Method for operating a hybrid-multilevel inverter, hybrid-multilevel invert, and computer program

Abstract

The invention relates to a method for operating a hybrid-multilevel inverter according to the preamble of claim 1. The invention also relates to a hybrid-multilevel inverter and to a computer program for carrying out the method.

Description

[0001] Method for operating a hybrid multilevel converter, hybrid multilevel converter and computer program

[0002] The invention relates to a method for operating a hybrid multilevel converter according to the preamble of claim 1. The invention also relates to a hybrid multilevel converter and a computer program for implementing the method.

[0003] Introduction

[0004] The topology of the hybrid multilevel converter is derived from the conventional T-type topology and the Q3L-MMC topology by adding a branch with full-bridge modules between the midpoint of the converter phase and the DC link neutral point. The process is based on the control method according to [1, 21], in which the branch currents and branch energies are actively controlled, ensuring small output voltage steps and the shutdown of the conventional switches at very low currents. This combines the advantages of conventional MMC topologies (voltage scalability, redundancy, small output voltage steps) and the output voltage curve of quasi-three-level operation, while simultaneously simplifying the topology and significantly reducing the required module capacity. This promises improvements in volume, power density, efficiency, and cost. Figures

[0005] Figure 1.1 : Two-level inverter for low voltage applications on the left and medium voltage applications on the right

[0006] Figure 1 .2: Example of pulse width modulation of the target output voltage u a * over one output period for two-level inverters, see [4]

[0007] Figure 1 .3: Three-phase T-type inverter [6]

[0008] Figure 1 .4: MMC topology with three inverter phases

[0009] Figure 1 .5: Half-bridge modules and full-bridge modules

[0010] Figure 1 .6: Single-phase model of the MMC

[0011] Figure 1.7: Example of a multilevel output voltage

[0010]

[0012] Figure 1 .8: Quasi-two-level operation of a single-phase MMC over a PWM period

[0013] Figure 1 .9: Topology of the single-phase MMC for quasi-three-level operation [2]

[0014] Figure 1.10: Output voltage curve in quasi-three-level operation over two output periods [2]

[0015] Figure 1 .11 : Module capacitor voltages top: Two A; middle: branch B; bottom: branch C [2]

[0016] Figure 2.1 : Single-phase circuit arrangement for quasi-three-level operation

[0017] Figure 2.2: Single-phase circuit arrangement for quasi-n-level operation

[0018] Figure 2.3: Voltage and current curves in quasi-three-level operation over one output period

[0019] Figure 2.4: Time-stretched enlargement of Figure 2.3

[0020] Figure 2.5: Control structure of the cascaded control

[0021] Figure 2.6: State machine of branch current control

[0022] Figure 2.7: Procedure for switching off the first converter branch

[0023] Figure 2.8: Procedure for switching on the first converter branch

[0024] 1. State of the art

[0025] This chapter introduces the converter topologies and operating modes relevant to this patent application. First, two-level inverters, three-level inverters, and MMCs are presented in conventional and quasi-two-level operation. Then, quasi-three-level operation according to [2] is presented.

[0026] 1.1 Two-point inverter

[0027] The two-level inverter is the most common inverter in low-voltage applications.

[0028] The two-level inverter is shown on the left in Figure 1.1 and is characterized by its simple control and design. The two-level inverter has six switching elements 1 (here, for example, an IGBT with an antiparallel diode), each of which converts the input voltage U eThe maximum blocking voltage of IGBTs is currently 6.5 kV, which limits voltage scalability. To overcome this, multiple switching elements 1 can be connected in series to minimize the voltage load on each individual switching element 1. This is shown in Figure 1.1 on the right. The challenge here is the uniform voltage distribution and simultaneous control of the switching elements 1 . This requires special, highly developed gate units [3],

[0029] Redundancy can only be achieved with press-pack switches, as these enter a defined short circuit in the event of a fault. The desired sinusoidal output voltage is achieved by pulse width modulation (PWM) as a short-term average. In this case, either the upper or the lower switch is switched on. This results in the typical two-point output voltage curve of the desired output voltage u, as shown in Figure 1.2. a *

[0030] Due to this two-stage modulation, the output voltage and output current are subject to relatively high harmonics, depending on the switching frequency, which can lead to additional losses in electrical machines. Since the medium-voltage semiconductor switching elements can only be switched at switching frequencies of a few hundred Hertz, the harmonics are relatively high. Also due to the two-stage operation, the output voltage jump with each switching operation is at the level of the input voltage. This voltage jump leads to wave reflection in long cables to the electrical machine. As a result, the insulation of the machine is loaded with up to twice the input voltage, which significantly increases the demands on the machine's insulation systems [5]. These problems are even more significant in medium-voltage applications, where the insulation systems for the machines are already very demanding.

[0031] Accordingly, this topology is not suitable for applications with electrical machines in the medium-voltage range. The limiting factor is not so much the harmonic current, which leads to slightly increased losses, but rather the increased insulation stress caused by line reflections, which can lead to premature machine failures and is therefore unacceptable.

[0032] Therefore, an additional filter is often installed at the output of the two-level inverter, which increases the size and cost of the entire system.

[0033] 1.2 Three-point inverter

[0034] With three-level inverters, the output voltage can be modulated across three voltage levels (Ue / 2, 0 V, -Ue / 2), which improves the harmonic characteristics of the output signals and halves the output voltage jump compared to a two-level inverter. This reduces the load on the electrical machine's insulation system.

[0035] Topologies for three-level inverters frequently mentioned in the literature are NPC inverters, flying capacitor inverters and T-type inverters [4]. The T-type inverter is described below as an example.

[0036] The design of a T-type inverter is shown in Figure 1.3. A capacitive voltage divider (Ci, C2) forms the neutral point (NP), to which another branch with a bidirectional power electronic switch is connected. This increases the number of switching elements. During operation, voltage balancing of the input capacitors must also be ensured, which increases the control effort [6].

[0037] To generate the three output voltages, either the upper switch, the lower switch or the middle switch is switched on, while the other switches are blocked.

[0038] The upper and lower switches must be able to block the entire input voltage. The middle switches, on the other hand, must be able to block half the input voltage [6]. For voltage scalability, several switches can be connected in series, as with a two-level inverter, but this poses the same challenges regarding voltage sharing and simultaneous control.

[0039] Compared to two-level inverters, the harmonic characteristics of the output signals are improved and the output voltage jump is halved. However, the number of power electronic switching elements increases, and the task of neutral point balancing is added. Voltage scalability presents the same challenges as with two-level inverters.

