Method for simulating multiphase electric drive for testing power electronic controller

Abstract

The invention relates to a computer-implemented method for simulating a multi-phase electric drive having an energizable branch by means of a hardware-in-loop simulator in order to test a power electronic controller having an integrated inverter. The simulator calculates the branch current in each branch of the driving device and the electric reaction potential of the non-potential branch terminal generated by the reaction of the driving device according to a mathematical model of the electric driving device, and the non-potential branch terminal is placed at the reaction potential by means of the voltage simulator. According to the method, the influence of the reaction potential which is not accurately calculated on the simulation is reduced in the following way: a mathematical model of the electric driving device is supplemented with a virtual switch in each branch; the virtual switch reduces the influence of the associated measured branch voltage on the calculated current in the respective branch in the open state and opens and / or closes the virtual switch in the respective branch of the electric drive by evaluating at least one branch voltage and / or branch current by the switching logic of the simulator.

Description

Technical Field

[0001] The present invention relates to a computer-implemented method for simulating, by means of a Hardware-in-Loop (HIL) simulator, a multiphase electric drive device having energizable branches in order to test a power electronics controller having an integrated inverter, the controller having at least three supply terminals and the inverter of the controller connecting, in a test run, one of the three supply terminals to a high inverter potential, another of the three supply terminals to a low inverter potential and yet another of the three supply terminals to ground potential in respective time intervals, connecting each supply terminal of the controller to a corresponding branch terminal of the simulator in the test run, detecting a branch voltage of the branch terminal by means of a measuring technique in the simulator, calculating, by means of the simulator, a corresponding branch current of the drive device from the measured branch voltage by means of a mathematical model of the electric drive device and its branches, determining, in the simulator, which branch terminal is connected to the ground-potential supply terminal of the controller and is thus the ground-potential branch terminal by evaluating the branch voltage and / or the branch current, determining, by the simulator, an electric reaction potential of the ground-potential branch terminal resulting from the reaction of the drive device and placing the ground-potential branch terminal at the determined reaction potential by means of a voltage simulator and feeding the calculated branch current into the non-ground-potential branch terminals by means of a current simulator. Background Art

[0002] The above computer-implemented method belongs to the technical field of real-time simulation of electrical circuits, which are currently in the form of multiphase electric drive devices for influencing or testing power electronics controllers, which are used, for example, in large numbers in motor vehicles, aircraft, energy production or energy distribution facilities, etc. The application scenario of the computer-implemented method under consideration is Hardware-in-Loop (HIL) simulation. If the computer-implemented method described at the beginning is carried out within the scope of HIL simulation, the simulation is carried out on one (or possibly also multiple) computing unit(s) of the HIL simulator by calculating a mathematical model of the electric drive device, i.e., a model in the form of numerically computable equations on an arithmetic unit. The HIL simulator simulates the technical environment for the connected power electronics controller in which the controller to be tested should later be actually used, here namely the multiphase electric drive device.

[0003] Since the current testing of the controller is carried out at the power level, the simulator includes not only the mathematical model of the drive device calculated on a suitable computing unit, but also power electronic components, namely power electronic components in the form of the voltage simulator and current simulator mentioned. In the test run, the controller is electrically connected to the simulator by connecting each supply terminal of the controller to a corresponding branch terminal of the simulator as described at the beginning.

[0004] In the simulator, the branch voltage at the branch terminal is detected by a measuring technique. The measured branch voltage is (in any case also) a numerical input parameter of the mathematical model of the electric drive device, and then the electrical and mechanical state parameters of the drive device are calculated with the aid of this mathematical model. The electrical state parameters include the branch currents generated in the respective branches of the simulated drive device. Depending on the operating state of the drive device, electrical power can be input into the simulator by the controller (motor operation of the drive device) or electrical power can also be input from the simulator into the controller (generatorisch operation of the drive device). The concept of the feed terminal of the controller should not be understood restrictively as a single direction of the energy flow.

