System for operating an inverter-fed electric rotating field machine

Abstract

System for operating an inverter-fed electric rotating field machine, where the motor current is measured and a stator current space vector is determined from it, wherein the stator voltage space vector is determined from at least one detected voltage value, in particular DC link voltage value, and optionally from a pulse width modulation signal output by the inverter, in particular pulse width modulation component, wherein the stator current space vector and the stator voltage space vector are supplied to an observer who determines a reference angle on the one hand and an actual speed value on the other hand, which is fed to a determining device which, taking into account the reference angle, determines a stress space vector which is transformed by a transformation element into another coordinate system, in particular into a rotor flux-oriented coordinate system, and this transformed stress space vector is then made available to the inverter as a manipulated variable.

Description

[0001] The invention relates to a system for operating an inverter-fed electric rotating field machine.

[0002] It is known to power an electric rotating field machine from an inverter.

[0003] From DE 10 2006 004 034 A1, a method for sensorless operation of a converter-fed, permanent excitation synchronous machine with a test signal is known as a generic prior art.

[0004] The invention is therefore based on the objective of saving energy during the operation of an electric machine.

[0005] According to the invention, the problem is solved in the system according to the features specified in claims 1, 2 and 3.

[0006] Key features of the invention in the system are that the system is designed to operate an inverter-fed electric rotating field machine, where the motor current is measured and a stator current space vector is determined from it, wherein the stator voltage space vector is determined from at least one detected voltage value, in particular DC link voltage value, and optionally from a pulse width modulation signal output by the inverter, in particular pulse width modulation component, wherein the stator current space vector and the stator voltage space vector are supplied to an observer who determines a reference angle on the one hand and an actual speed value on the other hand, which is fed to a determining device which, taking into account the reference angle, determines a stress space vector which is transformed by a transformation element into another coordinate system, in particular into a rotor flux-oriented coordinate system, and this transformed stress space vector is then made available to the inverter as a manipulated variable.

[0007] The advantage here is that the characteristic curve field can be selected in such a way that the operating points are chosen to be as energy-saving as possible.

[0008] In an advantageous embodiment, a torque value is also specified for the determining device. The advantage here is that the rotational speed and torque are predetermined, and the determining device uses these values ​​to calculate the stress space vector, taking into account the reference angle determined by the observer.

[0009] In an advantageous embodiment, the determining means determines, from a characteristic curve field, in particular from a characteristic curve field stored in a memory, an output-side value pair, i.e. voltage magnitude value and electrical frequency value, for each of the supplied value pairs, i.e., torque value and actual speed value. The electrical frequency value is fed to an integrator, whose output value, together with the voltage magnitude, is fed to another transformation element. This transformation element determines a voltage space vector, which is then transformed by the other coordinate system, in particular the rotor flux-oriented coordinate system. This transformed voltage space vector is then provided to the inverter as a control variable. An advantage of this method is that it enables a simple and rapid determination of the operating points.

[0010] In another advantageous embodiment, the determining means determines, for each of the supplied value pair, i.e., torque value and actual speed value, from a characteristic curve field, in particular from a characteristic curve field stored in a memory, an output-side value pair, i.e., two voltage space vector components of the voltage space vector of a representation in the rotor flux-oriented or stator flux-oriented coordinate system. The components are transformed by the transformation element into the other coordinate system, in particular into the rotor flux-oriented coordinate system, and this transformed voltage space vector is then provided to the inverter as a control variable. An advantage of this is that a simple and rapid determination of the operating points is possible.

[0011] In another advantageous embodiment, the determining means determines, for each of the supplied value pair, i.e., torque value and actual speed value, from a characteristic curve field, in particular from a characteristic curve field stored in a memory, an output-side value pair, i.e., two current space vector components of the current space vector of a representation in the rotor flux-oriented or stator flux-oriented coordinate system. The components are each fed as setpoints to a controller, which also receives an actual value of the torque-generating or flux-generating current component. The output signal of each controller is a component of a voltage space vector provided by the inverter. This voltage space vector is transformed by the transformation element into the other coordinate system, specifically the rotor flux-oriented coordinate system, and this transformed voltage space vector is then provided to the inverter as a manipulated variable. An advantage of this approach is that it enables a simple and rapid determination of the operating points, using fast current controllers and thus ensuring high control accuracy.