[0040] Compared to two-level inverters, three-level inverters can significantly reduce harmonic currents, and voltage jumps at the inverter output are halved. With modern insulation systems, a medium-voltage drive can thus be operated directly at the inverter output without a filter [7]. However, for motors with conventional insulation systems (including existing machines in the field that are to be retrofitted with an inverter, so-called retrofit applications), filters between the inverter and motor are also required.

[0041] Harmonics are no longer a limiting factor here, but voltage surges in combination with motor insulation are. 1.3 Modular Multilevel Inverter

[0042] The modular multilevel converter (MMC) [8] is shown schematically in Figure 1.4.

[0043] An MMC consists of several converter phases (typically three) connected to the input. Each of these converter phases contains two branches. The output system is connected between the branches of each converter phase. The branches consist of several n mO d modules and a branch inductance L z [9], As shown in [5], the inductors can be coupled together.

[0044] The most commonly used modules are half-bridge (HB) and full-bridge (VB) modules. These are shown in Figure 1.5.

[0045] Half-bridge modules consist of two switches and a capacitor. Depending on the switching state, the module voltage Umod can be either 0 V or the module capacitor voltage ucmod.

[0046] Full-bridge modules contain two additional switches, allowing the module voltage to assume 0 V, the positive module capacitor voltage ucmod and the negative module capacitor voltage (- ucmod).

[0047] If the modules are connected in series, as shown in Figure 1.4, high operating voltages can be achieved, while the power semiconductors in the modules only need to be designed for the module capacitor voltage. This enables simple voltage scaling.

[0048] Redundancy can be implemented relatively easily by adding several modules in series and bridging the defective modules in the event of a fault. The topology is therefore more complex than that of two-level and three-level inverters, and the control system is also more demanding. It is necessary to ensure that the module capacitor voltages are balanced during operation, which is achieved using branch modulation techniques.

[0049] For a simplified description, the relevant quantities are marked on the single-phase model of the MMC in Figure 1.6. The upper branch is referred to as branch A and the lower as branch B. The branch voltage is denoted by u z and the branch current with i z The output voltage of the inverter is indicated by u a and the output current with i a described.

[0050] 1.4 Conventional operation HO]

[0051] In conventional operation, the output voltage is finely modulated. This results in the output voltage curve shown in Figure 1.7.

[0052] This fine-tuned modulation results in lower harmonic loads on the output voltage and current, eliminating the need for an additional filter. It also reduces the voltage jump at the output terminals compared to two-level and three-level inverters, minimizing the voltage stress on the insulation of the electrical machine.

[0053] During DC / AC operation, branch powers p z which oscillate at one or two times the output frequency. This generally leads to relatively large branch energy fluctuations e z , which must be buffered in the module capacitors. The ratio H of the energy stored in the modules to the apparent power of the MMC can be used as a measure for this. For MMCs in conventional operation, this is approximately 55 mJ / VA [5]. Accordingly, the module capacitors for conventional MMC operation must be dimensioned comparatively very large. Consequently, the module capacitors represent a large portion of the converter's volume, weight, and cost.

[0054] The branch energy fluctuation and the associated module capacitor capacitance required are inversely proportional to the output frequency. To prevent the capacitor voltages from increasing excessively, which could lead to the destruction of the power semiconductors in the modules, special operating modes must be used in conventional operation at low frequencies close to zero. A typical operating mode is the "Low-Frequency Mode" by Korn et al., presented in

[0011] . Although the energy fluctuation in the capacitors is significantly reduced by this operating mode, as

[0012] shows, it leads to a significantly increased current load on the power semiconductors when high currents are required at low frequencies. This has a negative impact on dimensioning and efficiency.

[0055] 1.5 Quasi-two-point PWM operation M, 5, 13-15]

[0056] In quasi-two-level PWM operation (Q2L) according to

[0013] , the conventional MMC from Figure 1.4 is operated similarly to a two-level inverter (see Figure 1.8). The output voltage is also modulated by the voltage levels U e / 2 and -U e / 2. For this purpose, one branch represents approximately the value of the full input voltage (active branch) and the other branch represents 0 V (passive branch).

[0057] The basic idea is that the majority of the output current should always be conducted via the passive branch so that the branch power is as small as possible and correspondingly low values ​​for module capacitance are required.

[0058] The distribution of the output current between the branches is determined by the phase

[0059] The cross-phase current IQ (see Figure 1.6) is set, which flows through both branches as a superposition current. In quasi-two-level PWM operation, the phase cross-phase current is actively controlled.

[0060] For this purpose, the target branch voltage in the active branch is generated using high-frequency modulation. If the active and passive branches are to alternate, the branch voltages are set in

[0001] such that the phase shunt current changes as quickly as possible. For this purpose, either both branches generate 0V or the maximum branch voltage level. This is referred to below as a transient transition.

[0061] However, during transient transitions, high branch voltages overlap with high branch currents. This leads to power spikes that must be buffered by the capacitors and push the capacitor voltages away from the desired setpoints.

[0062] To compensate for this, a small current is injected into the active branch based on the phase cross-current to compensate for the branch energy swing (compensation current). This represents the second fundamental idea of ​​quasi-two-level PWM operation. The current level for compensation is set via a branch energy controller or calculated predictively.

[0063] Thus, the branch energies can be regulated within each PWM period, and the branch energy fluctuation is independent of the output frequency. Therefore, the module capacity required is significantly reduced (by more than an order of magnitude compared to conventional operation).

[0064] A waiting time is implemented between the connection and disconnection of the modules within a branch to ensure continued low output voltage steps. This limits the rate of rise of the output voltage and significantly reduces overvoltages caused by long machine cables [5].

[0065] The modular design, easy-to-implement redundancy, voltage scalability, and low output voltage jumps are retained in quasi-two-level PWM operation. This operation also enables a significant reduction in module capacitance, leading to a significant cost reduction and increased power density.

[0066] The disadvantage compared to conventional operation is the deteriorated harmonic spectrum of the output voltage, which can, however, be accepted in many machines as long as the frequency of the PWM is not too low.

[0067] It should be noted here that in addition to the quasi-two-level PWM operation presented here, there are also quasi-two-level operations that do not actively regulate the currents and energies, e.g. [16-20]. This usually leads to additional disadvantages such as higher peak currents in the branches, strongly influenced operating behavior by parasitic parameters (e.g. losses and stray inductances of the connection technology) and increased module capacitances.

[0068] 1.6 Modular multilevel converter for quasi-three-level operation [21

[0069] Quasi-three-level operation for MMCs was first introduced in 2018 in [2] and takes up the idea of ​​quasi-two-level operation. Here, the output voltage is to be divided into the voltage levels U e / 2, 0 V and -U e / 2 modulation to improve the converter's output spectra at the same switching frequency. Since both branches must provide half the input voltage at the 0 V output voltage level, an additional branch is necessary to allow the output current to be routed via a passive branch.