[0005] The inverter of the controller is fed via a DC intermediate circuit, which provides a high inverter potential and a low inverter potential. By connecting the different inverter potentials to the respective feed terminals of the controller in a specific time sequence, a voltage-time area can be achieved there and thus the desired current associated with the connected coil. What is important here is that the power electronic controller with an inverter implements such an excitation scheme in which one of the feed terminals (in terms of the direct influence of the controller) is connected without potential, i.e., neither to the high inverter potential nor to the low inverter potential, but is separated from the inverter potential with a high ohmic resistance; the relevant feed terminal is "floatend" and no current can flow through the feed terminal. This excitation scheme is used to drive, for example, a permanently magnetically excited three-phase synchronous motor, which is usually a brushless DC motor (BLDC = brushless DC) in practice and is driven in a block-commutated manner.

[0006] The feed terminal of the controller is connected without potential (in terms of the action of the controller), and thus the corresponding branch terminal of the simulator connected to this feed terminal of course also does not have an electrical potential directly predefined by the controller. Rather, in a real electric drive device, the electrical potential at these terminals is determined by the reaction of the drive device (i.e., the electromotive counter voltage induced in the switched-off branch). In the simulated drive device, it is determined by evaluating the branch voltage and / or the branch current which of the branch terminals is connected to the potential-free feed terminal of the controller and is thus the potential-free branch terminal. Then, the electrical counter potential generated by the reaction of the drive device at the potential-free branch terminal is determined with the aid of the mathematical model of the drive device, and the potential-free branch terminal is set to the determined counter potential with the aid of a voltage simulator. The simulator feeds the calculated branch current into the non-potential-free branch terminals with the aid of a current simulator, and current can also flow through these branch terminals because these branch terminals are each connected to one of the inverter potentials in the controller.

[0007] The calculation of the reaction potential generated by the reaction of the drive device is mostly carried out based on specific assumptions. Usually, the prerequisite assumptions are, for example, that the simulated drive device is magnetically symmetric, that is, the inductances of the windings of each branch of the energizable drive device are equal. In the dq coordinate system fixed to the rotor with only two phases (because in the case of a star connection of each branch of the drive device, the electrical parameters of one branch of the drive device are always derived from the electrical parameters of the other two branches), this means that for the transformed inductances Ld and Lq, Ld = Lq applies. However, if this is not the case, the reaction potential may be calculated unrealistically, which may lead to unrealistically calculated branch currents on the simulator side (even in branches that are originally without potential and current). Summary of the Invention

[0008] Therefore, the object of the present invention is to design and improve a method and a corresponding simulator for simulating a multiphase drive device in such a way that the influence of inaccurately calculated reaction potentials on the simulation is reduced.

[0009] In the method described at the beginning, the object is first solved by supplementing a virtual switch in each branch of the mathematical model of the electric drive device. The virtual switch reduces the influence of the measured branch voltage belonging to it on the current calculated in the corresponding branch in the open state and the virtual switch in each branch of the electric drive device is opened and / or closed by the switch logic of the simulator by evaluating at least one branch voltage and / or branch current.

[0010] By using a virtual switch, that is, by opening the relevant virtual switch (i.e., the switch in the branch of the drive device affected by the wrong branch voltage), the influence of a wrong branch voltage (or multiple wrong branch voltages) can be reduced. The wrong branch voltage is determined by measurement technology and transmitted to the mathematical model of the drive device as the measured branch voltage and thus also affects the calculation of the branch current. Through the switch logic of the simulator and the associated evaluation of at least one branch voltage and / or branch current, the virtual switch in each branch of the electric drive device can be operated as needed, that is, opened and / or closed, in order to specifically reduce the influence of the wrong branch voltage on the mathematical model within the scope of model calculation.

[0011] In an extended version of this method, the virtual switch is represented by an ohmic resistor in the model of the drive device branch. The resistance value of the ohmic resistor is related to the switch state of the virtual switch. This enables, when the switch state of one virtual switch or several virtual switches changes, no structural conversion of the mathematical model is required. This means that the equations on which the mathematical model is based remain unchanged in their structure, and only the parameters of the model, that is, the resistance values of the virtual switches, change according to the switch state.