[0012] In an advantageous embodiment, the actual speed value determined by the observer at the output, particularly after subsequent low-pass filtering, is fed not only to the measuring device but also to a controller. A setpoint is also fed to the controller, and the torque value is determined as the controller's output value from the difference between the setpoint and the actual value. This torque value is then fed back to the measuring device. An advantage of this design is that only the specification of a speed setpoint is necessary.

[0013] In an advantageous embodiment, one or each of the controllers has a proportional (P) controller component, a pi (PI) controller component, and / or a PID (Pulse-In) controller component. The advantage here is that a high level of control accuracy can be achieved.

[0014] In an advantageous embodiment, the characteristic curve is determined by selecting the output value pair associated with the input value pair (i.e., torque value and actual speed value) such that minimal power loss occurs during steady-state operation. This has the advantage of saving energy.

[0015] In an advantageous embodiment, the output-side value pair is selected to determine the characteristic curve field such that the power loss is minimal, subject to the constraint that the magnitude of the voltage and the magnitude of the current remain below or equal to a respective maximum permissible value. where the torque corresponds to the torque value and the speed to the actual speed value. An advantage of this is the reduced energy consumption.

[0016] In an advantageous design, the power loss is the sum of the stator heat losses, the rotor heat losses, the friction losses, the iron losses, and the additional losses. An advantage of this is that a simple model can be applied.

[0017] In an advantageous embodiment, the stator current heat losses P CuStator = 3R Stator (T)(|I Stator (U,f,s)|) 2 , where the complex RMS value of the stator current I stator is determined from an FEM calculation modeling the electric machine and R Stator (T) denotes the temperature-dependent stator resistance, where T is the temperature, including the rotor current heat losses PCu,Rotor=1P2πfRotorM(U,f,s) are, where the torque M(U,f,s) is from M(U,f,s)=3p⋅1ωsRe{(UStator−RStator(T)IStator(U,f,s))IStator*(U,f,s)} is calculated, where p is the number of pole pairs, ω s= 2πf the angular frequency of the voltage system, Re the real part of a complex quantity, U Stator the complex RMS value of the stator voltage and the operator * denotes the conjugation of a complex quantity, where the iron losses P Fe (U,f,s) are determined from the FEM calculation modeling the electric machine, where the friction losses PReib=PReib,0(nn0)C are, where P Reib,0 The friction losses at the rated speed n0 of the electric machine are described, and the exponent c reflects the speed-dependent behavior of the friction losses, with both quantities being determined by measurements on the electric machine, and where the additional losses P Zusatz This can be described as a percentage of the shaft power. An advantage of this approach is that it allows for the consideration of quickly and easily determined types of power loss.

[0018] Further advantages arise from the sub-claims.

[0019] The invention will now be explained in more detail with the help of illustrations: In the Fig. Figure 1 shows a system according to the invention, wherein a determining means 10 is provided. In the Fig. Figure 2 shows a first embodiment of the determining agent 10. In the Fig. Figure 3 shows an alternative embodiment of the determining agent 10. In the Fig. Figure 4 shows another alternative embodiment of the determining agent 10. In the Fig. 5 is a further development of another determining instrument 11 related to Fig. 4 shown. In the Fig. Figure 6 shows the torque-speed characteristics of an electric machine in the system.

[0020] As in Fig. Figure 1 shows an electric rotating field machine, in particular a three-phase motor 1, powered by a pulse-width modulated inverter 2, whose control signals are generated by a PWM modulation unit 3 operating according to the pulse-width space vector modulation method. The inverter is supplied from a unipolar voltage source, in particular an intermediate circuit. Preferably, this voltage source is represented by a mains-fed rectifier.

[0021] The current intermediate circuit voltage U measured in inverter 2 z is taken into account by the PWM modulation unit 3 when generating its output signals.

[0022] A determining instrument 10, which represents a functional dependence of its output variables on its input variables, generates a stress space vector (U). α , U β The output variable is the torque setpoint and the actual rotational speed.

[0023] The actual speed value and an angle of a reference coordinate system are determined by a speed / flux observer 4, which receives as input the actual value of the motor current space vector detected in the inverter 2 and the value from the PWM modulation unit 10, taking into account the DC link voltage U. z specific actual value of the motor voltage space pointer.

[0024] Since the electric machine has a shaft rotatably mounted via two bearings fixed to a housing, the torque setpoint can be described by a scalar and not by a space vector.

[0025] The functional relationship is at least two-dimensional. The stress space vector in the corresponding coordinate system is determined from the two scalars, i.e., the target torque value and the actual rotational speed.