[0070] The topology presented in Figure 1.9 consists of half-bridge modules in branches A and B of a converter phase and of full-bridge modules in the additional middle branch C (here: clamped arm). The topology is introduced as a "quasi three-level hybrid modular multilevel converter", where "hybrid" refers to the use of half-bridge and full-bridge modules. In this topology, the number of full-bridge modules to be installed in the middle branch C corresponds to half the number of half-bridge modules to be installed in the upper and lower branches A, B. To set the correct output voltage, the branch voltages are switched directly from one static state to the next. Accordingly, the branch currents are not actively regulated, but oscillate with the resonant circuit behavior of the connected module capacitors and branch inductance (similar to that described at the end of Chapter 1.5).

[0071] The output voltage waveform has three static output voltage levels, as shown in Figure 1.10.

[0072] A waiting time is also implemented between the switching on or bridging of the individual modules on the edges in order to ensure the small output voltage levels.

[0073] Since no compensation currents are implemented, the energy swings due to transient transitions are not compensated, resulting in a smaller module capacitor voltage fluctuation range than in conventional MMC operation, but still dependent on the output frequency. This can also be seen in Figure 1.11.

[0074] Furthermore, it should be noted that the branch currents have not been shown in [2] and their peak values ​​are expected to be very high.

[0075] Topologies and control methods for quasi three-level inverters based on the MMC are also known from patents DE 10 2020 108 034 B3 and DE 10 2020 108 035 B3. According to the proposals therein, voltage and current waveforms such as those shown in Figure 2.3 can be generated to synthesize the AC voltage. Likewise, the replacement of the MMC branch C with simple power electronic switches was proposed and filed in patent DE 10 2020 108 034 B3, in conjunction with the advantageous regulated branch currents. This variant is called Q3L hybrid MMC.

[0076] In addition, WO 2019 / 238443 A1 discloses further circuits of quasi-multilevel

[0077] Converters based on the MMC are known that can generate a higher number of stationary voltage levels at the output (QnL hybrid MMC, QnL stands for quasi-n-level).

[0078] 2. Invention

[0079] Compared to the prior art, further improvements and advantages can be realized by the features specified in the independent claims of this patent application.

[0080] The invention relates to a hybrid multilevel converter that delivers electrical output power comprising an output voltage and an output current, wherein the output power is transmitted by means of a circuit arrangement that a) has at least a first, a second, and a third converter branch, wherein b) the first converter branch is connected to a first input potential, c) the second converter branch is connected to a second input potential that differs from the first input potential, and d) the third converter branch is connected to a third input potential that lies between the first and the second input potential, wherein e) the first, the second, and the third converter branch are connected to one another at an output terminal of the circuit arrangement, to which the output power is transmitted,wherein f) the first and second converter branches each consist of a switch and at least one or no inductance, wherein g) the third converter branch is designed as a multilevel converter branch with several individual modules connected in series and at least one or no inductance.,

[0081] The invention also relates to a method for operating such a hybrid multilevel converter. The invention thus assumes that a modular multilevel converter is simplified in terms of its design by having the first and second converter branches each consist of a switch and one or no inductance. Only the third converter branch is designed as a multilevel converter branch with several individual modules connected in series. Accordingly, the hardware complexity for two half-bridge multilevel converter branches is eliminated, and only a comparatively simple circuit is provided for the first and second converter branches.

[0082] Existing two-level inverters can also be expanded to this topology with the additional third MMC branch and capacitive voltage divider.

[0083] In addition to the series connection of the individual modules, the third converter branch can also contain additional components, such as at least one series-connected inductor. The individual modules of the third converter branch can be identical or different modules. The individual modules of the third converter branch can be designed as full-bridge modules (VB) or similar. It is important that the individual modules can generate a positive voltage, a negative voltage, and zero volts.

[0084] The first and second converter branches, each consisting of a switch and at least one or no inductance, can be implemented with relatively simple control logic, since a switch only has the states "on" and "off." All types of switches can be used, in particular voltage-unidirectional, current-bidirectional switches or voltage-bidirectional, current-bidirectional switches. For cost reasons, implementation using voltage-unidirectional, current-bidirectional switches is preferable. Such a switch can be designed, for example, as a power semiconductor switching element with an antiparallel-connected freewheeling diode, e.g., as an IGBT or IGCT switching element with a freewheeling diode. The switch can also be designed as an arrangement of several power semiconductor switching elements with corresponding freewheeling diodes, e.g.,by two anti-serially connected power semiconductor switching elements, each with a freewheeling diode, to implement a voltage-bidirectional switch, or by a series connection of several similar switches to increase the blocking voltage strength. With the three converter branches mentioned, quasi-3-level operation can be implemented. The method can also be applied to embodiments in which additional converter branches are present, allowing quasi-n-level operation, as explained below.

[0085] One embodiment of the invention relates to a method for operating a hybrid multilevel converter of the type explained above.According to the invention, it is proposed that the operation of such a converter is controlled with the following features: h) the current intensity of the output current is controlled by a modulation method of the output voltage, wherein only one of the converter branches is always operated as a current-transmitting converter branch through which the output current flows and, accordingly, the output power is delivered, and the first, second and third converter branches alternately function as a current-transmitting converter branch, i) wherein, during a transition in which the third converter branch receives or loses its assignment as a current-transmitting converter branch, a transition of the output current is carried out by controlling the individual modules of the third converter branch from the converter branch previously functioning as a current-transmitting converter branch to the converter branch subsequently functioning as a current-transmitting converter branch.

[0086] In particular, in feature i), during a transition in which the third converter branch maintains or loses its assignment as a current-transmitting converter branch, the control of the individual modules of the third converter branch can be carried out depending on the branch current of the third converter branch, thus enabling a transition of the output current from the converter branch previously functioning as a current-transmitting converter branch to the converter branch subsequently functioning as a current-transmitting converter branch. The branch current of the third converter branch can be determined in real time by current measurements.

[0087] This allows for highly efficient commutation of a relatively simple and cost-effective converter hardware configuration, which requires only one multilevel converter branch. This type of control of the converter branches allows the converter to be used very universally. By controlling the transitions when switching the first and second converter branches on and off, using appropriate current regulation by the third converter branch, the transitions between the different voltage levels can be optimized.