[0012] A preferred design of the method is characterized in that the switching logic of the simulator identifies the branch terminal with no potential, that is, the branch terminal connected to the non-potential power supply terminal of the controller, and opens the virtual switch of the branch where the branch terminal is non-potential. The problem addressed by this specific design is that the reaction potential generated by the reaction of the drive device is incorrectly calculated, and thus the electrical potential at which the voltage simulator places the non-potential branch terminal is also incorrectly set.

[0013] In addition, in an extended version of the above method, it is also specified that the switching logic of the simulator evaluates whether the preconditions for correctly determining the reaction potential at the non-potential branch terminal are met in the actual operating state of the electric drive device and opens the virtual switch of the branch where the branch terminal is non-potential when the preconditions for correctly determining the reaction potential are not met. By this measure, the virtual switch can be used in a targeted manner, that is, restricted to the case where the preconditions for correctly determining the reaction potential are not met. The switching logic can, for example, check the mathematical model of the electric drive device to determine whether the electric drive device is magnetically symmetric, that is, whether the inductances Ld and Lq in each branch of the drive device are equal (Ld = Lq) in a two-phase dq system.

[0014] According to another advantageous design of the method, the resistance value of the virtual switch is selected for the on and off switch states such that the mathematical model of the electric drive device can be stably solved by an explicit numerical solution method, especially solved at a predetermined calculation step, preferably under real-time conditions.

[0015] Alternatively, an implicit numerical method can also be used for the motor model. In this case, the resistance value can be selected within a wider range. Then, since additional computational effort is generated in each integration step (for the iterative method for solving the numerical implicit equation or the analytical method using the matrix inversion required thereby), there may be a conflict with the real-time requirements of the calculation.

[0016] Preferably, the method shown is used in combination with the simulated electric drive device, which is a multi-phase, especially three-phase, permanent magnet excited synchronous motor, which is especially a brushless DC motor. The branches of the drive device are usually interconnected in star or delta.

[0017] The above task can also be solved with the aid of a hardware-in-the-loop simulator which is used to simulate a computer-implemented polyphase electric drive device with energizable branch lines in order to test a power electronics controller with an integrated inverter, and the above method is carried out with the aid of this hardware-in-the-loop simulator. The controller does not belong to the simulator, but the operating principle of the simulator can only be understood in combination with the controller. The controller has at least three power supply terminals and the inverter of the controller connects one of the three power supply terminals to a high inverter potential, connects another of the three power supply terminals to a low inverter potential and also connects another of the three power supply terminals to be without potential during a test run, i.e., when the simulator tests the controller, each power supply terminal of the controller is connected to a corresponding branch line terminal of the simulator. The branch line voltage of the branch line terminal is detected by measurement technology in the simulator, and the simulator calculates the corresponding branch line current of the drive device based on the measured branch line voltage with the aid of a mathematical model of the electric drive device and its respective branch lines on a computing unit. In the simulator, it is determined which branch line terminal is connected to the power supply terminal of the controller that is without potential and thus is the branch line terminal without potential by evaluating the branch line voltage and / or the branch line current. The simulator determines the electrical reaction potential generated by the reaction of the drive device on the branch line terminal without potential. The simulator places the branch line terminal without potential at the determined reaction potential with the aid of a voltage simulator and feeds the calculated branch line current into the branch line terminals that are not without potential with the aid of a current simulator. The simulator is designed such that it carries out the above method during a test run, i.e., when the power electronics controller is connected.

[0018] The invention also relates to a computer program which includes instructions that cause a computing unit of a hardware-in-the-loop simulator to carry out the above method when the program is executed on the computing unit.

