Method for controlling the operation of an inverter

Abstract

Method for controlling the operation of an inverter (3) for outputting an electrical signal to an electrical machine (4) of a motor vehicle (1), wherein the inverter (3) has at least three switch pairs (5-7), each having a first and a second switch (8-13), wherein the first and the second switch (8-13) of each switch pair (5-7) are alternately switched on and off based on a control signal for generating the electrical signal, wherein the switch pair (5-7) carrying the largest current is determined and the first switch (8-13) or the second switch (8-13) of the determined switch pair (5-7) is switched through for at least two switching periods carried out one after the other, characterized in that after the expiration of the at least two switching periods, in particular the defined number of switching periods,the switch (8-13) switched off during the switching process receives the higher proportion of an on-time for generating the electrical signal for at least two switching periods than the previously switched-on switch.

Description

[0001] The invention relates to a method for controlling the operation of an inverter for outputting an electrical signal to an electrical machine of a motor vehicle, wherein the inverter has at least three switch pairs, each having a first and a second switch, wherein the first and the second switch of each switch pair are alternately switched on and off based on a control signal for generating the electrical signal.

[0002] Methods for controlling the operation of inverters of electrical systems in motor vehicles, in particular for outputting an electrical signal to an electrical machine of the motor vehicle, are generally known from the prior art. In such control methods, the inverter is operated, for example, using a PWM method to output a specific electrical signal, specifically an electrical current, to the electrical machine for a specific operating point, so that the electrical machine can set a specific speed and torque at the specified operating point.

[0003] During the described generation of the electrical signal, the first and second switches of the switch pairs are operated alternately in the PWM method. As is well known, a dead time must be maintained between the first switch being turned off and the second switch being turned on in order to prevent short circuits that could be harmful to the inverter components. A dead time is also maintained between the second switch being turned off and the first switch being turned on.

[0004] DE 10 2011 017 705 A1, for example, discloses a method for operating a rotating field machine on an inverter by means of voltage vectors to be set on the rotating field machine, wherein the amounts of the phase currents on the rotating field machine are determined during a respective PWM period, wherein a power switch of the inverter is switched on over an entire PWM period of the subsequent PWM period as a function of the phase current amounts determined during a respective PWM period.

[0005] It is also known that, depending on the operating point of the electrical machine, different currents must be carried by the individual switches for different lengths of time. If, for example, the electrical machine is at an operating point at which the electrical machine has a comparatively high speed, the times during which the currents flow through the switches are correspondingly shorter compared to operating points at which the electrical machine has a comparatively low speed or is operating in the standstill range or at a standstill. If the electrical machine is operated at such low speeds in the standstill range, the time during which a particular switch carries a particular current is comparatively longer. In particular, the switch pair which carries the highest current at the current operating point also generates the highest switching losses when the two switches are switched alternately.As is well known, these not only reduce the efficiency of the inverter but also result in the switches heating up.

[0006] The invention is based on the object of providing an improved method for controlling the operation of an inverter, in which in particular the efficiency is increased and the heating is reduced.

[0007] The object is achieved by a method having the features of claim 1. Advantageous embodiments are the subject of the subclaims.

[0008] As described, the invention relates to a method for controlling the operation of an inverter for outputting an electrical signal to an electrical machine, in particular a rotating field machine for use in a motor vehicle. In such rotating field machines (PSM, ASM, etc.), voltage source inverters, for example, can be used to generate the necessary manipulated variables, for example voltages or currents. The first and second switches of the individual switch pairs, of which the inverter can in principle have any number, for example three or six, depending on the design of the electrical arrangement comprising the electrical machine and inverter, can in principle be designed in any way, for example as MOSFETs, IGBTs, thyristors, GTOs and the like.

[0009] As also previously described, the two complementary switches of the same switch pair must never be switched on at the same time, as this would lead to a short circuit, which could damage or destroy the DC link capacitor and the switches, so that a locking time or dead time or blocking time must usually be taken into account between the switching off of one switch and the switching on of the complementary switch of the same switch pair.

[0010] To control an electrical machine, the phase currents are typically measured to establish feedback in a control loop. As described, the control of an electrical machine can generally be operated using a pulse width modulation (PWM) method, whereby PWM values can be generated to generate a specific voltage vector that changes with the desired voltage frequency and voltage amplitude. The amplitude can be used as a measure of the length of the voltage vector.

