Method for determining a correction value which describes an angular difference between an assumed and an actual position of a d-axis, control device and inverter

Abstract

A method determines a correction value for a control device for an electric machine, which describes an angular difference between a position of a d-axis of the electric machine and an actual position of the d-axis. The correction value is determined in a rotating state of a rotor of the electric machine as a function of a d voltage value describing a d component of a stator voltage specified by the control device in the initial configuration, of a q voltage value describing a q component of the stator voltage specified by the control device in the initial configuration, of a flux value describing a magnetic flux of the rotor, a speed value describing the speed of the rotor in the rotating state, and of a calibration value describing a speed-dependent voltage error of the d component of the stator voltage.

Claims

1. Method for determining a correction value for a control device for an electric machine, the correction value describing an angular difference between a position of a d-axis of the electric machine assumed in an initial configuration of the control device on a basis of rotor position information of a rotor position sensor and an actual position of the d-axis, a zero current being impressed in stator windings of the electric machine and the correction value being determined in a rotating state of a rotor of the electric machine as a function of a d-voltage value describing a d-component of a stator voltage specified by the control device in the initial configuration, of a q-voltage value describing a q-component of the stator voltage specified by the control device in the initial configuration, of a flux value describing a magnetic flux of the rotor, of a speed value describing a speed of the rotor in the rotating state, and of a calibration value describing a speed-dependent voltage error of the d component of the stator voltage, wherein the flux value is determined as a function of the d-voltage value and the q-voltage value and the speed value as an additional function of the calibration value.

2. Method according to claim 1, wherein the calibration value is read out on a basis of the speed value from a stored characteristic diagram which assigns the calibration value to speeds of the electric machine in each case, or is fixed for a speed in the rotating state of the rotor.

3. Method according to claim 1, wherein setpoint voltage values of a control unit of the control device are used as the predetermined stator voltage or the predetermined stator voltage is determined from output signals of a modulator unit of the control device which provides switching signals for a power unit of an inverter.

4. Method according to claim 1, wherein the electric machine is a permanently excited electric machine and the flux value describes the magnetic flux of permanent magnets of the rotor, or wherein the electric machine is an electrically excited electric machine and the flux value describes the rotor flux concatenated with the stator.

5. Method according to claim 1, wherein a determined correction value is rejected as unreliable if the determined flow value is above a predetermined maximum flow value and/or if the determined flow value is below a predetermined minimum flow value.

6. Method according to claim 1, wherein the d-voltage value, the q-voltage value and the speed value are determined as a function of mean values of respective individual values acquired over an acquisition period comprising one or more complete electrical or mechanical periods.

7. Method according to claim 6, wherein a plausibility check verifying a quasi-stationary operation of the electric machine is performed during the acquisition period of the individual values and a correction value is rejected as unreliable if the plausibility check reveals a violation of a predetermined condition for the quasi-stationary operation.

8. Method of claim 7, wherein the condition comprises that individual values used to determine the speed value are within a predetermined speed interval and/or that a d-current value describing an average of individual values of a d-component of a stator current detected during the detection period and a q-current value describing an average of individual values of a the d-component of the stator current detected during the detection period are within a current interval comprising zero.

9. Method according to claim 7, wherein a plurality of correction values are determined in a predetermined or predeterminable number of successive determination cycles and an overall correction value is used as the average of the correction values determined in the determination cycles, if the number of correction values rejected as unreliable during the respective plausibility check does not exceed a predefined maximum value and/or the deviation of the correction values determined in the determination cycles from each other does not exceed a specified deviation measure.

10. A computer program comprising program code for carrying out a method according to claim 1, when the computer program is executed on a computing device.

11. Method for determining a correction value for a control device for an electric machine, the correction value describing an angular difference between a position of a d-axis of the electric machine assumed in an initial configuration of the control device on a basis of rotor position information of a rotor position sensor and an actual position of the d-axis, a zero current being impressed in stator windings of the electric machine and the correction value being determined in a rotating state of a rotor of the electric machine as a function: of a d-voltage value describing a d-component of a stator voltage specified by the control device in the initial configuration, of a q-voltage value describing a q-component of the stator voltage specified by the control device in the initial configuration, of a flux value describing a magnetic flux of the rotor, of a speed value describing a speed of the rotor in the rotating state, and of a calibration value describing a speed-dependent voltage error of the d component of the stator voltage, wherein the following equation is evaluated to determine the correction value:
Δγ=γ tan(ω.sub.el.Math.Ψ.Math.u.sub.d−Δu.sub.d.Math.u.sub.q, Δu.sub.d.Math.u.sub.d+ω.sub.el.Math.Ψ.Math.u.sub.q), where α tan(x, y) describes an arc tangent function or an arc cotangent function, in particular $\arctan \left(\frac{x}{y}\right)\mathrm{or}\mathrm{arccot}\left(\frac{y}{x}\right)$ or α tan 2(x, y) or α cot 2(y, x), u.sub.d describes the d-voltage value, u.sub.q describes the q-voltage value, Ψ describes the flux value, ω.sub.el describes the speed value, and Δu.sub.d describes the calibration value.