[0026] The more detailed structure of the determining agent 10 is described in Fig. Figure 2 shows that in a further determination device 11, a value pair consisting of a voltage magnitude and a frequency is determined from a characteristic curve array stored in a memory, based on the current value pair of target torque and actual rotational speed. The frequency signal thus generated is fed to an integrator 13, at whose output an angle value is generated. This angle value and the voltage magnitude are then fed to the transformation element 12, so that the voltage space vector can be determined in stator-fixed Cartesian coordinates as the output signal of the transformation element 12. Using the stored characteristic curve array, a particularly fast and simple determination of the output quantities, i.e., voltage magnitude and frequency, is possible.

[0027] In a further embodiment according to the invention, according to the in Fig. 3 shown alternative design of the further determining means 11 in contrast to the design according to Fig. 2. From the input-side value pair, i.e., again target torque and actual speed, a value pair is determined that consists of the two components (U d , U q ) of the stress space vector in a Cartesian rotor flux-oriented or stator flux-oriented coordinate system. The stress space vector is thus defined by the two-tuple (U d , U q ) is represented and the coordinate system, in particular its d-axis and q-axis directions, rotates with the flux. The stress space vector (U) represented in this rotating coordinate system d , U q ) is in turn fed to a transformation element 12, which, unlike the Fig. 2 now performs the correspondingly different coordinate transformation of the stress space vector into a stator-fixed Cartesian coordinate system. In doing so, the reference angle value γ determined by the rotational speed / flux observer 4 is again used. ref This is taken into account. Using the stored two-dimensional characteristic curve field, a particularly quick and easy determination of the output parameters, i.e., voltage magnitude and frequency, is again possible.

[0028] In a further embodiment according to the invention, according to Fig. 4 in the alternative version shown, further determining means 11 in contrast to the version according to Fig. 2. From the input-side value pair, i.e., again target torque and actual speed, a value pair is determined that consists of the two components (I d , I q) of the stator current space vector in a Cartesian rotor flux-oriented or stator flux-oriented coordinate system. The current space vector is thus defined by the two tuple (I d , I q The coordinate system, in particular its d-axis and q-axis directions, rotates with the flux. The two components are each supplied to a controller unit (13c, 14c) as their respective setpoints. These controller units (13c, 14c) determine the setpoint from this setpoint and the actual current value (I) supplied to them. d,ist , I d,soll ) each of the components of the stress space vector (U d , U q ), which in turn already stems from Fig. The 3 known transformation element 12 is fed to it. Thus, the stress space vector is transformed into a stator-fixed Cartesian coordinate system. The reference angle value γ determined by the speed / flux observer 4 is then used in this process. refThis is taken into account. Using the stored two-dimensional characteristic curve field, a particularly quick and easy determination of the output variables, i.e., voltage magnitude and frequency, is again possible. The controller units (13c, 14c) are implemented with either a proportional (P) controller component, a pi controller component, or a PID controller component.

[0029] In a further embodiment according to the invention, according to Fig. 5 The torque value supplied as input to the determining device 10 is determined by means of a torque controller. For this purpose, the actual speed n determined by the speed / flow observer 4 and subsequently low-pass filtered (5, 6) is used. ist fed to a speed controller 7, which also receives a setpoint for speed n Soll The speed controller 7 determines the setpoint from the control deviation, i.e., the difference between the setpoint n. Soll and actual value n ist, the torque value M used as input by the determining instrument 10 Soll and is implemented with a P controller component, a PI controller component, or a PID controller component.

[0030] To determine the characteristic curve field for the embodiment according to Fig. 2. The following determination procedure is applied: If the electric machine is designed as an asynchronous machine, a desired operating point, which corresponds to a specific speed value n, is Soll and a corresponding torque value M Soll exhibits, can be generated by different motor voltage values ​​U, slip values ​​s and frequency values ​​f generated on the output side of the inverter.

[0031] Using a finite element method (FEM) calculation, the stator current space vector (i.e., current magnitude and phase angle) is determined for steady-state operation. Furthermore, the iron losses are calculated from this according to a model.

[0032] To determine the optimal triple of values ​​(U, f, s), a minimum of the power loss P is sought. V determined under the constraints, - that the rotational speed n corresponds to the associated rotational speed value n Soll , - that the torque M corresponds to the associated torque value M Soll corresponds, - that the voltage U to be output by the inverter is less than or equal to the maximum permissible voltage value U max and - that the magnitude of the current space vector I is less than or equal to the maximum permissible current magnitude I max .