[0088] The aforementioned term "current-transmitting converter branch" refers to the converter branch through which the output current flows and, accordingly, via which the output power is delivered. Currents can then also flow in the other converter branches, although these are generally significantly lower than the output current. If, for example, the first or second converter branch is operated as a current-transmitting converter branch, in an advantageous embodiment of the invention, at least a small compensation current flows in the third converter branch (and also through the current-transmitting converter branch) in order to minimize the energy fluctuation in the module capacitors of the individual modules of the third converter branch within the framework of the energy control explained below.

[0089] According to an advantageous embodiment of the invention, it is provided that during a transition in which the third converter branch receives its assignment as a current-transmitting converter branch (i.e. becomes the current-transmitting converter branch), by controlling the individual modules of the third converter branch, a positive current flowing at the beginning of the transition through the first or second converter branch previously acting as the current-transmitting converter branch or a negative current flowing at the beginning of the transition through the first or second converter branch previously acting as the current-transmitting converter branch is controlled to a small negative current (close to 0 amperes), so that the freewheeling diode of the current-transmitting converter branch is brought into the conducting state, and then the switch of the first or second converter branch previously acting as the current-transmitting converter branch is switched off.In this way, the switch can be switched particularly gently, since a very low voltage (forward voltage of the anti-parallel diode) is applied to the switch, resulting in hardly any power loss. If, at the beginning of the transition, a positive current flows through the first or second converter branch, which previously functioned as the current-transmitting converter branch, the current is controlled towards a small negative current during the transition process. If, at the beginning of the transition, a negative current flows through the first or second converter branch, which previously functioned as the current-transmitting converter branch, the current is controlled towards a negative current with a reduced magnitude during the transition process.

[0090] According to an advantageous embodiment of the invention, during a transition in which the third converter branch receives its assignment as a current-transmitting converter branch (i.e., becomes the current-transmitting converter branch), by controlling the individual modules of the third converter branch, a positive current flowing at the beginning of the transition through the first or second converter branch, which previously functioned as the current-transmitting converter branch, or a negative current flowing at the beginning of the transition through the first or second converter branch, which previously functioned as the current-transmitting converter branch, is controlled to a small positive current (close to 0 amperes), and then the switch of the first or second converter branch, which previously functioned as the current-transmitting converter branch, is switched off. This also makes it possible to realize the advantages explained above.

[0091] The aforementioned small negative or positive current can be small in the sense that it is significantly lower than the average negative / positive current through the affected converter branch, for example, only a maximum of one-tenth of the average current. In practice, the aforementioned small current may be closer to one-hundredth of the average current. At least the freewheeling diode of the current-transmitting converter branch should be brought into conduction.

[0092] According to an advantageous embodiment of the invention, after the switch of the first or second converter branch, which previously functioned as the current-transmitting converter branch, is switched off, the current flowing through the first or second converter branch, which previously functioned as the current-transmitting converter branch, is gradually increased until it reaches zero by controlling the individual modules of the third converter branch (by generating a voltage which differs from the previous static voltage level by a maximum of the voltage of an individual module), in particular in such a way that the output voltage only changes by one module capacitor voltage when the zero crossing is reached and the freewheeling diode begins to block. In this way, particularly small voltage steps at the output of the converter can be ensured even when the switch is switched off.

[0093] According to an advantageous embodiment of the invention, it is provided that before the switch of the first or second converter branch, which previously functioned as the current-transmitting converter branch, is switched off, the current flowing through the third converter branch is gradually increased or decreased with a time gradient by controlling the individual modules of the third converter branch.

[0094] According to an advantageous embodiment of the invention, during a transition in which the third converter branch loses its assignment as a current-transmitting converter branch, the voltage applied to the switch of the first or second converter branch, which then functions as a current-transmitting converter branch, is controlled to a reduced amount by controlling the individual modules of the third converter branch, and then the switch of the first or second converter branch, which then functions as a current-transmitting converter branch, is switched on. In this way, particularly small voltage steps are ensured at the output of the converter even when the switch is switched on. For example, the voltage applied to the switch to be switched on can be controlled to a minimum possible amount, i.e. the smallest possible amount that is possible with the individual modules of the third converter branch, namely a module capacitor voltage.

[0095] According to an advantageous embodiment of the invention, it is provided that after the switch of the first or second converter branch, which subsequently functions as the current-transmitting converter branch, is switched on, the current flowing through the third converter branch is gradually reduced or increased with a time gradient by controlling the individual modules of the third converter branch. According to an advantageous embodiment of the invention, it is provided that the method has an output current control that regulates the current intensity of the output current by a modulation method of the output voltage, wherein the output current control is subordinate to an energy control of the third converter branch, by which temporal energy fluctuations in the module capacitors of the individual modules of the third converter branch are minimized. This has the advantage that the individual modules can be realized with relatively small module capacitors, i.e.The capacitance values ​​of the capacitors do not need to be particularly large. This allows for a particularly cost-effective implementation of the converter.

[0096] According to an advantageous embodiment of the invention, a branch current control is subordinate to the energy control, by which current setpoints specified by the output current control and / or the energy control are converted into actual currents flowing in the converter branches by high-frequency clocked switching on and off of individual modules of the third converter branch. Through such subordinate, particularly high-frequency actuation of the individual modules, the desired current setpoints can be implemented particularly precisely, i.e., the occurring control deviations are minimized. For the control of the branch currents, these are measured according to an advantageous embodiment of the invention and compared with setpoint values.

[0097] According to an advantageous embodiment of the invention, the branch current control is operated as a state machine which has at least three static operating states and at least four transition states via which it is possible to switch between different static operating states. The four transaction states thus capture all transitions between the static operating states. With such a state machine, a defined case differentiation can be realized in the actuation states of the individual switches and individual modules of the converter branches. Depending on the current operating state, i.e. one of the three static operating states or one of the four transition states, specific control and / or regulation measures can be carried out in a targeted manner, for example to ensure the previously explained favorable switching conditions with small currents close to 0A.In the state machine, for example, a state variable indicates the current operating state of the state machine. Depending on the value of this variable, the corresponding function can then be executed, e.g., one of the functions according to one of claims 2 to 6. In the transition states, functions according to one of claims 2 to 6 are executed, in particular. In the static operating states, control is carried out according to one of claims 7 and / or 8.

[0098] According to an advantageous embodiment of the invention, the output current control is operated with a clocking with a first period, and the branch current control is operated with a clocking with a second period, wherein the second period is significantly shorter than the first period. The first period can be significantly shorter than the period of the output current. For example, the period of the output current can be a multiple of the first period. The first period can be a multiple of the second period.

[0099] According to an advantageous embodiment of the invention, the bidirectional switch comprises an arrangement of power semiconductor switching elements that can accommodate unidirectional blocking voltage and conduct current symmetrically. In this way, a voltage-unidirectional and current-bidirectional switch with a simple design can be provided cost-effectively.

