METHOD FOR DETERMINING LOSSES, CONVERTER AND METHOD FOR OPERATING A CONVERTER

Description

[0001] The invention relates to a method for determining losses, converters and methods for operating a converter.

[0002] It is well known that improvements in plant design can be achieved by determining losses.

[0003] From XP 032985308, the closest prior art is a method for estimating losses in synchronous motors.

[0004] A reduction of losses in an inverter-fed electric motor is known from XP 033250597.

[0005] The invention is therefore based on the object of simplifying the determination of loss, in particular with regard to the effort involved in the determination.

[0006] According to the invention, the object is achieved by the method according to the features specified in claim 1 and by the converter according to the features specified in claim 7.

[0007] Important features of the invention in the method are that the method is provided for determining losses, in particular a converter-related loss component of the total losses, of at least a first and a second drive, wherein the first drive has a first motor, i.e. a three-phase electric motor, in particular a synchronous motor, fed by a pulse-width modulated converter, wherein the second drive comprises a second motor fed by the converter, i.e. a three-phase electric motor, in particular a synchronous motor, wherein the first motor has a number of windings w_1 and a wire cross-section A1 of its winding wire of its stator winding and the second motor has a number of windings w_2 and a wire cross-section A2 of its winding wire of its stator winding, wherein the wire cross-section A2 is equal to the product of the wire cross-section A1 and the quotient of the number of windings w_1 and the number of windings w_2, wherein in a first process stepthe losses P HL _1 of the first drive are determined according to P HL _ 1 = 3 ⋅ U DC 2 w _ 1 2 ⋅ g f PWM * m , where w_1 is the number of windings of the first motor, where U DC is the intermediate circuit voltage of the inverter, where g f PWM ∗ m = ∑ i ∑ j h fi j ⋅ g i , j 2 m applies and g f PWM ∗ m can be designated as a characteristic function, where i and j number the frequencies fi,j , in particular the non-vanishing frequencies fi,j , in the Fourier spectrum of the supply voltage of the first motor, in particular which is detected, where h fi j = cos ϕ z fi j ⋅ w 1 2 where cos (Φ) is the power factor dependent on the frequencies fi,j and Z is the impedance of the first motor dependent on the frequencies fi,j and is recorded as a function of the frequencies fi,j, where gi,j ( m ) a standardized output voltage of the converter, in particular depending on the degree of modulation m , at the respective frequency fi,j where the normalized output voltage gi,j ( m ) out of U i , j = U DC ⋅ g i , j m is determined, where the voltage components U i,j are determined from the Fourier spectrum of the motor supply voltage, wherein in a second process step the losses P HL _2 of the second drive are determined according to P HL _ 2 = 3 ⋅ U DC 2 w _ 2 2 ⋅ g f PWM * m where w_2 is the number of windings of the second motor, in particular where as characteristic function g f PWM ∗ m the characteristic function determined for the first engine g f PWM ∗ m is used.

[0008] The advantage here is that the characteristic function only needs to be determined for the first motor from the measured values. For the second motor, a calculation is sufficient, taking into account the inverter's DC link voltage and the number of windings of the second motor.

[0009] This way, no special effort is required for the second motor, allowing for a quick and inexpensive calculation. Instead of the second motor, other motors from a series can also be used, which also differ from the first motor only in the number of windings and the wire cross-section.

[0010] Although the invention also exhibits slight deviations from the losses determined by measurements, the achievable deviation is negligible. Thus, a quick and inexpensive determination of losses is achievable.

[0011] Another important aspect of the invention is that the wire cross-section is inversely proportional to the number of turns, and thus the ohmic DC resistance of the winding can be assumed to be proportional to the square of the number of turns. This effect allows the characteristic function to be used to determine losses in both motors.

[0012] Both motors are designed as three-phase motors. Preferably, the motors are synchronous motors.

[0013] In an advantageous embodiment, the quotient of the effective resistance and the square of the impedance is used instead of the quotient of the power factor and the magnitude of the impedance. This is advantageous because appropriate measurement methods, in particular the determination of voltage values, current values, and effective values ​​of current and voltage of the motor phases, can be used.

[0014] In an advantageous embodiment, the first and second motors are structurally identical, in particular identical, to each other except for the number of windings and the wire cross-section. This is advantageous because it allows for easy identification.