[0033] For each operating point (n soll ,M soll ) thus the optimal control (U,f,s) is considered the solution to the optimization problem. PV(U,f,s)→min!subject to the constraints=nshouldM=MshouldU≤UmaxI≤Imax Calculated.

[0034] The power loss P is Vformed from the individual power losses described in the respective models, including: PV(U,f,s)=PCu,Stator+PCu,Rotor+PFe+PReib+PFaddition

[0035] These include: The stator current heat losses P Cu.Stator = 3R Stator (T)(|I Stator (U,f,s)|) 2 , where the complex RMS value of the stator current I Stator emerges from the FEM calculation as a complex quantity, and R Stator (T) denotes the temperature-dependent stator resistance, the rotor current heat losses PCu,Rotor=1P2πfRotorM(U,f,s) where the torque M(U,f,s) from M(U,f,s)=3p⋅1ωsRe{(UStator−RStator(T)IStator(U,f,s))IStator*(U,f,s)} is calculated, where p is the number of pole pairs, ω s = 2πf the angular frequency of the voltage system, Re the real part of a complex quantity, U Stator the complex RMS value of the stator voltage and the operator * denotes the conjugation of a complex quantity, the iron losses P Fe (U,f,s), which arise as a direct result from the FEM calculation and are described, for example, via polynomial fits, the friction losses PReib=PReib,0(nn0)C where P Reib,0 The friction losses at the rated speed n0 are described, and the exponent C reflects the speed-dependent behavior of the friction losses, with both quantities being determined by measurements on the electric machine, and the additional losses P Zusatz , which are described as a percentage of the wave power.

[0036] If the electric machine is designed as a synchronous machine, the desired operating point is determined using the method described in DE 10 2011 013 128 A1 or DE 10 2011 009 935 A1. Reference symbol list 1 three-phase motor 2 pulse-width modulated inverters 3 PWM modulation units 4 speed / flow monitors 5 Low-pass filter 6 Low-pass filter 7 speed controllers 10 Determining tools 11 further means of determination 12 Transformation element 13 Integration element 13c control unit 14c control unit