[0100] The above-mentioned method can advantageously be carried out by a computer program by executing the computer program on a computer.

[0101] For example, it could be a computer in a converter's control unit. This can also realize the advantages discussed above.

[0102] The invention also relates to a hybrid multilevel converter of the type described above, wherein the circuit arrangement comprises at least one control device configured to control the power semiconductors of the converter branches, wherein the control device is configured to carry out a method of the type described above. This also allows the previously described advantages to be realized.

[0103] According to an advantageous embodiment of the invention, each individual module has a circuit arrangement connected to the module terminals of the individual module, consisting of at least four internal switching elements and a module capacitor. The individual modules can be designed as full-bridge modules. In the case of full-bridge modules, the module output voltage can be 0 volts or a positive or negative module capacitor voltage.

[0104] The invention also relates to a hybrid multilevel converter of the type described above, wherein the circuit arrangement has one or more further converter branches, wherein each further converter branch is connected on one side to a further input potential, which differs from the first, second, and third input potentials as well as from all other further input potentials, and on the other side to the output terminal of the circuit arrangement, to which the output power is transmitted, directly or indirectly via further switching elements. In this way, an advantageous QnL hybrid multilevel converter can be realized, i.e., a converter with more than 3 stages (n > 3).For a particularly cost-effective implementation, it is proposed that each additional converter branch consist of a switch and at least one or no inductance. Bidirectional voltage and bidirectional current switches are particularly suitable for this purpose. This also keeps the control logic of the switches simple. The switch can, for example, comprise an arrangement of power semiconductor switching elements that can symmetrically accommodate blocking voltage and conduct current.

[0105] The hybrid multilevel converter, also referred to as converter for short, can be connected to three independent input potentials (first, second, and third input potentials). The third input potential can also be formed from the first and second input potentials, e.g., by a capacitive voltage divider. According to an advantageous embodiment of the invention, the third input potential is provided by a circuit arrangement comprising a first capacitor, which is connected by a first terminal to the first input potential and a second terminal to the third input potential, and a second capacitor, which is connected by a first terminal to the second input potential and a second terminal to the third input potential.

[0106] Analogously, the aforementioned additional input potentials for the QnL hybrid multilevel converter can be provided by adding additional capacitors to the series circuit. For example, a respective additional input potential, to which a respective additional converter branch is connected, can be provided by a voltage divider circuit arrangement formed from series-connected capacitors.

[0107] For the purposes of the present invention, the indefinite term "a" is not to be understood as a numerical word. Therefore, if, for example, a component is mentioned, this is to be interpreted as "at least one component." Where angles are given in degrees, these refer to a circular dimension of 360 degrees (360°). Where a computer is mentioned, it may be configured to execute a computer program, e.g., in the sense of software. The computer may be a commercially available computer, e.g., a PC, laptop, notebook, tablet, or smartphone, or a microprocessor, microcontroller, or FPGA, or a combination of such elements. Where closed-loop control is mentioned, closed-loop control differs from open-loop control in that closed-loop control has a feedback or feedback of measured or internal values, which in turn influences the generated output values ​​of the control in the sense of a closed-loop control.In a control system, a quantity is simply controlled without any such feedback or loopback.

[0108] Implementation examples

[0109] Figure 2.1 shows a single-phase circuit arrangement for quasi-three-level operation of a multilevel converter. If the multilevel converter is designed as a single-phase converter, Figure 2.1 shows the entire converter. For example, in the case of a three-phase multilevel converter, a circuit arrangement as shown in Figure 2.1 is required for each of the three phases, although the same input potentials (and thus the same input capacitors) can be used for all three phases.

[0110] It can be seen that the multilevel converter according to Figure 2.1 has a first converter branch A, a second converter branch B and a third converter branch C. The first converter branch A is connected on the input side to a first input potential, shown here as a high input potential. The second converter branch B is connected to a second input potential, shown here as a low input potential. The third converter branch C is connected to a third input potential, shown here as a middle input potential, which lies between the first and second input potentials, e.g. exactly midway between these potentials. On the output side, the three converter branches A, B and C are connected to one another and to an output terminal of the converter.A load connected to an output terminal is then connected on the other side to the third input potential or to another converter phase or, in the case of a three-phase system, is connected in star or delta.

[0111] It can be seen that the first converter branch A is designed as a unidirectional voltage, bidirectional current switch 3. The second converter branch B is also designed as a unidirectional voltage, bidirectional current switch 3. The third converter branch C is designed as a multilevel converter branch with several individual modules 2 connected in series. Such a design is also referred to as a 2-hybrid MMC. In the illustrated embodiment, a quasi-3-level variant, such a converter is also referred to as a Q3L-2-hybrid MMC.

[0112] Figure 2.2 shows a further development of a multilevel converter compared to Figure 2.1. Here, too, there is a first converter branch A with a voltage unidirectional, current bidirectional switch 3 and a second converter branch B with a voltage unidirectional, current bidirectional switch 3. Furthermore, the aforementioned third converter branch C is also present as a multilevel converter branch with several individual modules 2 connected in series. In addition to the first input potential, to which the first converter branch A is connected on the input side, and the middle input potential immediately following in Figure 2.1, one or more additional input potentials are present in between, with Figure 2.2 showing an example of an additional potential in between.

[0113] For each additional input potential, there can be an additional converter branch W, which is connected on the input side to the additional input potential and on the output side to the output terminal. Likewise, one or more additional input potentials can be formed between the lower (second) input potential, to which the second converter branch B is connected on the input side, and the middle input potential, with Figure 2.2 also showing an additional input potential as an example. For this lower additional input potential, there is also an additional converter branch W, which is connected on the input side to this additional input potential and on the output side to the output terminal. Likewise, the additional converter branches W are designed as voltage-bidirectional, current-bidirectional switches 4.

[0114] The number of additional input potentials above and below the average input potential does not necessarily have to be identical, but this is advantageous for reasons of symmetry. Depending on how many additional input potentials are provided, abstractly speaking, n potentials can be connected to the output terminal. In this way, a quasi-n-level 2-hybrid MMC topology (abbreviated QnL-2-hybrid MMC) can be provided. In addition to this arrangement, any new or known multilevel converter arrangement from the literature can be used and supplemented with a multilevel branch C.