[0011] As mentioned above, switching losses occur when implementing the PWM method when switching the switches due to the non-ideal switching between the first and second switches of the same switch pair. The first switch can also be referred to as the upper switch ("high-side") and the second switch as the lower switch ("low-side") (or vice versa).

[0012] Since it is known that the switching losses are formed from the product of the voltage applied to the switch and the current flowing through it, and the voltage is essentially determined by the intermediate circuit voltage on the intermediate circuit capacitor, this means that the greater the current through the switch, the greater the switching losses at the switch of the switch pair.

[0013] Consequently, the switch that generates the greatest switching losses is the one that carries the greatest current at the current operating point.

[0014] Particularly at an operating point at standstill or close to standstill, i.e. at a stator frequency below a defined frequency value, a comparatively large phase current flows through a switch or through a pair of switches compared to the at least two other switch pairs. If this condition persists over a defined period of time, this leads to significant heating of the switch pair in addition to switching losses, whereas the at least two other switch pairs heat up far less and also generate significantly lower losses. In addition to the switching losses described, this leads to higher thermal loading of the switch pair carrying the highest current compared to the other switch pairs. The invention is therefore fundamentally based on the discovery of reducing or avoiding switching losses.

[0015] To this end, the invention proposes that the switch pair carrying the greatest current is determined in the current operating state, and that the first switch or the second switch of the determined switch pair is switched through for at least two switching periods carried out in succession. As described, the switch pair which currently carries the greatest current is identified for the current operating point. The identification of the switch pair carrying the greatest current can, for example, be carried out by means of a current detection which is already implemented. The first or second switch of the determined switch pair can then be switched through for at least two, in particular a plurality of, switching periods. The switching period is basically the time period orthe time interval in which the inverter switches between the first switch and the second switch of the switch pairs, that is, within the framework of the PWM process, switches off the previously switched on first switch and after the dead time switches on the previously switched off second switch or vice versa.

[0016] The switching of the first switch or second switch of the specific switch pair thus results in that switch remaining switched on and the other switch remaining switched off. For example, if the first switch is switched on for several switching periods, the first switch remains switched on, whereas the complementary second switch of the same switch pair remains switched off for the same number of switching periods. During the switched switching periods, the other switches of the other switch pairs alternate according to the usual PWM method. In particular, this means that while the first switch or the second switch of the specific switch pair is switched on for several switching periods, the other switches of the other switch pairs are alternately switched on and off.

[0017] Advantageously, by switching the first or second switch through, alternating switching is eliminated for the at least two consecutive switching periods, thus also eliminating the associated switching losses. This can significantly improve the efficiency of the inverter's operation, particularly at comparatively high currents that can occur during standstill or at stator frequencies below a defined frequency value.

[0018] Here, the switch pair that carries the highest current is explicitly identified, since the highest current also generates the highest switching losses. The other switch pairs, in contrast, carry significantly lower currents and thus also generate significantly lower switching losses during continued operation in the PWM process. Since the switch pair that carries the highest current is identified and remains switched on for at least two consecutive switching periods, in particular a large number of switching periods, the switching losses generated by the saved switching operations are also saved. In addition to determining the switch pair with the highest or lowest current, the switch pair with the highest or lowest PWM value can be determined.

[0019] As described, the first switch or the second switch of the specific switch pair can be switched on for at least two switching periods carried out one after the other. According to one embodiment of the method, it can be provided that the first switch or the second switch is switched on for a defined number of switching periods, in particular as a function of the current and / or a temperature and / or a temperature profile of the switch. In principle, it is thus possible to take into account which current is carried by the switch pair, so that ultimately a number of switching periods can also be determined as a function of the switching losses realized. The temperature or temperature profile of the switch carrying the current can also be taken into account, so that it is taken into account that the current flow through the switch also causes the switch to heat up.The currents and thus the temperature profiles are determined, for example, by the rotor position or the operating state of the electric machine. The currents and / or temperatures, or current and temperature profiles, can be determined or recorded by suitable sensors and taken into account when determining the operation of the inverter.

