Method for operating a converter to supply an electric machine and electric machine

Abstract

Method for operating a converter for supplying an electric machine and electric motor, where the motor current is recorded and an actual value of the motor current space indicator is calculated from the recorded values, wherein a current controller supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component, and in particular can be composed of these two current components, where a target torque value is specified, wherein the setpoint of the flux-generating current component is determined from the torque setpoint such that the magnitude of the stator current space vector is minimal at this torque setpoint, in particular with respect to the flux-generating current component and the torque-generating current component, and the setpoint of the flux-generating current component is subsequently corrected by means of a field weakening controller, in particular to maintain the voltage limit, in particular wherein the setpoint of the flux-generating current component corrected in this way is supplied to the current controller as a setpoint. wherein, for the corrected setpoint of the flux-generating current component, a setpoint of the torque-generating current component is determined such that the torque determinable from these current components and / or belonging to these current components equals the torque setpoint, in particular wherein the setpoint of the torque-generating current component determined in this way is supplied to the current controller as a setpoint.

Description

[0001] The invention relates to a method for operating a converter for supplying an electric machine and an electric machine.

[0002] It is generally known that an electric machine, such as IPMSM, i.e., Interior Permanent Magnetic Synchronous Motor, has a circumferentially anisotropic stator inductance and has buried permanent magnets, i.e., permanent magnets arranged radially within the rotor lamination stack.

[0003] From the seminal publication Bon-Ho Bae, Patel, N., Schulz S., Seumng-Ki Sul: “New field weakening technique for high saliency interior permanent magnet motor” in Industry Applications Conference, 2003, 38 thIAS Annual Meeting, Conference Record of the Vol. 2, pp. 898-905, 12-16 Oct. 2003, it is known to determine a setpoint for the torque-generating current component from a setpoint for torque and a maximum flux, using a 2D table. Independently of this determination, a setpoint for the flux-generating current component is determined from a setpoint for torque and a maximum flux, using a corresponding 2D table.

[0004] A method for controlling permanent magnet AC machines is known from DE 10 2008 013 799 A1.

[0005] From DE 10 2008 034 543 A1 a linear device for field weakening in a machine with an internal permanent magnet is known.

[0006] The invention is therefore based on the objective of achieving improved and extended control with simultaneously high torque and speed control range in an electric machine.

[0007] From "Electromagnets, Fundamentals, Calculation, Design" (ISBN: 978-3-8351-0138-8, Vieweg Teubner Verlag 2008, page 135) Fig. .3) The nonlinear behavior of the magnetic flux linkage Ψ in simple magnetic circuits is known. The nonlinearity and functional dependence of the linked magnetic flux on the current Ψ(I) is therefore also present in electric motors and must be taken into account when controlling these electric motors.

[0008] According to the invention, the problem is solved in the method according to claim 1, 2, or 3 and in the electric machine according to the features specified in claim 10.

[0009] Important features of the invention in the method for operating an inverter for supplying an electric machine according to claim 1 are that the motor current is detected and an actual value of the motor current space indicator is formed from the detected values, wherein a current controller supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component, and in particular can be composed of these two current components, where a target torque value is specified, wherein the setpoint of the flux-generating current component is determined from the torque setpoint such that the magnitude of the stator current space vector at this torque setpoint, in particular with respect to the flux-generating current component and the torque-generating current component, is minimal and thus efficiency-optimal, wherein an additional consideration of the machine speed to determine the loss-optimal setpoint currents with respect to the iron losses generated by the rotating field is also possible, and the setpoint of the flux-generating current component is subsequently corrected by means of a field weakening controller, in particular to maintain the voltage limit, in particular wherein the setpoint of the flux-generating current component corrected in this way is supplied to the current controller as a setpoint. wherein, for the corrected setpoint of the flux-generating current component, a setpoint of the torque-generating current component is determined such that the torque determinable from these current components and / or belonging to these current components equals the torque setpoint, in particular wherein the setpoint of the torque-generating current component determined in this way is supplied to the current controller as a setpoint.

[0010] An advantage of this is that, according to the invention, an improved, extended and in particular efficiency-optimized control can be achieved with simultaneously high torque and speed control range in an electric machine with pronounced anisotropic inductances and the resulting reluctance torque.

[0011] A further advantage is that the consideration of the voltage setpoint limit of the intermediate circuit by a controller, in particular a field weakening controller, in combination with a setpoint table for determining the torque-generating current component I, is possible. q I qThis is achieved by eliminating the need to consider the operating range of the electric machine, which depends on the voltage setpoint limit, when calculating the setpoint tables for optimal control. Instead, the machine-specific data is sufficient for calculating the setpoints. The advantage is twofold: firstly, the DC link voltage can vary and / or fluctuate due to the active field weakening control without requiring a recalculation of the tables. Secondly, the additional dimension of a voltage setpoint limit is eliminated when calculating the setpoint table, keeping the table size small and thus minimizing both the storage space required and the computation time needed for setpoint interpolation.