[0019] Specifically, there are now various possibilities for designing and improving the method according to the invention for simulating a polyphase electric drive device with energizable branch lines with the aid of a hardware-in-the-loop simulator and for designing and improving the corresponding hardware-in-the-loop simulator for testing a power electronics controller. For this purpose, reference is made on the one hand to the claims that are subordinate to the independent claim and on the other hand to the following embodiments in conjunction with the drawings. Description of the Drawings

[0020] Figure 1 Schematically shows a computer-implemented method for simulating a polyphase electric drive device with energizable branch lines with the aid of a hardware-in-the-loop simulator in order to test a power electronics controller and a corresponding simulator to which the controller is connected;

[0021] Figure 2shows the calculated branch voltages and branch currents for a magnetically asymmetric drive (Ld is not equal to Lq), once in the case of a magnetically symmetric drive being incorrectly assumed and once in the case of a drive with the magnetic asymmetry of the drive being correctly taken into account;

[0022] Figure 3 A method for simulating an electric drive using virtual switches is shown; and

[0023] Figure 4 FIG. 1 shows a case where the virtual switch is implemented as an ohmic resistor having a variable resistance value that depends on the switching state of the corresponding virtual switch. Figure 3 method. DETAILED DESCRIPTION

[0024] The drawing schematically shows different aspects of a computer-implemented method 1 for simulating a multiphase electric drive having energizable branches by means of a hardware-in-the-loop simulator 2 in order to test a power electronics controller 3 having an integrated inverter 4 .

[0025] Figure 1 A method 1 known from the prior art and a simulator 2 for implementing the method 1 are shown. The simulator 2 and the controller 3 to be tested are present as physically connected devices. The multiphase drive to be simulated does not exist per se, the drive being simulated by the simulator 2 in terms of its electrical characteristics. The arrangement thus allows testing of a power electronic controller 3, which is usually a mass-produced controller or a development controller, without the controller 3 being connected to its actual use environment and without having to provide an actual use environment at all, which is a significant advantage of hardware-in-the-loop simulation.

[0026] The inverter potential of the inverter 4 included in the controller 3 is formed by the intermediate circuit DC voltage U_DC. The controller is usually supplied with energy from the outside, in which case the intermediate circuit voltage U_DC is then derived directly or indirectly, which is not shown in detail here and is not of interest.

[0027] The controller 3 has three feed terminals 5. During test operation, the inverter 4 of the controller 3 connects one feed terminal 5u of the three feed terminals 5 to a high inverter potential, connects another feed terminal 5v of the three feed terminals 5 to a low inverter potential, and connects another feed terminal 5w of the three feed terminals 5 to a potential-free state in each time section. Figure 1 , only one switching configuration of the inverter is shown by way of example; it goes without saying that the three supply terminals 5 of the controller 3 are alternately applied with low inverter potential and high inverter potential or are connected to potential-free status in a temporal manner, which corresponds to the known operating mode of the inverter.

[0028] During the test run, the power supply terminals 5u, 5v, 5w of the controller 3 are connected to the corresponding branch terminals 6u, 6v, 6w of the simulator 2, so that the controller 3 can interact physically with the simulator 2. In the simulator 2, the branch voltages u_m,u, u_m,v, u_m,w of the branch terminals 6u, 6v, 6w are detected by measurement techniques. The simulator 2 calculates the corresponding branch currents i_m of the drive device based on the measured branch voltages u_m with the aid of the electric drive device and the mathematical model 7 of its respective branches 8.

[0029] In the present case, the simulated electric drive device is a three-phase permanent magnet excited synchronous motor. The mathematical model 7 of the electric drive device is represented by the equivalent circuit diagram of the electric drive device of three star-connected drive device branches 8 in Figure 1 Each drive device branch 8 is described in the equivalent circuit diagram by a series circuit of an ohmic branch resistance R_m, a branch inductance L_m and a counter electromotive voltage u_emf induced in the respective branch 8. The corresponding branch voltage u_m acts on the input side of each drive device branch 8, i.e. on the side of the series circuit facing away from the star point.

[0030] The equation-based transfer of the equivalent circuit thus represents the mathematical model 7, with the aid of which the state variables of the electric drive device can be calculated.