12. A control device for an electric machine, which is arranged to impress a zero current into stator windings of the electric machine, and to determine a correction value describing an angular difference between a position of a d-axis of the electric machine assumed in an initial configuration of the control device on a basis of rotor position information of a rotor position sensor and an actual position of the d-axis in a rotating state of a rotor of the electric machine in dependence of a d-voltage value describing a d-component of a stator voltage specified by the control device in the initial configuration, of a q-voltage value describing a q-component of the stator voltage specified by the control device in the initial configuration, of a flux value describing a magnetic flux of the rotor, of a speed value describing the speed of the rotor in the rotating state, and of a calibration value describing a speed-dependent voltage error of the d-component of the stator voltage, and wherein the flux value is determined as a function of the d-voltage value and the q-voltage value and the speed value as an additional function of the calibration value.

13. An inverter for the electric machine, comprising a power unit adapted to convert an input DC voltage into a polyphase AC current for the electric machine, and the control device according to claim 12.

14. A vehicle comprising the electric machine to drive the vehicle and the inverter according to claim 13 adapted to provide the alternating current to the electric machine.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a block diagram of an embodiment of the vehicle according to the invention, comprising an embodiment of the inverter according to the invention with an embodiment of the control device according to the invention;

(2) FIG. 2 is an initial and an actual dq coordinate system during operation of the control device according to the invention;

(3) FIG. 3 is a block diagram of a correction value determination unit and a writing unit of the control device shown in FIG. 1;

(4) FIG. 4 is a flow diagram of an embodiment of the method according to the invention; and

(5) FIG. 5 is an exemplary curve of a voltage error over a mechanical speed.

DETAILED DESCRIPTION OF THE INVENTION

(6) FIG. 1 is a block diagram of an embodiment of a vehicle 1, comprising an embodiment of an inverter 2 with an embodiment of a control device 3.

(7) The vehicle 1 also has a DC voltage source 4 in the form of a high-voltage battery and an electric machine 5 which is set up to drive the vehicle 1 by mechanical coupling with a load 6. In the present embodiment, the electric machine 5 is a permanently excited synchronous machine. The electric machine 5 is assigned a rotor position sensor 7, for example a resolver, which is set up to provide rotor position information 8 to the control device 3.

(8) In addition to the control device 3, the inverter 2 comprises a power unit 9, which has a plurality of power switching elements that can be controlled by means of the control device 3. On the basis of switching signals 10 from the control device 3, the power unit 9 converts the DC voltage provided by the DC voltage source 4 into a multiphase AC current for the electric machine 5. In addition, the inverter 2 comprises a current detection unit 11, which is set up to provide the control device 3 with current information 12 describing individual values of stator currents flowing along individual phases of the multiphase AC current.

(9) The control device 3 comprises a control unit 13 which is set up to determine setpoint voltages in dq coordinates as a function of an externally supplied torque demand 14. The control unit 13 is followed by a transformation unit 15, which is used to convert the setpoint voltages u*.sub.dq into nominal voltages u*.sub.abc in abc coordinates. The control unit 3 also has a modulator unit 16 which is set up to convert setpoint voltages determined by the control unit 13, in this case the setpoint voltages u*.sub.abc into the pulse-width-modulated switching signals 10 for the power unit 9, if necessary using limiting rules.

(10) In addition, the control unit 3 has a further transformation unit 17, which is set up to transform the current information 12, which describes the stator current in abc coordinates, into individual values of the stator current i.sub.dq in dq coordinates, which the control unit 13 receives as actual values for control. A further transformation unit 18 receives an output signal of the modulation unit 16, which describes the voltages generated by the switching signals 10 in abc coordinates, and is used to transform these individual values into individual values of the stator voltage u.sub.dq which take into account modifications with respect to the nominal voltages u*.sub.dq.

(11) The control device 3 further comprises an evaluation unit 19, which is set up to determine from the rotor position information 8 an electrical angular frequency ω.sub.el of the rotor motion and a position information 20 describing the position of the d-axis of the dq-coordinate system.

(12) FIG. 2 shows an initial and an actual dq coordinate system during operation of the control device 3.