[0012] Furthermore, the implicit determination of the moment-generating current component I ensures q I q depending on the specified target torque M* M* and the flux-forming current component I dMaximum utilization of the machine's torque and speed control range is ensured by the upstream use of a torque limiter, guaranteeing that only valid setpoints within the possible operating range are set, depending on the available DC link voltage and / or maximum inverter current.

[0013] Another advantage of the closed-loop control structure is that no iterative steps are necessary to obtain an optimal setpoint for the underlying current control. The optimal setpoints are unique and can be determined by interpolation using a simple linear, particularly one-dimensional (1D) table, or, with a clever selection of the support points, a bilinear, particularly two-dimensional (2D) table. Therefore, no functional relationships need to be calculated multiple times during the control process, thus keeping the response time short.

[0014] By using a flow observer, it is still possible to replace some tables with runtime calculations. This further reduces the dependence on stored table setpoints.

[0015] Despite the reference to a permanent magnet synchronous machine with reluctance effects, the application of the control loop structure to other types of electric machines is conceivable. However, this requires consideration of the physical differences and the corresponding functional relationships for generating the setpoints for the underlying current control.

[0016] Important features of the method for operating a converter for supplying an electric machine according to claim 10 are that the motor current is detected and an actual value of the motor current space pointer is formed from the detected values, wherein a current controller supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component, and in particular can be composed of these two current components, where a target torque value is specified, wherein the setpoint of the flux-generating current component is determined from the torque setpoint according to a first functional relationship and, - in particular when the voltage setpoint limit is reached - is subsequently corrected by means of a field weakening controller, in particular wherein the setpoint of the flux-generating current component corrected in this way is supplied to the current controller as a setpoint, wherein the setpoint of the torque-generating current component is determined from the corrected setpoint of the flux-generating current component and from the torque setpoint according to a second functional relationship, in particular wherein the setpoint of the torque-generating current component determined in this way is supplied to the current controller as a setpoint, wherein the minimum of the stator current, which can be represented as a function of the torque and the flux dependent on the torque-generating and flux-generating current components, is determined depending on the given torque and the dependence of the flux-generating current component on the torque is used as the first functional relationship, where, from the torque determinable from the stator flux space vector and current space vector, the dependence of the torque-generating current component on the torque is determined as a second functional relationship when a flux-generating current component is present.

[0017] An advantageous feature of the invention is that the voltage setpoint limit is only reached at high rotational speeds, and thus the setpoint of the flux-generating current component is only corrected at these speeds, meaning the correction value does not disappear. This is because the DC link voltage is supplied to the field weakening controller, and the correction value is determined from this. According to the invention, the torque-generating current component is determined as a function of the respective flux-generating current component, whereby the torque setpoint is maintained precisely – even when the field weakening controller applies a correction value to the flux-generating current component. Another advantage of the invention is that the setpoint values ​​of the current components can be determined using stored tables, thus enabling a low controller response time.To determine a target value for the respective current component, one only needs to find a corresponding output value of the functional element for each input value using the table.

[0018] In an advantageous embodiment, the electric machine has a circumferentially anisotropic stator inductance. Advantageously, the invention is thus applicable to an electric machine with a pronounced reluctance effect. For example, the invention is applicable to a permanent magnet synchronous machine whose permanent magnets are embedded in the rotor lamination stack. Such a machine is also referred to as an IPMSM, i.e., "Interior Permanent Magnetic Synchronous Motor".

[0019] In an advantageous embodiment, permanent magnets are embedded in the lamination stack of the rotor of the electric machine and / or permanent magnets are arranged radially further inward than the outer diameter of the rotor. The advantage of this is that a pronounced reluctance effect occurs, and therefore the control according to the invention is very effective, especially at high speeds.

[0020] In an advantageous embodiment, a table discretized to represent the first functional relationship is stored in a memory, and the respective functional relationship is determined by interpolating the table values. in particular where the table was determined offline, especially before commissioning, and / or a table discretized to represent the second functional relationship is stored in memory and the respective functional relationship is determined by interpolating the table values, In particular, the table was determined offline, especially before commissioning. The advantage here is that fast control is achievable and only minimal computational effort is required to process the control.

[0021] In an advantageous embodiment, the field weakening controller is supplied with the magnitude of the voltage space vector determined by the current controller, and the deviation of the magnitude from a predetermined voltage setting limit, in particular from the intermediate circuit voltage value reduced by a predetermined factor, is determined. The field weakening controller, in particular a linear controller such as a PL controller or the like, determines the correction value from the deviation by which the setpoint of the flux-generating current component is corrected. An advantage of this is that a voltage setpoint limit can be specified, and if this limit is exceeded, the torque setpoint is limited. If there is a deviation between the specified voltage setpoint limit and the physically maximum voltage, in particular the DC link voltage, control fluctuations extending into the intermediate range are permitted, thus eliminating the need for a hard cutoff of the voltage setpoint range.