[0020] According to a further embodiment of the method, it can be provided that after the expiration of at least two switching periods, in particular the defined number of switching periods, the switch that was switched off during the switching operation receives a higher proportion of an on-time for generating the electrical signal for at least two switching periods than the previously switched switch. In other words, after the first or second switch has been switched on, i.e. after the expiration of at least two switching periods or the plurality of switching periods in which the first or second switch was switched on, the regular PWM process is to be returned to.

[0021] As is well known, a specific ratio between the on-time and off-time is set depending on the ratio or PWM value to be achieved by the switch pair. The previously switched-off switch should bear the greater portion of the on-time, so that the previously switched-on switch, which was subjected to greater load due to the switching, is relieved and can, for example, cool down. In other words, the switch that was previously switched off for several switching periods is assigned the higher "on portion" of the PWM signal, whereas the previously switched-on switch is assigned the correspondingly lower "on portion" of the PWM signal.

[0022] According to a further development of the method, it can be provided that the switch pair with the second-largest current in terms of magnitude is determined, and the first or second switch of the switch pair with the second-largest current in terms of magnitude is switched through for the at least two consecutive switching periods after the first or second switch of the switch pair carrying the largest current has been switched through. The embodiment proposes that the switch pair carrying the second-largest current is also identified, wherein after the at least two switching periods or the plurality of switching periods, generally a first time interval in which the first or second switch of the switch pair carrying the largest current was switched through, the first or second switch of the switch pair with the second-largest current in terms of magnitude is switched through in a second time interval.

[0023] In other words, this approach allows for the switch pair, or one of the switches within the switch pair, that carries the highest current to be switched on in a first time interval to avoid the highest switching losses. Subsequently, in a second time interval, a switch within the switch pair that carries the second-highest current can be switched on in order to avoid at least the second-highest losses, while the switch pair carrying the highest current is relieved, particularly thermally. The first time interval and the second time interval can be performed alternately.

[0024] In both cases, the switch pair carrying the third-highest current continues to execute the regular PWM process. Since this is the switch pair carrying the lowest current, especially with three phases, the associated switching losses are also the lowest and thus easiest to tolerate. In this case, it is possible to alternate between switching the switch pair carrying the highest current and the switch pair carrying the second-highest current in order to at least alternately avoid the highest and second-highest switching losses. This also allows for alternating thermal equalization, so that heating that occurs when the first or second switch in a switch pair is switched on can be reduced.

[0025] As described, the number of switching periods for switching through a switch can be suitably set in order to reduce switching losses, wherein the switching through of the same switch should be limited with regard to heating due to the conduction of the comparatively large current. According to one embodiment, the number of switching periods for switching through a switch can be set based on at least one switch parameter, in particular based on a switch model and / or temperature model and / or based on predefined temperature data and / or an active temperature measurement. In principle, a time interval can thus be set or the number of switching periods can be set for which a specific switch may be switched through. This can ultimately be based on a switch model that describes the current at which the switch may be operated and for how long.how long a particular switch may conduct a particular current. Since this primarily depends on the magnitude of the current and the switch used, the number of switching periods can be determined based on the switch model, the current currently being conducted, and / or the switch's operating parameters.

[0026] Likewise, a temperature model can be created for the at least one switch, which, for example, based on historical data, takes into account heating or a temperature input into the switch. Predefined temperature data can also be taken into account, for example based on previous measurements that describe heating of the switch as a function of different currents for specific time periods. Another possibility is to provide active temperature measurement. This allows the temperature of the switch to be actively measured or recorded, so that the heating of the switch due to switching through can be actively measured. This makes it possible to measure during operation of the switch whether the switch can still be switched through or whether the previous switching through has already heated the switch above a specific temperature threshold.Active temperature measurement offers the advantage of actively determining the actual temperature of the switch independently of models and thus being able to thermally load the switch without exceeding a certain limit temperature.

[0027] The number of switching periods for switching through the first or second switch of the switch pair carrying the highest current and / or for switching through the first or second switch of the switch pair carrying the second highest current can be specified as a function of a limit temperature. The limit temperature can, for example, be the temperature of the current-carrying switch when switching through that rules out damage or premature aging of the switch. In other words, the switch can be operated up to the defined limit temperature without being damaged. In principle, the switch of the switch pair carrying the second highest current can be operated or switched through for longer than a switch of the switch pair carrying the highest current, since the comparatively lower current also leads to less heating.Thus, the number of switching periods for the highest current-carrying switch pair and the second highest current-carrying switch pair may differ. However, the limit temperatures may be the same for each switch.