Claims

[1] System for operating an inverter-fed electric rotating field machine, where the motor current is measured and a stator current space vector is determined from it, wherein the stator voltage space vector is determined from at least one detected voltage value, namely the intermediate circuit voltage value, and optionally from a pulse width modulation component output by the inverter (2), wherein the stator current space vector and the stator voltage space vector are supplied to an observer (4) who determines a reference angle on the one hand and an actual speed value on the other hand, which is fed to a determining means (10) which, taking into account the reference angle, determines a stress space vector which is transformed by a transformation element (12) into a rotor flux-oriented coordinate system and this transformed stress space vector is then made available to the inverter (2) as a control variable, characterized by , that the determining means (10) determines, from a characteristic curve field stored in a memory, an output-side value pair, i.e. voltage magnitude value and electrical frequency value, for the supplied pair of values, i.e. torque value and actual speed value, wherein the electrical frequency value is fed to an integrator, the output value of which, together with the voltage magnitude value, is fed to a further transformation element (12), which determines a voltage space vector, which is converted by the transformation element (12) into the rotor flux-oriented coordinate system and this transformed voltage space vector is then made available to the inverter (2) as a manipulated variable, wherein the characteristic curve field is determined by selecting the output-side value pair assigned to the input-side value pair, i.e., torque value and actual speed value, such that a minimum power loss occurs in steady-state operation, wherein the determining means (10) is also given a torque value, where, to determine the characteristic curve field, the output-side value pair is chosen such that the power loss is minimal, subject to the constraint that the magnitude of the voltage and the magnitude of the current remain below or equal to a respective maximum permissible value. where the torque corresponds to the torque value and the speed corresponds to the actual speed value. [2] System for operating an inverter-fed electric rotating field machine, where the motor current is measured and a stator current space vector is determined from it, wherein the stator voltage space vector is determined from at least one detected voltage value, namely the intermediate circuit voltage value, and optionally from a pulse width modulation component output by the inverter (2), wherein the stator current space vector and the stator voltage space vector are supplied to an observer (4) who determines a reference angle on the one hand and an actual speed value on the other hand, which is fed to a determining means (10) which, taking into account the reference angle, determines a stress space vector which is transformed by a transformation element (12) into a rotor flux-oriented coordinate system, and this transformed stress space vector is then made available to the inverter (2) as a control variable, characterized by , that the determining means (10) for the supplied pair of values, i.e. torque value and actual speed value, from a characteristic curve field stored in a memory, determines one output-side pair of values, i.e. two voltage space vector components of the voltage space vector of a representation in the rotor flux-oriented or stator flux-oriented coordinate system, wherein the components are transformed by the transformation element (12) into the rotor flux-oriented coordinate system, and this transformed voltage space vector is then provided to the inverter (2) as a control variable, wherein the characteristic curve field is determined by selecting the output-side value pair assigned to the input-side value pair, i.e., torque value and actual speed value, such that a minimum power loss occurs in steady-state operation, wherein the determining means (10) is also given a torque value, where, to determine the characteristic curve field, the output-side value pair is chosen such that the power loss is minimal, subject to the constraint that the magnitude of the voltage and the magnitude of the current remain below or equal to a respective maximum permissible value. where the torque corresponds to the torque value and the speed corresponds to the actual speed value. [3] System for operating an inverter-fed electric rotating field machine, where the motor current is measured and a stator current space vector is determined from it, wherein the stator voltage space vector is determined from at least one detected voltage value, namely the intermediate circuit voltage value, and optionally from a pulse width modulation component output by the inverter (2), wherein the stator current space vector and the stator voltage space vector are supplied to an observer (4) who determines a reference angle on the one hand and an actual speed value on the other hand, which is fed to a determining means (10) which, taking into account the reference angle, determines a stress space vector which is transformed by a transformation element (12) into a rotor flux-oriented coordinate system, and this transformed stress space vector is then made available to the inverter (2) as a control variable, characterized by , that the determining means (10) for the supplied pair of values, i.e. torque value and actual speed value, from a characteristic curve field stored in a memory, determines one output-side pair of values, i.e. two current space vector components of the current space vector of a representation in the rotor flux-oriented or stator flux-oriented coordinate system, wherein the components are each supplied as setpoint values ​​to a controller, to which an actual value of the torque-generating or flux-generating current component is also supplied, and whose output signal is each a component of a voltage space vector to be provided by the inverter (2), which is converted by the transformation element (12) into the rotor flux-oriented coordinate system, and this transformed voltage space vector is then made available to the inverter (2) as a manipulated variable, wherein the characteristic curve field is determined by selecting the output-side value pair assigned to the input-side value pair, i.e., torque value and actual speed value, such that a minimum power loss occurs in steady-state operation, wherein the determining means (10) is also given a torque value, where, to determine the characteristic curve field, the output-side value pair is chosen such that the power loss is minimal, subject to the constraint that the magnitude of the voltage and the magnitude of the current remain below or equal to a respective maximum permissible value. where the torque corresponds to the torque value and the speed corresponds to the actual speed value. [4] System according to any one of claims 1 to 3, characterized by , that the actual speed value determined by the observer (4) on the output side, in particular after a subsequent low-pass filtering, is supplied not only to the determining means (10), but also to a controller, to which a setpoint is also supplied and from the difference value between setpoint and actual value the torque value is determined as the output value of the controller, wherein the torque value is supplied to the determining means (10). [5] System according to at least one of the preceding claims, characterized bythat one or each of the controllers has a P-controller component, a PL-controller component and / or a PID-controller component. [6] System according to at least one of the preceding claims, characterized by , that the power loss is the sum of the stator heat losses, the rotor heat losses, the friction losses, the iron losses and the additional losses. [7] System according to at least one of the preceding claims, characterized by , that the stator current heat losses P Cu,Stator = 3R Stator (T)(|I Stator (U,f,s)|) 2 are, where the complex RMS value of the stator current I Stator is determined from an FEM calculation modeling the electric machine and R Stator (T) denotes the temperature-dependent stator resistance, where T is the temperature, where the rotor current heat losses PCu,Rotor=1P2πfRotorM(U,f,s) are, where the torque M(U,f,s) is from M(U,f,s)=3p⋅1ωsRe{(UStator−RStator(T)IStator(U,f,s))IStator*(U,f,s)} is calculated, where p is the number of pole pairs, ω s = 2πf the angular frequency of the voltage system, Re the real part of a complex quantity, U Stator the complex RMS value of the stator voltage and the operator * denotes the conjugation of a complex quantity, where the iron losses P Fe (U,f,s) are determined from the FEM calculation modeling the electric machine, where the friction losses PReib=PReib,0(nn0)C are, where P Reib,0 The friction losses at the rated speed n0 of the electric machine are described, and the exponent C reflects the speed-dependent behavior of the friction losses, with both quantities being determined by measurements on the electric machine, and where the additional losses P Zusatz can be described as a percentage of the wave power.

Images