(13) In an initial configuration of the control device, in which an offset of a zero axis 21 of the rotor position encoder with respect to an actual position d axis marked with d is not known, a position of the d axis marked with d.sub.init is assumed, which has an initial offset φ.sub.init with respect to the zero axis 21. Thereby FIG. 2 also shows corresponding orthogonal q-axes, marked with q for the actual position of the d-axis and with q.sub.init for the assumed position of the d-axis.

(14) Consequently, it is necessary to determine a total correction value Δγ which allows a correction of the initial offset φ.sub.init to an offset φ better corresponding to the actual position of the d-axis.

(15) Referring again to FIG. 1, the control device 3 has for this purpose a correction value determination unit 22 for determining the offset φ. By means of a writing unit 23 of the control device 3, the evaluation unit 19 can be configured in such a way that the position information 20 is no longer determined on the basis of the initial offset φ.sub.init, but on the basis of the offset φ.

(16) In this context, the control device 3 is set up to carry out a method which is explained below on the basis of a block diagram of the correction value determination unit 22 and the writing unit 23 shown in FIG. 3 and a flow diagram of the method shown in FIG. 4:

(17) In an initial state S1 at the start of the method, the vehicle 1 is driving, with a rotor of the electric machine rotating at a certain speed and the control device being in the initial configuration. In a step S2, it is cyclically checked whether the control device 3 receives an external trigger signal T from a higher-level control device (not shown) of the vehicle 1, which starts the determination. After receiving the trigger signal T, in a step S3 the control unit 13 specifies the target voltages u.sub.dq* such that a zero current is impressed in stator windings of the electric machine 5.

(18) In a subsequent step S4, an averaging block 24 of the correction value determination unit 22 obtains, over an acquisition period comprising one or more complete electrical or mechanical periods of the electrical machine 5, a plurality of individual values of a d component of the stator voltage u.sub.d,init and a q-component of the stator voltage u.sub.q,init from the modulator unit 16 via the transformation unit 18, respectively. In addition, the averaging block 24 receives single values of a d-component of the stator current i.sub.d,init and a q-component of the stator current i.sub.q,init from the current detection unit 11 via the transformation unit 17 in each case. In addition, the first averaging block receives 24 individual values of the electrical angular frequency ω.sub.el,init. In a subsequent step S5, the averaging block 24 generates a d-voltage value u.sub.d, a q-voltage value u.sub.q, a speed value ω.sub.el, a d-voltage value i.sub.d and a q-voltage value i.sub.q from the respective individual values by averaging over the duration of the acquisition period.

(19) In a following step S6, a calibration value determination block 25 of the correction value determination unit 22 determines from the speed value ω.sub.el calibration value Δu.sub.d, which describes a speed-dependent voltage error of the d component of the stator voltage. For this purpose, the calibration value determination block 25 reads, on the basis of the speed value ω.sub.el a stored characteristic field, which assigns a calibration value to each of the speeds of the electric machine 5. FIG. 5 shows an exemplary course of a voltage error, which is described by the calibration value Δu.sub.d over a mechanical speed. The voltage error results from iron losses and timing errors and has been determined for the type of the electric machine 5 and the type of the inverter 2 before the start of the process and stored in a memory (not shown) of the control unit 3.

(20) In a subsequent step S7, a flux value determination block 26 of the correction value determination unit 22 determines from the d-voltage value u.sub.d, the q-voltage value u.sub.q, the speed value ω.sub.el and the calibration value Δu.sub.d according to the equation

(21) $\Psi =\frac{\sqrt{u_d^2+u_q^2-\Delta u_d^2}}{{\omega }_{el}}$
a flux value Ψ, which describes a magnetic flux of the permanent magnets of the rotor.

(22) Subsequently, in a step S8, from the d-voltage value u.sub.d, the q-voltage value u.sub.q, the flux value Ψ, the speed value ω.sub.el and the calibration value Δu.sub.d by means of a correction value calculation block 27 a correction value Δγ.sub.i according to the equation
Δγ.sub.i=Δ tan 2(ω.sub.el.Math.Ψ.Math.u.sub.d−Δu.sub.d.Math.q.sub.q, Δu.sub.d.Math.u.sub.d+ω.sub.el.Math.Ψ.Math.u.sub.q)
is calculated.