[0031] In the simulator 2, in a manner known per se, it is determined by evaluating the branch voltage u_m and / or the branch current i_m which branch terminal 6 is connected to the floating power supply terminal 5w of the controller 3 and is thus the floating branch terminal 6w. This is important for simulating the electric drive device because the floating branch terminal 6w and the energized terminals 6u, 6v are treated differently. The simulator 2 calculates the electric counter electromotive potential u_emf,w of the floating branch terminal 6w generated by the drive reaction and places the floating branch terminal 6w at the determined counter electromotive potential u_emf,w with the aid of the voltage simulator 9. Here it should be noted that the counter electromotive potential u_emf is not referenced to ground (in contrast to the branch voltage u_m), but to the common star point of the drive device branches; this is also shown in the figure. Usually, the voltage u_m on the open (offenliegend) branch terminal 6 is generated by the superposition of the voltage u_m on the connected power supply terminal 5 (or the corresponding branch terminal 6) and the counter electromotive potential u_emf applied separately by the voltage simulator. The simulator 2 feeds the calculated branch currents i_m,u, i_m,v to the non-floating branch terminals 6u, 6v with the aid of the current simulator 10. Figure 1This is also schematic in this regard, and for the sake of clarity, further details for implementing the simulation are omitted. Thus, for example, not shown is that switching devices are usually provided between the terminals of the current simulator 9 and the branch terminal 6 of the simulator 2 (and switches are provided between the terminals of the voltage simulator 9 and the branch terminal 6 of the simulator 2), and these switching devices can achieve disconnecting the current source connected to the branch terminal 6w with no potential from the branch terminal 6w with no potential, thereby preventing the current simulator 10 and the voltage simulator 9 from working against each other at a branch terminal.

[0032] As already explained at the beginning, if certain preconditions for calculating the reaction potential generated by the reaction of the drive device are not met or are not fully met, problems will occur, resulting in incorrect calculation of the reaction potential, which in turn will have an adverse impact on the calculation of the branch current. To illustrate this problem, it is considered in equation form Figure 1 the electrical conditions of the equivalent circuit shown. Starting from the following: the branches 8u and 8v are energized and the branch 8w is connected without potential on the side of the controller 3 and thus has no current (except for the current flowing briefly through the freewheeling diode in the inverter). Therefore, the branch voltage u_m,w and the reaction voltage u_emf,w are of concern. For the star point voltage, considering the branches 8u and 8v, we get (Equation 1):

[0033]

[0034] For the branch voltage u_m,w of the branch 8w connected without potential (which has no current), the relationship u_m,w = u_emf,w + u_st applies. If it is simply assumed that R_m,u is equal to R_m,v and it is known that i_m,u is equal to -i_m,v, then as follows (Equation 2):

[0035]

[0036] If only the fundamental wave of the magnetic flux in the drive device is considered and based on a magnetically symmetric motor, the inductance L_m can be regarded as the same constant based on the symmetric characteristics of the three-phase system, that is, L_m,u = L_m,v = Lm,w. In the dq coordinates fixed on the rotor, which are usually used for the mathematical description of the electric drive device, Ld = Lq applies. The voltage drops caused by the inductance cancel each other out, so the assumption for the magnetically symmetric motor applies (Equation 3):

[0037]

[0038] This relationship of course applies to each branch connected without potential, whether it is the branch 8u, 8v or 8w. Figure 1In the case of the switch, u_m and v correspond to the low inverter potential and usually to the electrical device ground (elektrischen Gerätemasse), and u_m and u correspond to the high inverter potential. This is not a constant DC voltage but a high-frequency PWM signal in order to be able to adjust the branch current within a wide range.

[0039] In the embodiment according to Figure 1 the reaction potential of the branches, which are respectively floatingly grounded from the perspective of the controller 3, is calculated according to Equation 3 and the voltage simulator 9 applies the corresponding voltage to the relevant floating branches.

[0040] Figure 2 Shows a comparison of the calculated branch voltage u_m and the calculated branch current i_m for a actually magnetically asymmetric drive device (Ld not equal to Lq), once under the false assumption of a magnetically symmetric drive device (applying Equation 3, curves u_m, i_m) and once under the correct consideration of the magnetic asymmetry of the drive device (applying Equation 2, curves u_m,ref, i_m,ref). The upper left curve shows the branch voltage over a period of more than one inverter cycle, within which, i.e., all branches experience all switching states of the inverter 4 twice. The voltage curves u_m and u_m,ref appear as shaded areas here because the applied voltage is actually a high-frequency PWM signal, i.e., the voltage jumps back and forth between the upper and lower envelopes. In Figure 2 the lower figure below, the two voltages u_m and u_m,ref are shown together in an enlarged detail so that it can be seen that the voltage is variable at high frequencies. It can also be seen in the lower figure that the calculations of the voltages are significantly different from each other. This also affects the calculation of the current, as can be seen in Figure 2 the upper right figure in