[0022] In an advantageous embodiment, to determine the second functional relationship, the stator flux space vector is represented as a function, in particular an integral function, of the torque-generating and flux-generating current components and numerically solved for the torque-generating current component. An advantage of this approach is that a table can be determined which is usable in operation and requires only minimal computational effort.

[0023] In an alternative advantageous embodiment, to determine the second functional relationship, an observer determines the stator flux space vector, in particular wherein - either the motor current and / or motor voltage values ​​recorded by the observer are supplied - or the observed with a magnetic flux-detecting sensor is provided with the measured values ​​of the components of the stator flux.

[0024] An advantage of this approach is that it enables an improved determination of the torque-generating current component. Here, the value used as stator flux in the control system corresponds as closely as possible to the actual physical value present in the machine.

[0025] According to the invention, the setpoint of the torque-generating current component is determined from the corrected setpoint of the flux-generating current component and from the stator flux space vector determined by an observer, in particular wherein - either the motor current and / or motor voltage values ​​recorded by the observer are supplied - or the observed with a magnetic flux-detecting sensor is provided with the measured values ​​of the components of the stator flux.

[0026] The advantage here is that the determination is as close as possible to the actual value of the physical quantity.

[0027] In an advantageous embodiment, the dependency applicable for determining the torque from the stator flux space vector and the current space vector is used M=m2⋅ZP⋅(Ψ→S×I→S) the dependence of the torque-generating current component on the torque setpoint, used in particular as the second functional relationship, is determined for a given flux-generating current component, whereby M M = Torque, especially internal torque, Z P = Number of pole pairs of the synchronous machine, Ψ S = Stator flux space indicator and I S = Stator current space pointer m = number of strands. An advantage of this is that a physically exact relationship is applied and maintained using the method according to the invention.

[0028] In an advantageous embodiment, the torque setpoint is determined from a torque preset value, which is limited to comply with the voltage setting limit and the current limit. in particular wherein, in order to comply with the current limit, the torque setpoint is limited to a first limiting value which corresponds to the torque maximum which can be generated at the maximum permissible stator current, in particular wherein the torque maximum is determined as the solution of an optimization problem in which the stator current is the maximum permissible value ILim=|ISmax→|ILim=ISmax achieved, in particular wherein, to comply with the voltage setpoint, the torque setpoint is limited to a second limit value which corresponds to the maximum torque that can be generated at the maximum permissible stator flux, wherein the maximum permissible stator flux is determined from the DC link voltage value and the rotational speed, in particular wherein the maximum torque is determined as the solution of an optimization problem in which the stator flux reaches the maximum permissible value: ΨLim=|ΨSmax→|

[0029] The advantage here is that the voltage and current limits are easily maintained by limiting the setpoint value. Unacceptably high setpoints are therefore not passed on to the current controller as target values, only permissible ones.

[0030] Key characteristics of the electric machine used to carry out the aforementioned method are that the electric machine has a circumferentially anisotropic stator inductance. An advantage of this is that effective control is possible even at high speeds.

[0031] In an advantageous embodiment, permanent magnets are embedded in the lamination stack of the rotor of the electric machine and / or permanent magnets are arranged radially further inward than the outer diameter of the rotor. An advantage of this is that reluctance values ​​that can be varied in the circumferential direction can be generated.

[0032] In an advantageous embodiment, the electric machine is an IPMSM. The advantage here is that the invention provides a particularly effective control system for such a machine.

[0033] Further advantages arise from the dependent claims. The invention is not limited to the combination of features of the claims. For those skilled in the art, further meaningful combinations of claims and / or individual claim features and / or features of the description and / or the figures will become apparent, in particular from the problem statement and / or the problem arising from a comparison with the prior art.

[0034] The invention will now be explained in more detail with the help of illustrations: In the Fig. Figure 1 shows a first embodiment of the invention in which, based on offline calculated look-up tables in function blocks 1 and 2, a setpoint current for a current controller is determined from a torque setpoint M*, which provides a voltage with which an electric machine, in particular an IPMSM, can be supplied. The voltage to be provided is generated via an inverter whose power switches receive pulse-width modulated control signals. In the Fig. 2 shows a second embodiment of the invention, in which, unlike the Fig. 1. The current stator flux is determined by an observer. In the Fig. Figure 3 shows a third embodiment of the invention, in which, unlike the Fig. 1 and Fig. 2. A torque setpoint is determined from a torque specification value depending on the current limit and voltage setpoint.