[0015] In an advantageous embodiment, the stator of the respective motor comprises a laminated core with radially inwardly projecting teeth, each tooth being wound with a number of turns of a winding, the number being the number of windings of the motor. Advantageously, the motors can be designed as synchronous motors with tooth-wound windings.

[0016] In an advantageous embodiment, the number of windings w_1 differs from the number of windings w_2. This is advantageous because the loss determination is easy to implement.

[0017] In an advantageous embodiment, i and j number those frequencies fi,j of the Fourier spectrum of the supply voltage of the first motor at which non-zero voltage components U i,j are detected. The advantage here is that a simple determination can be carried out.

[0018] Important features of the converter for carrying out the aforementioned method are that the determination of the losses of the second motor is carried out by the signal electronics of the converter, in particular is executable.

[0019] The advantage here is that the losses can be determined during ongoing operation. In an advantageous embodiment, the characteristic function is stored in a memory of the converter, and the converter has a means for detecting the converter's DC link voltage. The advantage here is that the characteristic function can be used for all motors. Important features of the method for operating an aforementioned converter are that the losses determined as a function of the detected DC link voltage of the converter are used in a temperature model of the converter's motor control. The advantage here is that the determined losses can be used during ongoing operation.

[0020] Further advantages arise from the subclaims.

[0021] The invention will now be explained in more detail: In order to determine losses of a first drive, comprising a first three-phase electric motor, in particular a synchronous motor, fed by a first converter, the electric motor is first measured.

[0022] The losses mentioned refer only to the converter-related losses, which of course only represent a portion of the total loss.

[0023] Due to the frequency dependence, the following generally applies to the losses P HL of a drive: P HL = 3 ⋅ ∫ f > f 1 ∞ cos ϕ z f ⋅ U 2 f df , where Z is the impedance of the stator winding of the electric motor, R is the effective resistance of the stator winding, U is the supply voltage of the motor, cos (Φ) is the power factor and f 1 is the fundamental frequency.

[0024] When measuring, the quotient R Z 2 = cos ϕ Z for different frequencies fi,j, where i and j number the frequencies, the numbered frequencies being according to fi,j = j · f PWM ± i · f 1 is the pulse width modulation frequency f PWM , the fundamental frequency f 1 of the converter. Additionally, i+j = 2h+1, where h is a natural number.

[0025] It would be possible to determine the losses according to: P HL = 3 ⋅ U DC 2 ⋅ ∑ i ∑ j cos ϕ Z f i , j ⋅ g i , j 2 m where U DC the intermediate circuit voltage and gi,j ( m ) the standardized output voltage of the converter as a function of the modulation degree m at the frequency fi,j is.

[0026] From the Fourier spectrum of the motor supply voltage, the voltage components U i,j that occur at the respective frequencies fi,j can be determined.

[0027] The standardized output voltage gi,j ( m ) can therefore be determined from U i,j = U DC · gi,j( m ). m = 2 ⋅ 2 ⋅ U S U DC , where US is the phase voltage of the stator winding.

[0028] At the frequencies fi,j, according to R Z 2 = cos ϕ Z The quotient of the power factor and the magnitude of the impedance is determined. Together with the voltage components U i,j determined from the Fourier spectrum, the above-mentioned double sum can thus be determined.

[0029] However, this determination would have to be carried out for every other motor as well, since these other motors differ from the first motor in their number of windings and wire cross-section.

[0030] Each of the motors comprises a stator core having radially inwardly projecting teeth, with a number w of windings. Preferably, the teeth are evenly spaced from one another in the circumferential direction.

[0031] The number of windings w is therefore characteristic of the respective motor. The wire cross-section is selected inversely proportional to the number of windings.

[0032] According to the invention, the determination of losses is advantageously carried out according to: P HL = 3 ⋅ U DC 2 w 2 ⋅ g f PWM * m , where g f PWM ∗ m = ∑ i ∑ j h f ⋅ g i , j 2 m and h f = cos ϕ Z f ⋅ w 2 applies.

[0033] Therefore, the determination of g f PWM ∗ m , especially for a given fundamental frequency f 1 to determine a characteristic function for the first drive.

[0034] For all engines, the characteristic function is g f PWM ∗ m even.

[0035] Therefore, the determination of losses for a particular P HL = 3 ⋅ U DC 2 w 2 ⋅ g f PWM * m

[0036] Purely mathematically possible, since the characteristic function g f PWM ∗ m does not need to be determined again.