[0115] The following describes an advantageous control of the hybrid multilevel converter according to Figure 2.1. The method is also suitable for a hybrid multilevel converter according to Figure 2.2, although additional control steps must be implemented for the other converter branches. As already mentioned, the output alternating current is synthesized by high-frequency switching of the converter branches. Figure 2.3 shows the curve of the output voltage and the output current over a period of the output current To. The output voltage is generated by switching the converter branches with a pulse modulation with a period TPWM. Figure 2.4 shows a time-stretched section from Figure 2.3 over a period of the output voltage modulation TPWM. The corresponding switching points for generating the pulse modulation are generated, for example, by predetermined carrier signals Carrier 1, Carrier 2, which, for example,can be designed as triangular signals or sawtooth signals (upper diagram in Figure 2.4). If at least one of the carrier signals reaches a certain threshold, this triggers a transition T from one of the static states A, B, or C. The transitions between the states are explained below using Figure 2.6.

[0116] Figure 2.4 illustrates the transition through the states A->T ->C and additionally the states C->T->A for a positive output current. The lower labels in Figure 2.4 indicate that the first transition T is shown with an additional time extension in Figure 2.7, and the second transition T is shown with a time extension in Figure 2.8.

[0117] The middle graph of Figure 2.4 shows the output voltage vo and the output current io. The output current varies practically unchanged over the illustrated, time-stretched range. The lower graph shows the branch voltage Vbc and the branch current ibc of the third converter branch C. It can be seen that in the steady-state state A, the branch voltage Vbc is modulated with a high-frequency pulse modulation with a period THF. This period THF is used to implement high-frequency branch current control in the steady-state state (SZ).

[0118] The quantities used in Figures 2.3 and 2.4 are summarized below. Period of the output current: T o (Example: T o = 20ms)

[0119] Period of the output voltage modulation: T PWM (Example: T PWM = 1ms) Period of the high-frequency branch current control in the static states: T HF(Example: T HF = 40|is)

[0120] Time between two switchings of the MMC modules: T d (Example: T d = l|is)

[0121] Figure 2.5 illustrates the hierarchy of the various controllers within the control system, i.e., a control structure of a cascaded control system. The highest hierarchy level contains an output current controller 11. Output current controller 11 regulates the output current by modulating the output voltage with a modulation period TPWM. In this case, one of the converter branches is always operated as the current-transmitting converter branch. The current-transmitting converter branch provides the converter's output current, i.e., the output current flows through this converter branch.

[0122] The output current control 11 is subordinate to an energy control 12 of the third converter branch, i.e., the multilevel converter branch. The energy control 12 calculates the setpoint currents to compensate for energy fluctuations of the module capacitors in the individual modules.

[0123] A branch current control 13 is subordinate to the energy control 12. The branch current control 13 is implemented by a state machine, as explained below using Figure 2.6. The branch current control 13 converts the setpoints specified by the output current control 11 and the energy control 12 into control signals for the switches of the first and second converter branches and for controlling the third converter branch.

[0124] The branch current control 13 can also be subordinate to a branch balancing function 14, by means of which the switching signals for individual modules of the multilevel converter branch are generated in the third converter branch, i.e. a selection is made as to which individual modules of the third converter branch are switched on or off in order to create an even voltage distribution of the module voltages. In addition, a minimum waiting time Td between the switching operations of the multilevel converter branch can be guaranteed in the branch balancing. Figure 2.6 shows an advantageous embodiment of the branch current control 13 as a state machine. There are three static states A, B, C. These static states correspond to states in which the respective first converter branch A, second converter branch B or third converter branch C functions as the current-transmitting converter branch. A switch can be made between these static states as shown in Figure 2.6 can be switched back and forth. When switching between different static states, a respective transition state is briefly assumed, whereby four transition states are distinguished: A—>C, C—>A, B—>C, C—>B.

[0125] If the state machine is in the static states “A” or “B”, the branch current control is characterized by the following features:

[0126] • In the multilevel converter branch, as well as in the closed switch,

[0127] Currents that serve to compensate for energy fluctuations of the module capacitors in the multilevel converter branches are regulated by a modulation method of the multilevel converter branch (modulation period of the modulation method of the multilevel converter branch in the static states is called T HF through which the output current does not flow.

[0128] If the state machine is in transitions “T”, the branch current control and the required activities in the transition sections can be described using the following table. For the transitions from C->A or C->B (for both output current directions), "Transition Start" must first be carried out in order to guarantee the small voltage levels at the output before a switch ^ or S ß may be closed.

[0129] Only then does the current commutation in the desired direction take place by setting the voltage of the multilevel converter branch v bc instead of.

[0130] In the transitions A->C or B->C, the current commutation in the desired direction is first carried out by setting the voltage of the multilevel converter branch v bc instead of.

[0131] In the transitions A->C or B->C, where previously the transistor of S Aor S B conducted (current in S A or S ß is positive), “Transition End” includes the following three steps:

[0132] 1 ) The current in S A or S B is reduced until it becomes negative.

[0133] 2) Then the diode of S A or S B , and the switching signal from S A or S B can be set to 0.

[0134] 3) The current in S A or S B back to 0A regulated by setting the voltage of the multilevel converter branch v bc , which is a maximum of a module capacitor voltage v mod from the next static voltage level (at transition A->C or at transition C->B). Thus, even the small Voltage levels at the output when switching off S A or S B guaranteed.

[0135] In the transitions A->C or B->C, where previously the diode of S A or S ß conducted (current in S A or S ß is negative), "Transition End" only includes steps 2) and 3) of the three steps mentioned above. These are initiated when the current in S A or S B is close to 0A.

[0136] Definitions: k Number of full-bridge modules (individual modules) in the multilevel converter branch g Number of full-bridge modules (individual modules) to be switched on to generate half the input voltage. Calculation example: g = round II with v mod as the average voltage of all module capacitor voltages \ v mod / voltages of the full bridge modules (individual modules).

[0137] Figure 2.7 shows the voltage and current waveforms in the three converter branches and at the output terminal during transition A^C. The selected time scale is greatly enlarged and includes the left-hand transition region T marked in Figure 2.4. The lower diagram in Figure 2.7 shows the switching signal for turning off the switch in the first converter branch A. It can be seen how, by controlling the third converter branch, in which the current is slowly increased, the current in the first converter branch A is slowly reduced in a corresponding manner. As soon as the current in the first converter branch A becomes slightly negative, the freewheeling diode of the switch in the first converter branch becomes conductive. The switching signal for turning off the switch is then output.The current in the first converter branch A is then regulated back to zero by adjusting the voltage in the third converter branch so that the voltage differs from the next static voltage level by a maximum of one module capacitor voltage. This completes the transition A^C.

[0138] Figure 2.8 shows a transition C^A corresponding to the time segment marked in the right transition T in Figure 2.4. Thus, a "Transition Start" is executed according to the pattern described above.