[0035] In three exemplary embodiments of the invention, a flux-generating current component is determined from a torque setpoint using a table preferably determined offline and stored in memory. The table values ​​are calculated using an optimization function under the condition of a minimum current value per torque and are thus designated as the optimal setpoint current values, representing the respective optimal operating point of the machine. The flux-generating current component is adjusted by a field weakening controller when the control limit, i.e., the DC link voltage of the inverter, is reached. In another exemplary embodiment, the speed value of the electric motor is used to determine the flux-generating current component in addition to the torque setpoint. A corresponding setpoint for the torque-generating current component is determined for the flux-generating current component thus determined by enforcing compliance with the torque setpoint.The torque setpoint is limited taking into account the possible inverter voltage and the maximum possible inverter current. For this purpose, a table, preferably determined offline, is stored in memory or an observer is used.

[0036] A key advantage of the invention is that, despite the field weakening controller's intervention on the offline-determined flux-generating current component, the torque-generating current component is always determined in such a way as to achieve the permissible target torque. This results, on the one hand, in very high utilization of the speed and torque control range of the electric motor. On the other hand, the control loop structure is independent of specific inverter parameters, such as DC link voltage and maximum inverter current, and therefore is not dependent on a particular inverter topology. Another important advantage of the invention is that the control structure shown in the figures allows for a very small table size for the previously offline calculated operating points, thus keeping the control computation time short.

[0037] The machine is supplied with three-phase current, with the control variable determined by the current controller being fed to a pulse-width modulated inverter stage, which is supplied from a DC link. The machine is powered from the inverter's output voltage. The current controller thus adjusts the motor voltage so that the detected stator current vector of the motor is regulated to match the target stator current vector.

[0038] The stator current of the motor is detected by means of a suitable current sensor, and the stator current space vector formed from the detected measured values ​​is transformed, for example, into a rotor flux-oriented coordinate system with a torque-generating current component I by means of a space vector transformation. q I q and flow-forming current component I d I q transferred.

[0039] In the first embodiment according to Fig. 1. The target torque M* is supplied as an input variable to function block 1 and is determined by means of a functional dependency. Id*=f(M*)Id*=f(M*) into a field-forming target current ID*Id* I d The functional dependency is determined as an interpolation of the values ​​stored in a table, especially a look-up table, i.e., in discrete representation.

[0040] To determine the table values, the solution to the optimization problem is calculated for each torque value M*. |I→S|→min! under the equation constraints M=M* calculated. In full, this expression is the functional minimum of the stator current magnitude. |I→S| subject to the constraint of a constant target torque M*.

[0041] This formulation of the optimization problem is described below using a Lagrangian function. The objective function, i.e., the function for which a minimum or maximum is to be calculated, is f(x). n ) with n as the numerator of the variable quantities, in the above case reads |I→S|=Id2+Iq2 Is2=Id2+Iq2 f(xn)=d(Id,Iq)=Id2+Iq2

[0042] For this expression, according to the above condition |I s |→ min! |I s | → min! a local minimum is sought while satisfying the equation constraint g(x) n ) = 0 g(xn)=g(Id,Iq)=M*−m2⋅ZP(Ψd(Id,Iq)⋅Iq−Ψq(Id,Iq)⋅Id)=0

[0043] Accordingly, the flux components that occur are Ψ d (I d ,I g ) and Ψ(I d ,I q) itself depends on the flow in each spatial direction. This means that the nonlinear flow patterns of the two spatial directions d and q, which are known to be true according to the state of the art, are taken into account when calculating the optimization problem.

[0044] The Lagrange function is then: L(Id,Iq,λ)=Id2+Iq2+λ⋅[M*−m2⋅ZPΨd(Id,Iq)⋅Iq−Ψq(Id,Iq)⋅Id]

[0045] The operator λ is called the Lagrange multiplier. For each torque value M = M*, a corresponding stator current space vector results at the minimum of the stator current magnitude. I→s(M*) with longitudinal component I d (M*), real part, and I q (M*), imaginary part, which represents the optimal operating points of the motor. According to the invention, only the dependence I of this is taken into account. d I d of M but not the dependence I q I qused by MM. Thus, a first table for function block 1 is determined numerically, where I d This depends on the mass. Due to the consideration of the nonlinear flux profiles Ψ(I) Ψ(I), a simple analytical solution of the Lagrangian is not possible. This is done numerically with suitable programs offline, i.e., before commissioning the drive system. During operation, only minimal computational effort is required, as the values ​​only need to be retrieved from the look-up table and interpolated if necessary.