[0028] As already described several times, during switching, a pair of switches is operated continuously, i.e., over several switching periods, with a PWM value of 100% or 0%, so that one of the two switches is permanently loaded and the other is completely unloaded for the duration of the switching, i.e., one switch carries the entire current while the other switch carries no current. To prevent the switch carrying the current from overheating, thus ruling out the possibility of a reduction in service life or damage to the switch, the thermal load on the switches should be equalized. For this purpose, a first and second time interval can be defined, or a load time can be defined for which switching is carried out at its maximum. Ultimately, the first time interval is determined by the number of switching periods for which switching is carried out. The first time interval is therefore calculated as follows: ΔT1=n1∗TPWM

[0029] Here, T PWM the PWM period, i.e., the duration of the switching period, of the inverter. This is equal to the inverse of the inverter's frequency. The value n1 thus determines the number of switching periods for which one of the switches can be switched on, so that the time interval is derived from the number of switching periods and the duration of the switching period. As previously described, the time interval can be selected so that a desired limit temperature is not exceeded, i.e., a switch is only switched on long enough to prevent it from heating up above a desired limit temperature.

[0030] As also described above, to relieve the load on the previously switched-on switch, another pair of switches, namely the pair carrying the second-highest current, can be switched on. Similarly, the second time interval for switching on the second pair of switches is determined by: ΔT2=n2∗TPWM with ΔT2=ΔT1∗PWMZPWMZ−PWMmaxmod

[0031] Here PWM Z for the average PWM value for the modulation method used. At this PWM value for a phase, a voltage equal to 0 is generated in this phase. Ideally, the average PWM value is 50%. The PWM value maxmodis a modeled PWM value of the phase or switch pair that carries the largest phase current, i.e., the switch pair to which the largest PWM value is assigned, in the time interval ΔT2. As previously explained, the PWM value of the switch pair that carries the largest current is switched through and thus receives a PWM value of PWM maxmod = 100%.

[0032] Similarly, the PWM value for the phase with the highest phase current can be corrected to 0% if the phase is assigned the lowest PWM value. The complementary switch should conduct the current for longer in the second time interval than the other switch was conducting, so that the previously conducting switch is relieved. Since the switch conducting the highest current is subjected to a comparatively greater load when conducting than a switch in another switch pair when conducting, the second time interval should be extended accordingly. This results, for example, from: ΔT2=ΔT1∗PWMZPWMZ−PWMminmod

[0033] Modifying the PWM value of the phase that originally carries the largest current in the operating state would generate distortions with respect to the desired voltage vector or current vector. To prevent such distortions, the control signals for the other switch pairs can be adjusted to maintain the originally set voltage vector. In other words, the method can provide that when one switch in a switch pair is switched on, the control signals of the other switch pairs are adjusted, in particular such that the voltage vector defined by the original control signal is maintained.

[0034] As also described above, the control signals of the switch pair carrying the highest current can be adjusted, in particular by switching one of the switches of the switch pair carrying the highest current through. Subsequently, the switch pair carrying the second highest current can be switched through, particularly for thermal relief, so that the corresponding control signals of the remaining switch pairs must be adjusted. This can prevent distortion and oscillations.

[0035] As described above, the switch pair that should carry the largest current in the current operating state can be determined first. This switch pair will, as described above, have the largest PWM value (or the smallest PWM value) and will be switched through, so that the previously largest PWM value is increased to 100%, ie, PWM maxmod = 100% (or alternatively PWM minmod= 0%). In order to avoid distortions in the electrical current or voltage, the other two control signals are corrected or recalculated and are obtained for the switch pair that carries the second largest current (PWM mid is corrected to PWM midmon ) and the pair of switches that carries the lowest current (PWM min is corrected to PWM minmod ) to: PWMminmod=(PWMmaxmod−PWMmax)+PWMmin PWMmidmod=(PWMmaxmod−PWMmax)+PWMmid

[0036] Furthermore, the phase with the smallest PWM value can be set to 0% in order to switch on the first or second switch, especially the “lower switch”, so that the phase with the smallest PWM value is: PWMminmod=0%

[0037] To avoid any distortion in the current or voltage, the other two phases are corrected. This results in: PWMmaxmod=PWMmax−(PWMmin−PWMminmod) PWMmidmod=PWMmid−(PWMmin−PWMminmod)

[0038] Similarly, the corresponding phase currents can be compared to determine the phase or switch pair with the largest phase current, the smallest phase current, and the average phase current. The phase with the largest phase current should be unloaded, as described above.