[0041] In Figure 3 Method 1 and simulator 2 are shown, by means of which the influence of the wrongly calculated reaction potential u_emf described above can be greatly reduced and even avoided. The illustration of the simulator 2 basically corresponds to Figure 1Illustration of the simulator 2 in [reference], but the controller is omitted. To solve the problem, a virtual switch 11 is added to the mathematical model 7 of the electric drive device in each branch line 8. These virtual switches 11 reduce the influence of the measured branch line voltage u_m of the corresponding branch line on the calculated current i_m in the corresponding branch line 8 in the open state. The virtual switches 11 in each branch line 8 of the electric drive device are opened and / or closed by the switch logic 12 of the simulator 2 by evaluating at least one branch line voltage u_m and / or branch line current i_m.

[0042] In the shown case, the switch logic 12 of the simulator 2 is designed such that the switch logic opens the virtual switch 11 of the branch line 8 whose branch line terminal 6 is at zero potential. If starting from the Figure 1 conditions in [reference], the supply terminals 5u, 5v of the controller 2 are placed at a defined electrical potential, which thus also applies to the branch line terminals 6u, 6v of the simulator 2. In addition, the supply terminal 5w of the controller 2 is connected at zero potential, which thus also applies to the branch line terminal 6w of the simulator 2. The switch logic 12 has recognized these relationships and thus closes the virtual switches 11u and 11v and opens the virtual switch 11w.

[0043] The switch changes the circuit structure in principle because usually circuit elements are activated or deactivated by using the switch. Therefore, depending on which switches in the switch are open or closed, the equation-based description of the circuit also changes, so the circuit is structurally variable and different mathematical models must be considered to calculate the circuit.

[0044] In the method 1 according to Figure 4 and the design of the simulator 2, the virtual switches 11 in the models of each branch line 8 of the drive device are each represented by an ohmic resistor 12, and the resistance value R_sw of this ohmic resistor depends on the switch state of the virtual switch 11. Thus, the mathematical model 8 of the drive device is structurally invariant because the mathematical description of the drive device does not change with the switch state of the virtual switch 11 (regardless of the switch state, the ohmic resistor 12 always exists), and only the parameter of the model, i.e., the resistance value R_sw, depends on the switch state.

[0045] By again describing the equivalent circuit diagram of the electric drive device equationally (as already according to Figure 1As done, it can well represent the effect of the virtual switch 11 in the form of the resistor 12, except that the resistance value R_sw of the ohmic resistor 12 must be additionally considered. The starting point is still that the controller 3 connects the branch terminal 6w without potential. The resistance values R_sw,u and R_sw,v are set to zero because the corresponding resistors 12u and 12v represent the closed virtual switches 11u, 11v. Additionally, an internal branch voltage u'_m is introduced, which indicates the voltage directly behind the resistor 12 of the virtual switch 11, and the branch voltage u_m acts directly on the other end of the resistor 12 of the virtual switch 11.

[0046] The current in the branch 8w connected floatingly according to the preconditions is obtained as follows (Equation 4):

[0047]

[0048] The internal branch voltage u'_m,w of the branch 8w connected without potential is obtained as follows (Equation 5):

[0049]

[0050] If the resistance value R_sw,w is selected to be infinitely large, the reaction voltage of the wrong simulation (which is subsequently detected again by measurement technology and used as an input parameter of the model 7 of the electric drive device) no longer has an impact on the current calculation. From Equation 4, i_m,w = 0 is obtained. Therefore, Equation 5 becomes (Equation 6):

[0051]

[0052] Equation 6 corresponds to the general result according to Equation 2 and thus represents the correct solution of the magnetically asymmetric electric drive device (equivalent to Ld not equal to Lq in the coordinates fixed on the rotor).