[0046] According to the invention, the value determined in functional block 1 is ID* corrected by a field weakening controller path to maintain the voltage setpoint limit, wherein the setpoint supplied to a current controller 3 for the flux-generating current component ID**Id** deviates from the optimal target current shown in the tables in function block 1 Id*Id*. The deviation is determined by the field weakening controller to maintain the voltage setting limit. This is because the voltage to be set by current controller 3 must not exceed the DC link voltage. Therefore, the magnitude of the deviation is derived from the voltage space vector determined by current controller 3. |U→S| The signal is generated and fed to the field weakening controller 4, which is preferably designed as a linear controller, in particular with a PI controller element. The input to the field weakening controller 4 is therefore the magnitude of the signal. |U→S| supplied and the deviation from the maximum value of the DC link voltage of the inverter U, modified by a control factor of 5 z U z determined. The setpoint U supplied to the field weakening controller. Smax U Smax is calculated according to: USmax=Uz⋅2π⋅mMod⋅k, where the proportion 2π⋅mMod The factor k is determined depending on the selected modulation type. It sets the margin to the actual voltage limit and must be chosen between k = 0.1...1, i.e., 10-100% k = 0.1...1.

[0047] The current value provided on the output side by the field weakening controller 4 IdFSW*IdFSW* It may only sensibly take negative values ​​and is set to a value between by function block 6. IdFSWMax≤IdFSW*≤0 It is limited and therefore 0 or negative. It is measured at the summation point. ID*Id* added, thereby increasing the setpoint for the flux-generating current component, which is adjusted to the setpoint limit. ID** is determined. The field weakening controller actually restricts the valid solution range of the functional relationship in function block 1. Id*(M*)Id*(M*) a.

[0048] The effect is therefore equivalent to an additional constraint on the optimization function as a voltage control limit. For this reason, the changed current component ID**Id** to continue to be considered optimal, since it represents the solution for a valid solution domain of the optimization function.

[0049] The voltage space vector determined by the current regulator 3 is fed to the final stage 8 after the coordinate transformation 7, which supplies the electric machine, i.e. the IPMSM.

[0050] In the case of the electric machine, i.e. the IPMSM, it is important that the stator inductance is anisotropic in the circumferential direction and that permanent magnets are buried in the laminated core of the rotor, i.e. radially further inwards than the outer diameter of the rotor.

[0051] The field weakening controller 4 controls from the point at which the voltage limit U is reached. Smax = f(U z ) (U Smax = f (U z)) an additional contribution IdFSW* to the field-forming current ID*Id* This achieves a parameter-independent field weakening.

[0052] The stator-oriented stator current space vector, detected by means of at least one current sensor I→SS After the coordinate transformation 9, the angle is changed by the transformation angle Θ. f Θ f into a flux-oriented stator current space vector I→Sf transformed and fed to the current controller 3 as an actual value, so that this actual value corresponds to the setpoint of the flux-oriented stator current space vector, which the components ID** and Iq* exhibits, is regulated.

[0053] Coordinate transformations 7 and 8 are inverses of each other.

[0054] The second table, i.e., the table effective in function block 2, determines the torque-generating target current component. Iq* as a function of the target torque M* M* and the flux-forming current component ID** Thus, the flux-forming current component influenced by the field weakening controller 4 is ID** a moment-forming component Iq* assigned so that the specified torque setpoint M* M* is maintained.

[0055] For this purpose, the implicit equation is calculated offline, i.e., before operating the electric machine. M*−m2⋅ZP(Ψd(Id,Iq)⋅Iq−Ψq(Id,Iq)⋅Id)=0 according to the target current Iq=Iq* I q resolved, where Z P is the number of pole pairs of the electric motor and M = M* is the target torque. Iq*=2⋅M*m⋅ZP⋅Ψd(Id**,Iq*)+Ψq(Id**,Iq*)Ψd(Id**,Iq*)⋅Id**

[0056] The solution of the equation for discrete values ​​of M* and ID** This also does not occur at runtime, as the nonlinear flux relationships of the motor are also taken into account.

[0057] In summary, the following are therefore in the Fig. One further element of a field-oriented control system is described: - a subordinate current control (3) - a coordinate transformation T(x) T(x) (7) and T(x) -1 (8) - an inverter (9) - a transformation angle Θ f - a stator current pointer ◯ in stator current space vector in stator coordinates I→SSISS ◯ in stator current space vector in flux coordinates I→Sf - a rotating field machine in the embodiment of a permanent magnet synchronous machine with buried magnets (IPMSM) - a voltage U Z at the intermediate circuit capacitor - Conversion of the intermediate circuit voltage U Z to a maximum permissible stator voltage U Smaxdepending on the selected modulation type, whereby a voltage reserve of the underlying current regulator can be specified. - The output voltage magnitude of the current regulator |U→S||US|

[0058] In the embodiment according to the invention according to Fig. 2 is different from the Fig. In functional block 1, the current speed of the electric motor is supplied as an input variable, which is determined by a suitable measuring device or by sensorless methods. The functional relationship for determining the optimal target current ID*Id* The text then reads: Id(M*,N*)