[0039] As described above, the method is intended to be carried out in particular in operating states of the electrical machine in which the electrical machine is operated close to standstill, i.e. at a comparatively low speed and / or stator frequency. One embodiment can in particular provide that the switching through of at least one switch is carried out as a function of the speed and / or the stator frequency, in particular below a defined limit speed. Thus, in operating states close to standstill or at a comparatively low speed, the generation of switching losses or heating of individual switches can be effectively prevented or at least avoided. The definition of the limit speed can differ depending on the electrical machine used, the inverter and the operating point. For example, the limit speed can be in the range of 1% of the nominal speed.In principle, this can also be defined depending on the current vector or the current magnitude. It can be determined that the higher the current magnitude, the higher the limit speed can be selected at which the process should be carried out.

[0040] The invention also relates to a control device for controlling the operation of an inverter for outputting an electrical signal to an electrical machine of a motor vehicle, wherein the control device is designed to carry out the method described above. Furthermore, the invention relates to a motor vehicle comprising such a control device. The control device, the inverter, and the electrical machine can be components of an electrical arrangement of the motor vehicle. The electrical machine can be used in particular as a traction drive. For example, the electrical arrangement can be components of an electric axle or electric axle or an electric axle drive.

[0041] The invention is explained below using an exemplary embodiment with reference to the figures. The figures are schematic representations and show: Fig. 1 a motor vehicle; Fig. 2 an exemplary flow chart; and Fig. 3 an example diagram of control signals of an inverter.

[0042] Fig. 1 shows, by way of example, a motor vehicle 1 having a control device 2, an inverter 3, and an electric machine 4. The control device 2, the inverter 3, and the electric machine 4 can be part of an electrical arrangement, for example, an electric axle drive or a so-called "E-axle." In other words, the electric machine 4 is used as a traction drive to drive the motor vehicle 1.

[0043] In principle, the control device 2 can generate control signals for the inverter 3. The inverter 3 is operated in particular using a PWM method, wherein an exemplary diagram of the PWM values is shown in Fig. 3. The inverter 3 is selected merely as an example, with the inverter 3 and the electric machine 4 being three-phase in this embodiment. In principle, any desired number of phases can be implemented, for example, six phases instead of three.

[0044] In the exemplary embodiment shown, the inverter 3 has three switch pairs 5-7, each having a first switch 8-10 and a second switch 11-13. The individual switch pairs 5-7 are assigned to the respective phases. During operation of the inverter 3, the first switches 8-10 and the second switches 11-13 are switched alternately within the framework of the basically known PWM method, whereby a dead time or blocking time is observed between switching on and off the switches 8-10, 11-13 of the same switch pair 5-7. The first switches 8-10 can also be referred to as the upper switch, whereby the second switches 11-13 can also be referred to as the lower switch. In principle, the assignment of the terms "first" and "second" can be changed as desired.

[0045] Depending on the control signals that the inverter 3 receives from the control device 2, the inverter 3 provides output variables for the electrical machine 4, in particular currents or phase currents of the electrical machine 4, whereupon the electrical machine 4 can generate or provide a rotational speed or a torque.

[0046] In particular, in operating states of the electric machine 4 in which the electric machine 4 is operated at a standstill or at an operating point close to a standstill, in particular at frequencies below a defined frequency value or a defined limit speed, the switching losses in the switches 8-13 increase. The control device 2 is designed to carry out the method, which is schematically shown as a flow chart in Fig. 2, in order to reduce switching losses and avoid excessive heating of the switches 8-13. Various sensors, which are not explicitly shown for the sake of simplicity, can be assigned to the control device 2 for this purpose. Such sensors, for example, temperature sensors, current sensors, voltage sensors, rotor position sensors, and the like, are generally known from the prior art, which is why such sensors and their mode of operation will not be discussed in detail.