[0139] It was discovered that the same principle can also be applied to a novel QnL-2 hybrid MMC topology, as shown in Fig. 2.2. Here, the MMC branch has the primary task of shaping the transitions between the different output voltage levels. This is possible because, in the steady state, only one of the switches 3 is switched on at a time, connecting the output to one of the intermediate circuit voltage levels. Only the zero output voltage state would be realized by the MMC branch—if it is not bypassed in parallel by an additional switch. The advantage over state-of-the-art topologies is that a smaller number of MMC branches are used. MMC branches are more expensive and require more control and regulation effort (including sensors) than simple power electronic switches.It is therefore expected that the size and cost of the novel converters will be significantly lower than those of conventional MMC and QnL (hybrid) MMC converters from previous work, so that they can represent an attractive alternative for variable-frequency applications with inductive load, especially for large drives.

[0140] The novel QnL-2Hybrid MMC topologies with unidirectional or bidirectional switch arrangements (see Fig. 2.1 , 2.2) also promise many advantages over their conventional counterparts (n-level converters):

[0141] • Because switching occurs at zero or very low current, more inductance can be allowed in the commutation paths of the switch configuration. This simplifies the converter design compared to conventional technology, which requires a large number of commutation paths to be low-inductance, complicating the design of the DC link busbar.

[0142] • While the individual MMC branch can be designed redundantly, a fail-safe design and operation of the other switches is also facilitated, since the series connection of several power semiconductors in one switch is simplified by the almost voltage-free / current-free switching.

[0143] • The arrangement of the unidirectional or bidirectional switches 3 can differ from the example in Fig. 2.2. The important thing is that the switches 3 are capable of connecting the DC taps to the converter output. The output voltage is transitioned from one steady state to the next by the MMC branch while all switches 3 are off. The new switching state can then be switched on by closing the corresponding switch, and the MMC branch is controlled so that the output current commutates to that switch.

[0144] • In principle, the MMC branch of the topologies shown in Fig. 2.1 and Fig. 2.2 could be an addition to an existing converter, active only during transitions, reducing the effects of steep voltage transients and reducing switching losses to almost zero. • In Fig. 2.1, 2.2, the additional effort for designing the transitions is moderate, while the potential savings in switching losses of the high-voltage switches is quite high. In conventional medium-voltage converters with HV IGBTs, the switching losses amount to approximately 30% - 50% of the total losses, even at low switching frequencies below 1 kHz. The MMC branch can be implemented with low-voltage semiconductors; even the use of MOSFETs (Si or SiC) would be conceivable, so their potential switching losses are relatively low in comparison, and redundancy can be high thanks to a large number of MMC modules connected in series.

[0145] 4. Literatur

[0146]

[0001] J. Kucka and A. Mertens, “Improved Current Control of a Quasi-Two-Level PWM-Operated Modular Multilevel Converter,” in 2018 IEEE Energy Conversion Congress and Exposition (ECCE), Portland, OR, Sep. 2018 - Sep. 2018, pp. 2948-2955 .

[0147] [2] X. Wei, S. Lu,

[0148] [3] M. Bruckmann, R. Sommer, M. Fasching and J. Sigg, “Series Connection of High Voltage IGBT Modules” in Conference records of the 1998 IEEE Industry Applications Conference.

[0149] [4] D. Schröder, Power Electronic Circuits: Function, Design and Application, 3rd ed. Berlin: Springer, 2012.

[0150] [5] J. Kucka, “Quasi-two-level PWM operation of modular multilevel converters: implementation, analysis, and application to medium-voltage drives”. Dissertation, Leibniz University Hannover.

[0151] [6] S. Brüske, Performance Comparison and Neutral Point Balancing for 3-Level Inverter Topologies for Use in Electric Vehicles. Dissertation. Kiel: Christian-Albrechts-University of Kiel, 2016.

[0152] [7] R. Sommer et al., “New medium voltage drive systems using three-level neutral point clamped inverter With high voltage IGBT” in Conference Record of the 1999 IEEE Industry' Applications Conference. Thirty-Forth IAS Annual Meeting (Cat. NO.99CH363 70), Phoenix, AZ, USA, Oct. 1999, pp. 1513-1519. [8] A. Lesnicar and R. Marquardt, “An innovative modular multilevel converter topology suitable for a wide power range,” in 2003 IEEE Bologna Power Tech Conference Proceedings, Bologna, Italy, Jun. 2003, pp. 272-277.

[0153] [9] K. Sharifabadi, L. Hamefors, H. P. Nee, S. Norrga und R. Teodorescu, Design, control, and application of modular multilevel convertersfor HVDC transmission systems. Chichester West Sussex United Kingdom: IEEE Press Wiley, 2016.

[0154]

[0010] D. Karwatzki, Analyse und Regelung einer Klasse von modularen Multilevelum- richter-Topologien. Dissertation. Hannover: Gottfried Wilhelm Leibniz Universität Hannover, 2017.

[0155]

[0011] A. J. Korn, M. Winkelnkemper und P. Steimer, „Low output frequency operation of the Modular Multi-Level Converter“ in 2010 IEEE Energy Conversion Congress and Exposition, Atlanta, GA, Sep. 2010 - Sep. 2010, S. 3993-3997.

[0156]

[0012] A. Antonopoulos et al., „Modular Multilevel Converter AC Motor Drives With Constant Torque From Zero to Nominal Speed“, IEEE Trans, on Ind Applicat., Jg. 50, Nr.

[0157] 3, S. 1982-1 9931 , 2014.

[0158]

[0013] A. Mertens und J. Kucka, „Quasi Two-Level. PWM Operation of an MMC Phase Leg With Reduced Module Capacitance“, IEEE Trans. Power Electron., Jg. 31 , Nr.

[0159] 10, S. 6765-6769, 2016.

[0160]

[0014] J. Kucka und A. Mertens, „Common-Mode Voltage Injection Techniques for Quasi Two-Level PWM-Operated Modular Multilevel Converters“ in 2018 International Power Electronics Conference (IPEC-Niigata 2018 -ECCE Asia), Niigata, Mai. 2018 - Mai. 2018, S. 1904-1911.

[0161]

[0015] J. Kucka und A. Mertens, „Control for Quasi Two-Level PWM Operation of Modular Multilevel Converter“ in 2016 IEEE 25th International Symposium on Industrial Electronics (ISIE), Santa Clara, CA, USA, Jun. 2016 - Jun. 2016, S. 448-453.

[0162]

[0016] M. F. Aiello und W. Barie, „High resonate pole frequency M2LC converter for medium voltage drive and energy sector applications“ in 2015 IEEE Energy Conversion Congress and Exposition (ECCE): 20-24 Sept. 2015, Montreal, Quebec, Canada, Montreal, QC, Canada, 2015, S. 3515-3521.