[0059] The additional information about the motor speed is needed when determining the flux-forming current component. ID* the optimization problem: |PGes|→min! under the equation constraints M=M* is set up. Whereby PGes P ges The total losses of the electric motor are and result from the electrical losses P el , and the speed-dependent iron losses of the motor sheet P Fe P Fe The objective function seeks the minimum of total losses. f(Id,Iq)=Pel(Id,Iq)+PFe(Id,Iq,N*) The equation constraint remains: g(Id,Iq)=M*−m2⋅ZP(Ψd(Id,Iq)⋅Iq−Ψq(Id,Iq)⋅Id)=0

[0060] The optimization problem is solved numerically for a discretized set of target torque M* M* and speed N* N. The consideration of iron losses is to be decided based on the motor-specific loss characteristics.

[0061] In the embodiment according to the invention according to Fig. 3 is different from the Fig. In function block 2, the observed flux linkage is additionally fed in, thereby reducing the above implicit torque equation to: M−m2⋅ZP(Ψd_obs⋅Iq−Ψq_obs⋅Id)=0 which analytically determines the target current Iq*Iq* can be changed. Iq*=2⋅M*m⋅ZP⋅Ψd_obs+Ψq_obsΨd_obs⋅Id**

[0062] Using this dependency, the target current is determined in block 2. Iq* calculated for the underlying system, where observer 8 determines the components of the flux Ψ q_obs and Ψ d_obs This is determined, for example, by direct acquisition or by modeling from the acquired voltage and current values. This measure reduces the storage and / or computing power requirements of the embedded system. Furthermore, the parameter dependency of the method decreases.

[0063] In the embodiment according to the invention according to Fig. 4 is different from the Fig. 1. Additionally, the target torque M* M* is limited so that the current limit and voltage set limit are observed:

[0064] For this purpose, the torque setpoint is first supplied to the limiting element 12, whereby the limiting value is determined by the functional element 11 depending on the current limit. ILim=|ISmax→|, therefore, depending on the maximum permissible stator current of the motor and / or inverter current, the limiting element 12 is specified.

[0065] The dependency is represented as a table, which is calculated offline. This involves the optimization problem. M*→max! under the equation constraints |I→S|=ILim

[0066] In this case, the objective function seeks the maximum, hence the negative sign, of the torque. f(M)=−m2⋅ZP(Ψd(Id,Iq)⋅Iq−Ψq(Id,Iq)⋅Id)=f(Id,Iq) for a given current limit: g(Id,Iq)=ILim2−Id2−Iq2

[0067] The torque setpoint is limited by the limiting element 12 depending on the current limit. Mbegrl*Mbegrl* is used as the input value of the limiting element 13 and is additionally limited there to the value to maintain the voltage setting limit: MLim=f(ΨLim) , where the function f is again a table and the intermediate quantity Ψ Lim Ψ Lim in Division Block 15 according to ΨLim=USmax(UZ)ωel is determined, whereby the maximum stator voltage on the input side to functional block 14 is determined according to USmax=UZ⋅2π⋅mMod⋅k is calculated in the P-term 5. Thus, in the Fig. 3 lower path (13, 14, 15) a limit value is determined which limits the torque setpoint by means of the limiting element 13 and thus the voltage setpoint limit is maintained.

[0068] The table used in function block 16 is also determined offline by calculating the maximum torque through the solution of the optimization problem. M*→max! under the constraints |Ψ→S|=ΨLim The calculation is performed so that the table assigns a stator flux value to each torque value.

[0069] The objective function is the same as the objective function of the optimization problem in function block 11. The equation constraint is: |Ψ→S|=Ψd2(Id,Iq)+Ψq2(Id,Iq) ΨS2=Ψd2(Id,Iq)+Ψq2(Id,Iq) g(Id,Iq)=ΨLim2−Ψd2(Id,Iq)−Ψq2(Id,Iq)

[0070] The tables stored in the function blocks of the exemplary implementations are each defined as discrete support points for an interpolation function. This allows for the determination of values ​​corresponding to intermediate values. Depending on the dimension (1D, 2D), the table values, i.e., support points, are interpolated linearly or bilinearly.

[0071] The optimization problems described are preferably solved using numerical solvers, in particular Matlab optimization toolbox, Acado toolkit, or the like, generally offline.