[0047] The method starts in a block 14, in which various values of the electrical arrangement are supplied to the control device 2. In particular, the PWM values for the individual switch pairs 5-7 for the current operating state are supplied. Likewise, in block 14, the phase currents for the current operating state are supplied to the control device 2. In a block 15, the control device 2 can then compare the supplied PWM values and thus determine the largest PWM value, the average PWM value or the second-largest PWM value, and the smallest or third-largest PWM value in a block 16.

[0048] Similarly, a comparison or evaluation of the phase currents can be performed in block 17, and the largest phase current, the second largest phase current, and the smallest phase current can be determined in block 18. The determination results from blocks 16 and 18 can then be fed to block 19, which determines whether the largest phase current is assigned to the largest or smallest PWM value. In other words, block 19 determines whether the switch pair 5-7 intended to carry the largest phase current in the current operating state is operating with the largest or smallest PWM value.Depending on the result of the determination from block 19, the process branches from block 19 to block 20 if the switch pair 5-7 that carries the largest phase current also has the largest PWM value, or it branches from block 19 to block 21 if the switch pair 5-7 that carries the largest phase current has the smallest PWM value.

[0049] Starting from block 20, i.e. in the case where the switch pair 5-7, which carries the largest phase current, is also assigned the largest PWM value, in block 22 the largest PWM value, which is also known as PWM max can be described as PWM maxmod = 100%. To avoid distortions caused by changing the largest PWM value, the other PWM values are adjusted as described above. For an initial number of switching periods or a first time interval, the PWM values can be determined as follows: PWMminmod1=(PWMmaxmod1−PWMmax)+PWMmin PWMmidmod1=(PWMmaxmod1−PWMmax)+PWMmid

[0050] For the second time interval we get: PWMminmod2=0% PWMmaxmod2=(PWMmax−PWMmin−PWMminmod2) PWMmidmod2=PWMmid−(PWMmin−PWMminmod2)

[0051] Subsequently, in a block 23, the time intervals ΔT1 and ΔT2, in which the switch 8-13 carrying the largest current remains switched on and then the switch 8-13 carrying the second largest current is switched on, can be determined as follows: ΔT1=n1∗TPWM ΔT2=n2∗TPWM with ΔT2=ΔT1∗PWMzPWMz−PWMmaxmod

[0052] As described, the time intervals ensure, or rather the time intervals are selected, that switch 8-13 of switch pair 5-7, which carries the highest current when switched on, is relieved of load in the second time interval following the first time interval, while the complementary switch 8-13 of the same switch pair 5-7 does not carry any current. The time intervals can be selected so that a desired limit temperature is not exceeded. To even out the load, a different modulation of the PWM values can be selected in the second time interval. In this case, switching does not occur in switch pair 5-7, which carries the highest phase current, but rather a switch 8-13 of switch pair 5-7, which carries the second highest current value and is assigned, for example, the smallest PWM value, is switched on.This can be set to 0% as described above, whereby the other PWM values are again compensated to obtain the originally desired voltage pointer.

[0053] In this case, the times for conducting the phase currents in the switch pair 5-7, which conducts the largest phase current, can be determined to be the same for both switches over the entire time. The time intervals can then be repeated.

[0054] If the switch pair 5-7, which carries the largest phase current, has the smallest PWM value, the program branches from block 21 to block 24, in which the PWM values for the individual switch pairs 5-7 for the first time interval are determined as follows: PWMminmod1=0% PWMmaxmod1=PWMmax−(PWMmin−PWMminmod1) PWMmidmod1=PWMmid−(PWMmin−PWMminmod1)

[0055] For the second time interval we get: PWMmaxmod2=100% PWMmidmod2=(PWMmaxmod2−PWMmax)+PWMmin PWMmidmod2=(PWMmaxmod2−PWMmax)+PWMmid

[0056] Accordingly, the time intervals are: ΔT1=n1∗TPWM ΔT2=n2∗TPWM with ΔT2=ΔT1∗PWMZPWMZ−PWMminmod

[0057] Analogous to the first case described above, for the second case, in which the phase with the largest phase current has the smallest PWM value, this value is set to 0%, and to prevent distortion, the remaining PWM values are adjusted. The time intervals, or rather the PWM value settings in the time intervals, are therefore reversed. The description is therefore transferable.