(23) Subsequent to step S8 or in parallel with steps S5 to S8, an extreme value determination block 28 of the correction value determination unit 22 determines, in a step S9, a minimum value of the individual values of the electric angular frequency ω.sub.el,min occurring during the detection period and a maximum value of the individual values of the electric angular frequency ω.sub.el,max occurring during the detection period. In a subsequent step S10, which can also be performed in parallel with steps S5 to S8, a plausibility check block 29 of the correction value determination unit 22 checks a condition for quasi-stationary operation of the electric machine 5 during the detection period. For this purpose, on the one hand, based on the minimum value ω.sub.el,min and the maximum value ω.sub.el,max, it is determined whether the individual values detected during the detection period are within a predetermined speed interval and, on the other hand, whether the d-current value i.sub.d and the q current value i.sub.q lie within a current interval comprising zero.

(24) Subsequent to step S10, in a step S11, which can alternatively be performed in parallel with steps S7 and S8, a flow value checking block 30 of the correction value determining unit 22 checks whether the determined flow value is ψ is below a predetermined maximum flow value and above a predetermined minimum flow value. If the determined flow value ψ is below the minimum flux value, the flux value checking block 30 additionally outputs demagnetization information 31 indicating demagnetization of the permanent magnets.

(25) The subsequent steps S12 to S19 described below are performed by an overall correction value calculation block 33 of the correction value determination unit 22. In step S12, a first counter which counts the number of determined correction values Δγ.sub.i is incremented. Then, in a step S13, it is evaluated whether the conditions checked in steps S10 and 511 and cumulatively linked by a logic block 32 are satisfied. If this is not the case, the determined correction value Δγ.sub.i is considered unreliable and a new correction value Δγ.sub.i+1 is determined by returning to step S4.

(26) If the evaluation in step S13 is positive, the correction value Δγ.sub.i can be considered as reliable, so that in a subsequent step S14 a second counter counting the number of reliable correction values Δγ.sub.i is incremented and the correction value Δγ.sub.i is stored. In a step S15, it is then evaluated whether the first counter has reached a predetermined value of, for example, twenty correction value determinations. If this is not the case, the program returns to step S4 so that a further correction value Δγ.sub.i+1 is determined.

(27) If the evaluation in step S15 shows that a sufficient number of correction value determinations have been carried out, it is evaluated in a step S16 whether the second counter has a predetermined minimum value of, for example, eighteen correction values Δγ.sub.i considered reliable. If this is not the case, the counters are reset in step S17 and previously stored correction values Δγ.sub.i are deleted. The program then returns to step S4 to determine a new set of correction values Δγ.sub.i.

(28) If the evaluation in step S16 is successful, an average value of the stored correction values Δγ.sub.i, i.e., the values considered reliable, is determined in step S18. In a step S19, a deviation measure for the mean value is evaluated by checking whether a minimum value and a maximum value of the correction values Δγ.sub.i lie within an interval defined around the mean value. If this is not the case, a jump is made to step S17 and a new set of correction values Δγ.sub.i is determined. If the evaluation in step S19 is positive, the mean value is output in a step S20 as total correction value Δγ to the writing unit 23.

(29) In a step S21, the writing unit 23 determines from the initial offset φ.sub.init and the total correction value Δγ by difference formation the offset φ and writes this to the evaluation unit 19, thus ending the process.

(30) In the further operation of the electric machine 5 or of the vehicle 1, the position information 20 is thus determined much more precisely on the basis of the offset φ which enables a more precise control of the electric machine 5. The procedure is carried out for the first time immediately after initial commissioning of the vehicle 1, in order to correct the factory-set initial offset φ.sub.init as quickly as possible. Thereafter, the procedure is restarted by the trigger signal T, for example, when after maintenance or repair of the vehicle 1 or after expiration of a predetermined use of the vehicle, there is a need to check the reliability of the offset φ which is used as the initial offset φ.sub.init when the procedure is carried out again.

(31) According to further embodiments, the electric machine 5 is an electrically excited synchronous machine. In this case the flux value W describes the rotor flux concatenated with the stator at a stator current of zero value and the demagnetization information 31 is omitted.

(32) According to a further embodiment, averaging by the total correction value calculation block 33 in steps S11 to S18 is omitted, and a correction value Δγ.sub.i determined to be reliable for correction of the initial offset φ.sub.init is output to the writing unit 23.

(33) According to a further embodiment, in step S1 the electric machine 5 is operated on a test bench, for example as part of an end-of-line test, against a load machine at a predetermined speed. The determination of the calibration value Δu.sub.d is omitted in favor of a calibration value Δu.sub.d which is fixed for the rotational speed. Steps S4, S5, S9 and S10 can then be omitted and individual values can be used as the d voltage value, q voltage value and speed value.

(34) According to a further embodiment, there is no return after step S17, but a signal indicating unsuccessful determination is output to the higher-level control unit (not shown). The vehicle control unit can then output the trigger signal T again.