[0072] In an alternative embodiment according to the invention, the solution can also be implemented online, i.e., in controlled operation, provided that the computer used has sufficient computing power. Reference symbol list 1 Functional block I d (M*) or I d (M*,N*) I d (M*) 2 Function block Iq(M*,Id**)Iq(M*,Id) 3 current regulators 4 field weakening regulators 5 Function block 6 Limitation 7 Coordinate transformation 8 inverse coordinate transformation 9 inverters 10 observers 11 Function block M(I) M(I) 12 Limiter element 13 Limiter element 14 function block, reserve factor m Mod · km mod · k 15 Function block, stator flux calculation 16 Function block M(Ψ) M(Ψ) Field weakening controller Magnitude of the stator current space pointer Magnitude of the stator voltage space vector magnitude of the stator flux space pointer M torque M* Target torque limited target torque depending on the stator current limited target torque depending on the stator flux Current setpoint of the longitudinal axis (output of the control condition) Current setpoint of the longitudinal axis (actuator value of the FSW controller) FSW Field Weakening total current setpoint along the longitudinal axis, Id**=Id*+IdFSW* total current setpoint of the transverse axis I Lim Maximum permissible inverter and / or stator current of the motor m number of strands of the electric machine Z P Number of pole pairs of the electric machine Ψ d (I d ,I q Current-dependent, non-linear flux linkage in the longitudinal axis Ψ q (I d,I q Current-dependent, non-linear flux linkage in the transverse axis Θ f Transformation angle Ψ d_obs observed flow along the longitudinal axis Ψ q_obs observed flow in transverse axis ω el Electrical circulation frequency (ωel=N⋅2⋅π⋅ZP60) N mechanical speed in rpm U Smax maximum possible stator voltage U Z Inverter intermediate circuit voltage m Mod Maximum modulation level k Reserve factor (0.1 ... 1)