[0053] In numerical reality, the resistance value R_sw,w cannot be selected to be infinitely large, so the branch current i_m,w is not equal to zero. In the embodiment according to Figure 4 , the calculation is not performed according to Equation 6, but after converting to u'_m,w according to Equation 4.

[0054] The switch logic 12 (which ensures the correct operation of the switch 11 in time or the correct change of the resistance value R_sw of the ohmic resistor 12 in time) is implemented in Figure 4 as known per se for the identification of the wiring change at the branch. Based on this, the voltage simulator 9 and the current simulator 10 are instructed to energize the energized branch and apply the reaction potential u_emf to the floating branch disconnected by the controller 3.

[0055] A variant of Method 1 and Simulator 2 (not shown in detail here) stipulates that the switching logic 12 of Simulator 2 evaluates whether the preconditions for correctly determining the reaction potential u_emf,w at the unpowered branch terminal 6w are met in the actual operating state of the electric drive device and opens the virtual switch 11w of the branch 8w whose branch terminal 6w is unpowered when the preconditions for correctly determining the reaction potential u_emf are not met. Specifically, it is selected that the electric drive device is magnetically symmetric as a precondition for correctly determining the reaction potential u_emf at the unpowered branch terminal 6w.

[0056] In Method 1 and Simulator 2 according to Figure 4 , the resistance value R_sw of the virtual switch 11 in the form of an ohmic resistor 12 is selected for the open and closed switch states such that the mathematical model 7 of the electric drive device can be stably solved by means of an explicit numerical solution method. In particular, the value is selected in consideration of a predetermined calculation step size such that the calculation can be performed in real time. List of reference numerals

[0057] 1 Computer-implemented method

[0058] 2 Hardware-in-the-loop simulator

[0059] 3 Controller

[0060] 4 Inverter of the controller

[0061] 5 Feeding terminal of the controller

[0062] 5u, 5v Non-unpowered feeding terminals

[0063] 5w Unpowered feeding terminal

[0064] 6 Branch terminals of the simulator

[0065] 6u, 6v Non-unpowered branch terminals

[0066] 6 Unpowered branch terminal

[0067] 7 Mathematical model of the electric drive device

[0068] 8 Mathematical models of the branches of the drive device

[0069] 9 Voltage simulator

[0070] 10 Current simulator

[0071] 11 Virtual switch

[0072] 12 Switching logic for operating the virtual switch

[0073] U_DC Intermediate circuit voltage

[0074] Branch voltage of u_m

[0075] Branch current of i_m

[0076] Branch resistance of R_m

[0077] Branch inductance of L_m

[0078] Star point voltage of u_st

[0079] Electromotive reaction potential of u_emf

[0080] Ohmic resistance of R_sw for simulating virtual switch

[0081] Internal branch voltage of u'_m

Claims

1. A computer-implemented method (1) for simulating a multiphase electric drive having an energizable branch line by means of a hardware-in-the-loop simulator (2) in order to test a power electronic controller (3) having an integrated inverter (4), wherein the controller (3) has at least three feed terminals (5) and the inverter (4) of the controller (3) connects one of the three feed terminals (5) to a high inverter potential, connects another of the three feed terminals (5) to a low inverter potential and connects another of the three feed terminals (5) to a potential-free state in each time section during a test operation, wherein each feed terminal (5) of the controller (3) is connected to a corresponding branch line terminal (6) of the simulator (2), the branch line voltage (u_m) of the branch line terminal (6) is detected in the simulator (2) by means of measurement technology, and the simulator ( 2) using a mathematical model (7) of the electric drive and its branches (8) the corresponding branch current (i_m) of the drive is calculated based on the measured branch voltage (u_m), in the simulator (2) by evaluating the branch voltage (u_m) and / or the branch current (i_m) it is determined which branch terminal (6) is connected to the potential-free supply terminal (5w) of the controller (3) and is therefore the potential-free branch terminal (6w), the simulator (2) determines the electrical reaction potential (u_emf,w) of the potential-free branch terminal (6w) caused by the reaction of the drive and uses a voltage simulator (9) to set the potential-free branch terminal (6w) to the determined reaction potential (u_emf,w), and the simulator (2) uses a current simulator (10) to feed the calculated branch currents (i_m,u), (i_m,v) into the non-potential-free branch terminals (6u), (6v), characterized in that The mathematical model (7) of the electric drive is supplemented in each branch (8) by a virtual switch (11), which, in the open state, reduces the influence of the associated measured branch voltage (u_m) on the calculated current (i_m) in the corresponding branch (8). Virtual switches (11) in the individual branches (8) of the electric drive are opened and / or closed by a switching logic (12) of the simulator (2) by evaluating at least one branch voltage (u_m) and / or branch current (i_m).