[0058] Depending on the case, the algorithm can then branch from block 23 or block 25 to block 26, where the values determined in blocks 22-25 are output to inverter 3. The algorithm can then end in block 27. The algorithm can run at specified intervals, periodically, continuously, or event-based, or start in block 14.

[0059] Fig.3 shows an example diagram of the adjusted PWM values, with the method starting at a time T0, wherein in an exemplary first time interval ΔT1, phase 28, which carries the largest phase current, with the lowest PWM value is set to 0% and the remaining phases 29, 30 are determined and adjusted according to the equations presented above. In a second time interval ΔT2 following the first time interval ΔT1, phase 30 with the second largest phase current and the largest PWM value is set to 100% and the remaining phases 28, 29 are adjusted according to the equations presented above. The time intervals ΔT1 and ΔT2 can alternate. During the time intervals, switching losses are reduced and negative thermal effects are compensated for by the alternating switching.

[0060] In other words, it is possible to achieve an optimal reduction in switching losses in the switches when the machine is at standstill or at low stator frequencies without causing distortions in the voltages and phase currents. The method described does not result in any power losses or torque fluctuations. The electrical arrangement has an extended service life thanks to the method described. The reduction in power loss increases efficiency. Furthermore, due to the reduction in switching losses and the described implementation of the switching, the cooling effort is reduced. By preventing the introduction of distortions into the phase currents, the reactive power is reduced. By adjusting the modulation of the PWM values, no harmonics are introduced into the phase currents, thus reducing acoustic problems.The described method can be used in particular for rotating field machines and switches of any type. Reference symbol 1 motor vehicle 2 Control device 3 inverters 4 electric machine 5-7 pair of switches 8-10 first switch 11-13 second switch 14-27 Blocks 28-30 phase

Claims

[1] Method for controlling the operation of an inverter (3) for outputting an electrical signal to an electrical machine (4) of a motor vehicle (1), wherein the inverter (3) has at least three switch pairs (5-7), each having a first and a second switch (8-13), wherein the first and the second switch (8-13) of each switch pair (5-7) are alternately switched on and off based on a control signal for generating the electrical signal, wherein the switch pair (5-7) carrying the largest current is determined and the first switch (8-13) or the second switch (8-13) of the determined switch pair (5-7) is switched on for at least two switching periods carried out in succession, characterized bythat after the expiry of the at least two switching periods, in particular the defined number of switching periods, the switch (8-13) switched off during the switching through receives the higher proportion of a switch-on time for generating the electrical signal for at least two switching periods than the previously switched through switch. [2] Method according to claim 1, characterized by that the switching through of the first switch (8-13) or second switch (8-13) is carried out for a defined number of switching periods, in particular depending on the current and / or a temperature and / or a temperature profile of the switch (8-13). [3] Method according to one of the preceding claims, characterized bythat the switch pair (5-7) with the second largest current in terms of magnitude is determined and the first or second switch (8-13) of the switch pair (5-7) with the second largest current in terms of magnitude is switched through for at least two switching periods carried out in succession after the first or second switch (8-13) of the switch pair (5-7) carrying the largest current has been switched through. [4] Method according to one of the preceding claims, characterized by that the number of switching periods for switching through a switch (8-13) is determined based on at least one switch parameter, in particular based on a switch model and / or temperature model and / or based on predefined temperature data and / or an active temperature measurement. [5] Method according to one of the preceding claims, characterized bythat the number of switching periods for switching through the first or second switch (8-13) of the switch pair (5-7) carrying the highest current and / or for switching through the first or second switch (8-13) of the switch pair (5-7) carrying the second highest current is determined as a function of a limit temperature. [6] Method according to one of the preceding claims, characterized by that when one switch (8-13) of a switch pair (5-7) is switched on, the control signals of the other switch pairs (5-7) are adapted, in particular in such a way that the voltage vector defined by the original control signal is retained. [7] Method according to one of the preceding claims, characterized by that the switching through of at least one switch (8-13) is carried out as a function of the speed and / or the stator frequency, in particular below a defined limit speed. [8] Control device (2) for controlling the operation of an inverter (3) for outputting an electrical signal to an electrical machine (4) of a motor vehicle (1), characterized by that the control device is designed to carry out the method according to one of the preceding claims. [9] Motor vehicle (1) comprising a control device (2) according to the preceding claim.

Images