The method can be used with traction drives with inaccurate current measurement in the small current range and low speed range. Of course, further applications are conceivable.

LIST OF REFERENCE NUMBERS

1, 1 '

inverter

2

DC

3

Link capacitor

4, 4 '

Induction machine

5, 5 '

contraption

6, 6 '

drive system

7, 7 '

tax structure

8, 8 '

functional unit

9, 9 '

entrance 8th

10, 10 '

another entrance 8th

11, 11 '

position sensor

12, 12 '

functional unit

13, 13 '

connecting line

14, 14 '

output 8th

15, 15 '

functional unit

16, 16 '

entrance 15

17, 17 '

functional unit

18, 18 '

entrance 15

19, 19 '

output 15

20, 20 '

functional unit

21, 21 '

entrance 20

22, 22 '

another entrance 20

23, 23 '

output 20

24

entrance 25

25

functional unit

26

output 25

27

filter

28

entrance 27

29

output 27

30

functional unit

31

entrance 30

32

another exit 15

33

another entrance 30

34

output 30

35

another entrance 25

S1 ... S6

 breakers

AND MANY MORE

phases

Claims (16)

Method for operating a rotating field machine ( 4 ), in particular a permanent-magnet synchronous machine (PMSM) or an asynchronous machine, to an inverter ( 1 ), wherein the operation of the induction machine ( 4 ) Default values per phase (U, V, W) of the same are determined for setting a voltage vector on the induction machine ( 4 ) via the circuit breaker (S1, ... S6) of the inverter ( 1 ), corresponding to a respective switching period for compensating the dead time (t ₀ ), a default value error (ΔPWM _1,2,3 , ΔUs _1,2,3 ) per phase (U, V, W) is determined by sign and magnitude, which with the default value (PWM1, PWM2, PWM3; Us1, Us2, Us3) of the phase for providing a corrected default value (PWM1n, PWM2n, PWM3n; Us1n, Us2n, Us3n) of the phase (U, V, W) for setting the voltage vector in Relationship is set, characterized in that the default value error (ΔPWM _1,2,3 , ΔUs _1,2,3 ) per phase (U, V, W) based on a means of target current specifications (I _sdRef , I _sqRef ), in particular by means of filtering of set current specifications (I _sdRef , I _sqRef ), simulated phase _{current value} (I _s1Ref , I _s2Ref , I _s3Ref ) of the phase (U, V, W) is determined. Method according to Claim 1, characterized in that the sign of the standard value error (ΔPWM _1,2,3 ; ΔUs _1,2,3 ) per phase (U, V, W) is derived from the sign of the simulated phase _{current value} (I _s1Ref , I _s2Ref , I _s3Ref ) of the phase (U, V, W) is determined. Method according to claim 1 or 2, characterized in that the amount of the default value error (ΔPWM _1,2,3 ; ΔUs _1,2,3 ) of the respective phase (U, V, W) is based on the dead time (t ₀ ) and a Period (T _s ) is determined, wherein the period (T _s ) in particular with a switching period of the inverter ( 1 ) corresponds. Method according to one of the preceding claims, characterized in that a simulated phase _{current value} (I _s1Ref , I _s2Ref , I _s3Ref ) by means of filtering the nominal current specifications (I _sdRef , I _sqRef ) with a Low pass filter ( 27 ), in particular a digital filter, wherein the setpoint current specifications (I _sdRef , I _sqRef ) are provided in particular as time-discrete variables. Method according to one of the preceding claims, characterized in that the phase _{current values} simulated by the filtering (I _s1Ref , I _s2Ref , I _s3Ref ) _{correspond to} the actual phase _{current values of} the phases (U, V, W) of the _{induction machine} ( 4 ) correspond. Method according to one of claims 4 or 5, characterized in that the time constant of the filter ( 27 ) is adjustable, in particular corresponding to a dynamics of the control of the induction machine ( 4 ). Method according to one of the preceding claims, characterized in that a default value error (ΔPWM1,2,3) per phase (U, V, W) according to equation 1):

or according to equation 2):

where ΔPWM _{1,2,3 is} a PWM default value error and ΔU _{1,2,3 is} a voltage default value _error , sign (I _s1,2,3Ref ) the sign of the respective replica phase current value obtained, in particular, by the filtering (I _s1Ref , I _s2Ref , I _s3Ref ), and where t _{0 is} the dead time, T _S the period _duration , which in particular with a switching period of the inverter ( 1 ) and U _{DC is} the _DC link voltage of the inverter ( 1 ) designated. Method according to one of the preceding claims, characterized in that the default values (PWM1, PWM2, PWM3, Us1, Us2, Us3) are determined based on the desired current specifications (I _sdRef , I _SqRef ). Method according to one of the preceding claims, characterized in that the setpoint current specifications (I _sdRef , I _SqRef ) for simulating the phase _currents (I _s1Ref , I _s2Ref , _lS3Ref ), in particular for filtering, are provided in a d, q two-size system, and or the simulated phase _currents (I _s1Ref , I _s2Ref , I _s3Ref ) per phase (U, V, W) are provided in a three-size system. Method according to one of the preceding claims, characterized in that the induction machine ( 4 ) is operated in a controlled manner during the method according to a field-oriented control, in particular with adjustable dynamics. Method according to one of the preceding claims, characterized in that the corrected default value (PWM1n, PWM2n, PWM3n; Us1n, Us2n, Us3n) for each phase (U, V, W) for or to the inverter ( 1 ) is output for setting a voltage vector. Contraption ( 5 ) for the operation of a rotary field machine ( 4 ), in particular with an inverter ( 1 ), characterized in that the device ( 5 ) for carrying out the method according to one of the preceding claims, wherein for the operation of the inverter ( 1 ) corrected default values (PWM1n, PWM2n, PWM3n, Us1n, Us2n, Us3n) by the device ( 5 ) to be provided. Contraption ( 5 ) according to claim 12, characterized in that the device has a control structure ( 10 ) for field-oriented control, in particular for field-oriented control with adjustable dynamics, wherein the device ( 5 ) provides corrected values (PWM1n, PWM2n, PWM3n, Us1n, Us2n, Us3n) in the course of the field-oriented control. Contraption ( 5 ) according to one of claims 12 or 13, characterized in that the device is a filter ( 27 ), in particular a digital filter, further comprising in particular a low-pass filter for filtering the setpoint current specifications (I _sdRef , I _SqRef ). Drive system ( 6 ) with a rotating field machine ( 4 ) and an inverter ( 1 ), characterized in that the drive system ( 6 ) for the operation of the induction machine ( 4 ) is designed according to one of claims 1 to 11. Method according to one of claims 1 to 11 or device ( 5 ) according to one of claims 12 to 14 or drive system according to claim 15, characterized in that a default value is a PWM default value (PWM1, PWM2, PWM3) or a voltage default value (Us1, Us2, Us3), the default value error is a PWM default value error (ΔPWM _1,2,3 ) or a voltage default value error (ΔUs _1,2,3 ) and / or the corrected default value, a corrected PWM default value (PWM1n, PWM2n, PWM3n) or a corrected voltage default value (Us1n, Us2n, Us3n) is.

Images