2. The method (1) according to claim 1, characterized in that The virtual switch (11) is represented in a model of the branch line (8) of the drive by an ohmic resistor (12), the resistance value (R_sw) of which is dependent on the switching state of the virtual switch (11).

3. The method (1) according to claim 1 or 2, characterized in that: The switching logic (12) of the simulator (2) opens the virtual switch (11w) of the branch line (8w) whose branch line terminal (6w) is potential-free.

4. The method (1) according to claim 3, characterized in that The switching logic (12) of the simulator (2) evaluates whether the prerequisites for correctly determining the reaction potential (u_emf,w) at the potential-free branch terminal (6w) are met in the actual operating state of the electric drive, and opens a virtual switch (11w) of the branch (8w) whose branch terminal (6w) is potential-free if the prerequisites for correctly determining the reaction potential (u_emf) are not met.

5. The method (1) according to claim 4, characterized in that A prerequisite for the correct determination of the reaction potential (u_emf) at the potential-free spur terminal (6w) is that the electric drive is magnetically symmetrical.

6. The method (1) according to any one of claims 1 to 5, characterized in that The resistance value (R_sw) of the virtual switch (11) in the form of an ohmic resistor (12) is selected for the open and closed switching states so that the mathematical model (7) of the electric drive can be solved stably by means of an explicit numerical solution method, in particular with a predetermined calculation step size, preferably under real-time conditions.

7. The method (1) according to any one of claims 1 to 6, characterized in that The simulated electric drive is a multi-phase, in particular a three-phase, permanently excited synchronous machine, in particular a brushless DC motor.

8. A hardware-in-the-loop simulator (2) for computer-implemented simulation of a multiphase electric drive having an energizable branch line in order to test a power electronic controller (3) having an integrated inverter (4), the controller (3) having at least three feed terminals (5) and the inverter (4) of the controller (3) connecting one of the three feed terminals (5) to a high inverter potential, another of the three feed terminals (5) to a low inverter potential and another of the three feed terminals (5) to a potential-free state in each time section during test operation, the feed terminals (5) of the controller (3) being connected to corresponding branch line terminals (6) of the simulator (2), the branch line voltage (u_m) of the branch line terminal (6) being detected in the simulator (2) by means of measurement technology, the simulator ( 2) using a mathematical model (7) of the electric drive and its branches (8) the corresponding branch current (i_m) of the drive is calculated based on the measured branch voltage (u_m), in the simulator (2) by evaluating the branch voltage (u_m) and / or the branch current (i_m) it is determined which branch terminal (6) is connected to the potential-free supply terminal (5w) of the controller (3) and is therefore the potential-free branch terminal (6w), the simulator (2) determines the electrical reaction potential (u_emf) of the potential-free branch terminal (6w) caused by the reaction of the drive and sets the potential-free branch terminal (6w) to the determined reaction potential (u_emf,w) using a voltage simulator (9), and the simulator (2) feeds the calculated branch currents (i_m,u), (i_m,v) into the non-potential-free branch terminals (6u), (6v) using a current simulator (10), It is characterized in that The simulator (2) is designed such that it carries out the method (1) according to any one of claims 1 to 7 in a test mode, ie in connection with a power electronics controller (3).

9. A computer program comprising instructions which, when a computing unit of a hardware-in-the-loop simulator (2) executes the program, cause the computing unit to execute the method (1) according to any one of claims 1 to 7.

Images