Claims

[1] Method for operating a converter to supply an electric machine, where the motor current is recorded and an actual value of the motor current space indicator is calculated from the recorded values, wherein a current controller (3) supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component and can be composed of these two current components, where a target torque value is specified, characterized by , that The setpoint of the flux-generating current component is determined from the torque setpoint such that the magnitude of the stator current space vector is minimal at this torque setpoint with respect to the flux-generating current component and the torque-generating current component, and the setpoint of the flux-generating current component is subsequently corrected additively when the voltage setpoint limit is reached by summing a current value provided on the output side by a field weakening controller (4). IdFSW*, to maintain the voltage setpoint limit, wherein the setpoint of the flux-generating current component corrected in this way is supplied to the current controller (3) as the setpoint, wherein, for the corrected setpoint of the flux-generating current component, a setpoint of the torque-generating current component is determined such that the torque determinable from and / or associated with these current components equals the torque setpoint, wherein the setpoint of the torque-generating current component determined in this way is supplied to the current controller (3) as a setpoint, wherein the setpoint of the torque-generating current component is determined from the corrected setpoint of the flux-generating current component and from the stator flux space vector determined by an observer (10), wherein - either the motor current and / or motor voltage values ​​recorded by the observer (10) are supplied - or the observed (10) with a magnetic flux-detecting sensor is supplied with the detected values ​​of the components of the stator flux. [2] Method for operating an inverter to supply an electric machine, where the motor current is recorded and an actual value of the motor current space indicator is calculated from the recorded values, wherein a current controller (3) supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component, and can be composed of these two current components, where a target torque value is specified, characterized by , that The optimal setpoint of the flux-generating current component is determined from the torque setpoint and the current speed of the electric motor such that the sum of the total losses, consisting of electrical losses and iron losses, is minimal with respect to the flux-generating current component and the torque-generating current component at this torque setpoint and the current speed of the machine, and the setpoint of the flux-generating current component is subsequently corrected additively when the voltage setpoint limit is reached by summing a current value provided on the output side by a field weakening controller (4). IdFSW*, to maintain the voltage setpoint limit, wherein the setpoint of the flux-generating current component corrected in this way is supplied to the current controller (3) as the setpoint, wherein, for the corrected setpoint of the flux-generating current component, a setpoint of the torque-generating current component is determined such that the torque determinable from and / or associated with these current components equals the torque setpoint, wherein the setpoint of the torque-generating current component determined in this way is also optimally supplied to the current controller (3) as a setpoint, wherein the setpoint of the torque-generating current component is determined from the corrected setpoint of the flux-generating current component and from the stator flux space vector determined by an observer (10), wherein - either the motor current and / or motor voltage values ​​recorded by the observer (10) are supplied - or the observed (10) with a magnetic flux-detecting sensor is supplied with the detected values ​​of the components of the stator flux. [3] Method for operating an inverter to supply an electric machine, where the motor current is recorded and an actual value of the motor current space indicator is calculated from the recorded values, wherein a current controller (3) supplies the electric machine with such a voltage space vector that the actual value of the motor current space vector is regulated to a setpoint value of the motor current space vector, wherein the setpoint of the motor current space indicator has a flux-forming current component and a torque-forming current component, and can be composed of these two current components, where a target torque value is specified, characterized by , that The setpoint of the flux-generating current component is determined from the torque setpoint according to a first functional relationship and, - when the voltage control limit is reached - is subsequently corrected additively by summing a current value provided on the output side by a field weakening controller (4). IdFSW*, wherein the setpoint of the flux-forming current component corrected in this way is supplied to the current controller (3) as the setpoint, wherein the setpoint of the torque-generating current component is determined from the thus corrected setpoint of the flux-generating current component and from the torque setpoint according to a second functional relationship, wherein the setpoint of the torque-generating current component determined in this way is supplied to the current controller (3) as a setpoint, wherein the minimum of the stator current, which can be represented as a function of the torque and the flux dependent on the torque-generating and flux-generating current components, is determined depending on the given torque and the dependence of the flux-generating current component on the torque is used as the first functional relationship, where, from the torque determinable from the stator flux space vector and current space vector, the dependence of the torque-generating current component on the torque is determined as a second functional relationship when a flux-generating current component is present. wherein the setpoint of the torque-generating current component is determined from the corrected setpoint of the flux-generating current component and from the stator flux space vector determined by an observer (10), wherein - either the motor current and / or motor voltage values ​​recorded by the observer (10) are supplied - or the observed (10) with a magnetic flux-detecting sensor is supplied with the detected values ​​of the components of the stator flux. [4] Method according to one of claims 1, 2 or 3, characterized by , that the electric machine has a circumferentially anisotropic stator inductance and / or that Permanent magnets are buried in the laminated core of the rotor of the electric machine and / or permanent magnets are arranged radially further inwards than the outer diameter of the rotor. [5] Method according to at least one of the preceding claims, characterized by , that A table discretized to represent the first functional relationship is stored in a memory, and the respective functional relationship is determined by interpolating the table values. in particular where the table was determined offline, especially before commissioning, and / or that A table discretized to represent the second functional relationship is stored in memory, and the respective functional relationship is determined by interpolating the table values. in particular where the table was determined offline, especially before commissioning. [6] Method according to at least one of the preceding claims, characterized by , that the field weakening controller (4) is supplied with the magnitude of the voltage space vector determined by the current controller (3) and the deviation of the magnitude from a predetermined voltage setting limit, in particular from the intermediate circuit voltage value reduced by a predetermined factor, is determined. wherein the field weakening controller (4), in particular a linear controller such as a PL controller or the like, determines from the deviation the correction value by which the setpoint of the flux-forming current component is corrected. [7] Method according to at least one of the preceding claims, characterized by , that to determine the second functional relationship, the stator flux space vector is represented as a function, in particular an integral function, of the torque-generating and flux-generating current component and is numerically solved for the torque-generating current component. or that To determine the second functional relationship, an observer (10) determines the stator flux space vector, in particular wherein - either the motor current and / or motor voltage values ​​recorded by the observer (10) are supplied - or the observed (10) with a magnetic flux-detecting sensor is supplied with the detected values ​​of the components of the stator flux. [8] Method according to at least one of the preceding claims, characterized by , that from the dependency applicable for determining the torque from the stator flux space vector and the current space vector M=m2⋅ZP⋅(ψs_×Is_) the dependence of the torque-generating current component on the torque setpoint, used in particular as the second functional relationship, is determined for a given flux-generating current component. where M Torque, especially internal torque, Z P Number of pole pairs of the synchronous machine Ψ S Stator flux space pointer and I S Stator current space indicator m strand number is. [9] Method according to at least one of the preceding claims, characterized by , that The torque setpoint is determined from a torque target value, which is limited to comply with the voltage setting limit and the current limit. in particular wherein, in order to comply with the current limit, the torque setpoint is limited to a first limiting value which corresponds to the torque maximum that can be generated at the maximum permissible stator current, in particular wherein the torque maximum is determined as the solution of an optimization problem in which the Stator current magnitude to the maximum permissible value ILim=|ISmax→| reached, in particular wherein, in order to comply with the voltage setpoint, the torque setpoint is limited to a second limit value which corresponds to the maximum torque that can be generated at the maximum permissible stator flux, wherein the maximum permissible stator flux is determined from the DC link voltage value and the rotational speed, in particular wherein the maximum torque is determined as the solution of an optimization problem in which the stator flux equals the maximum permissible value Ψ Lim reached, where I Lim the maximum permissible output current of the inverter and / or the maximum permissible stator current of the motor is, where Ψ Lim is specified, whereby |ISmax→| the maximum permissible stator current. [10] Electric machine for carrying out a method according to at least one of the preceding claims, characterized by , that the electric machine has a circumferentially anisotropic stator inductance and / or that Permanent magnets are buried in the laminated core of the rotor of the electric machine and / or permanent magnets are arranged radially further inwards than the outer diameter of the rotor. [11] Electric machine according to claim 10, characterized by that the electric machine is an IPMSM.

Images