[0037] The advantage of the invention is therefore that converter-related losses from the first motor can be converted to corresponding losses in the other motors with different numbers of turns without the need for further measurement determinations.

[0038] Preferably, this conversion can be stored or executed in the signal electronics of the converter in order, for example, to improve temperature models used for motor control and / or to determine energy efficiency.

[0039] However, the calculation of losses can also be used in the design of systems with a single drive, where the calculation of losses is now simplified. For example, from several available drives, the drive with the lowest losses can be selected for the system. List of reference symbols

[0040] g f PWM ∗ m characteristic function gi,j ( m) standardized output voltage cos (Φ) power factor f 1 fundamental frequency U DC intermediate circuit voltage of the converter US phase voltage

Claims

1. Method for determining losses, in particular a converterrelated loss component of the total losses, of at least a first and a second drive, wherein the first drive comprises a first motor, i.e. a three-phase electric motor, in particular a synchronous motor, fed by a pulse-width-modulated converter, wherein the second drive comprises a second motor, i.e. a three-phase electric motor, in particular a synchronous motor, fed by the converter, wherein the first motor has a number of windings w_1 and a wire cross-section A1 of its winding wire of its stator winding, and the second motor has a number of windings w_2 and a wire cross-section A2 of its winding wire of its stator winding, wherein the wire cross-section A2 is equal to the product of the wire cross-section A1 and the quotient of the number of windings w_1 and the number of windings w_2, wherein in a first method step the losses PHL_1 of the first drive are determined according to P HL _ 1 = 3 ⋅ U DC 2 w _ 1 2 ⋅ g f PWM * m , wherein w_1 is the number of windings of the first motor, wherein UDC is the DC link voltage of the converter, wherein g f PWM ∗ m = ∑ i ∑ j h fi j ⋅ g i , j 2 m , and g f PWM ∗ m can be designated as a characteristic function, wherein i and j number the frequencies fi,j, in particular the non-vanishing frequencies fi,j, in the Fourier spectrum of the supply voltage of the first motor, in particular which is detected, wherein h fi j = cosϕ Z fi j ⋅ w 1 2 , wherein cos(Φ) is the power factor dependent on the frequencies fi,j, and Z is the impedance of the first motor dependent on the frequencies fi,j, and these are detected as a function of the frequencies fi,j, wherein gi,j(m) is a standardized output voltage of the converter, in particular as a function of the degree of modulation m, at the respective frequency fi,j, wherein the standardized output voltage gi,j(m) is determined from U i , j = U DC ⋅ g i , j m , wherein the voltage components Ui,j are determined from the Fourier spectrum of the supply voltage of the motor, wherein in a second method step the losses PHL_2 of the second drive are determined according to P HL _ 2 = 3 ⋅ U DC 2 w _ 2 2 ⋅ g f PWM ∗ m , wherein w_2 is the number of windings of the second motor, in particular wherein the characteristic function g f PWM ∗ m determined for the first motor is used as the characteristic function g f PWM ∗ m .

2. Method according to claim 1, characterized in that the quotient of the effective resistance and the square of the impedance is used instead of the quotient of the power factor and the magnitude of the impedance.

3. Method according to any one of the preceding claims, characterized in that the first and the second motor are of the same construction, in particular are identical to each other, apart from the number of windings and the wire cross-section.

4. Method according to any one of the preceding claims, characterized in that the stator of each motor comprises a laminated core with radially inwardly projecting teeth, wherein each tooth is wound with a number of turns of a winding, the number being the number of windings of the motor.

5. Method according to any one of the preceding claims, characterized in that the number of windings w_1 differs from the number of windings w_2.

6. Method according to any one of the preceding claims, characterized in that wherein i and j number those frequencies fi,j of the Fourier spectrum of the supply voltage of the first motor at which non-vanishing voltage components Ui,j are detected.

7. Converter configured to carry out a method according to any one of claims 1 to 6 or according to claim 9, characterized in that the losses of the second motor are determined, in particular can be determined, by the signal electronics of the converter, in particular wherein the signal electronics of the converter are suitable for determining the losses of the second motor.

8. Converter according to claim 7, characterized in that the characteristic function is stored in a memory of the signal electronics of the converter, and the converter has a means for detecting the DC link voltage of the converter.

9. Method according to any one of claims 1 to 6, characterized in that the losses determined as a function of the detected DC link voltage of the converter are used in a temperature model of the motor control of the converter.