[0163]

[0017] M. F. Aiello, K. S. Berton und W. G. Barie, „M2LC System and Method for Controlling Same,“ US 9372 493 B2.

[0018] F. Bertoldi, M. Pathmanathan, R. S. Kanchan, K. Spiliotis und J. Driesen, „Quasi- Two-Level Converter for overvoltage mitigation in medium voltage drives“ in 2018 International Power Electronics Conference (IPEC-Niigata 2018 -ECCE Asia), Niigata, Mai. 2018 - Mai. 2018, S. 488-494.

[0019] F. Bertoldi, M. Pathmanathan und R. ,S. Kanchan, „Quasi-two-level converter operation strategy for overvoltage mitigation in long cable applications“ in 2019 IEEE International Electric Machines & Drives Conference (IEMDC), San Diego, CA, USA, Mai. 2019 - Mai. 2019, S. 1621-1627.

[0164]

[0020] M. S. Diab, G. P. Adam, B. W. Williams, A. M. Massoud und S. Ahmed, „Quasi two-level PWM operation of a nine-arm modular multilevel converter for six-phase medium-voltage motor drives“ in 2018 IEEE Applied Power Electronics Conference and Exposition (APEC), San Antonio, TX, USA, Mrz. 2018 - Mrz. 2018, S. 1641 - 1648.

[0165]

[0021] M. Lorenz, J. Kucka und A. Mertens, „A Modular Multilevel Converter with a Clamping Switch for Quasi-Three-Level Operation“ in 2021 22nd IEEE International Conference on Industrial Technology (ICIT), Valencia, Spain, 10.03.2021 - 12.03.2021 , S. 260-265, doi: 10.1109 / ICIT46573.2021 .9453540.

Claims

Patent claims:

1. A method for operating a hybrid multilevel converter that delivers electrical output power comprising an output voltage and an output current, wherein the output power is transmitted by means of a circuit arrangement that a) has at least a first, a second, and a third converter branch, wherein b) the first converter branch is connected to a first input potential, c) the second converter branch is connected to a second input potential that differs from the first input potential, and d) the third converter branch is connected to a third input potential that lies between the first and the second input potential, wherein e) the first, the second, and the third converter branch are connected to one another at an output terminal of the circuit arrangement to which the output power is transmitted,wherein f) the first and the second converter branch each consist of a switch and at least one or no inductance, wherein g) the third converter branch is designed as a multilevel converter branch with several series-connected individual modules and at least one or no inductance, characterized in that a control of the operation of the converter is carried out with the following features: h) the current intensity of the output current is controlled by a modulation method of the output voltage, wherein only one of the converter branches is operated as a current-transmitting converter branch, by which, the output current flows and the output power is delivered accordingly, and the first, second and third converter branches alternately function as current-transmitting converter branches, i) wherein, during a transition in which the third converter branch receives or loses its assignment as a current-transmitting converter branch, a transition of the output current is carried out from the converter branch previously functioning as a current-transmitting converter branch to the converter branch subsequently functioning as a current-transmitting converter branch by controlling the individual modules of the third converter branch.

2. Method according to claim 1, characterized in that during a transition in which the third converter branch receives its assignment as a current-transmitting converter branch, by controlling the individual modules of the third converter branch, a positive current flowing at the beginning of the transition through the first or second converter branch previously acting as a current-transmitting converter branch or a negative current flowing at the beginning of the transition through the first or second converter branch previously acting as a current-transmitting converter branch is controlled to a small negative current, so that the freewheeling diode is brought into the conducting state, and then the switch of the first or second converter branch previously acting as a current-transmitting converter branch is switched off.

3. Method according to claim 2, characterized in that after switching off the switch of the first or second converter branch previously acting as the current-transmitting converter branch, by controlling the individual modules of the third converter branch, the current flowing through the first or second converter branch previously acting as the current-transmitting converter branch is gradually increased with a time gradient until it crosses zero.

4. Method according to one of the preceding claims, characterized in that during a transition in which the third converter branch receives its assignment as a current-transmitting converter branch, by controlling the individual modules of the third converter branch, a A positive current flowing through the first or second converter branch previously functioning as a current-transmitting converter branch, or a negative current flowing at the beginning of the transition through the first or second converter branch previously functioning as a current-transmitting converter branch, is controlled to a small positive current. Method according to one of the preceding claims, characterized in that, before the switch of the first or second converter branch previously functioning as a current-transmitting converter branch is switched off, the current flowing through the third converter branch is gradually increased or decreased with a time gradient by controlling the individual modules of the third converter branch.Method according to one of the preceding claims, characterized in that, during a transition in which the third converter branch loses its assignment as a current-transmitting converter branch, the voltage applied to the switch of the first or second converter branch subsequently functioning as a current-transmitting converter branch is controlled to a reduced value by controlling the individual modules of the third converter branch, and then the switch of the first or second converter branch subsequently functioning as a current-transmitting converter branch is switched on. Method according to claim 6, characterized in that, after the switch of the first or second converter branch subsequently functioning as a current-transmitting converter branch is switched on, the current flowing through the third converter branch is gradually increased or decreased with a time gradient by controlling the individual modules of the third converter branch.Method according to one of the preceding claims, characterized in that the method has an output current control which controls the current intensity of the output current by means of a modulation method of the output voltage, wherein the output current control is subordinate to an energy control of the third converter branch, by means of which temporal energy fluctuations in the. Module capacitors of the individual modules of the third converter branch are minimized. Method according to claim 8, characterized in that the energy control is subordinate to a branch current control, by which current setpoints specified by the output current control and / or the energy control are converted into currents flowing through the third converter branch by high-frequency clocked switching on and off of individual modules of the third converter branch. Method according to claim 9, characterized in that the branch current control is operated as a state machine having at least three static operating states and at least four transition states, via which it is possible to switch between different static operating states.Method according to claim 9 or 10, characterized in that the output current control is operated with a clocking with a first period duration, and the branch current control is operated with a clocking with a second period duration, wherein the second period duration is substantially shorter than the first period duration. A computer program with program code means configured to carry out a method according to one of claims 1 to 11 when the computer program is executed on a computer.A hybrid multilevel converter which delivers an electrical output comprising an output voltage and an output current, wherein the output power is transmitted by means of a circuit arrangement which a) has at least a first, a second and a third converter branch, wherein b) the first converter branch is connected to a first input potential, c) the second converter branch is connected to a second input potential which is different from the first input potential, and. d) the third converter branch is connected to a third input potential that lies between the first and the second input potential, wherein e) the first, the second, and the third converter branches are connected to one another at an output terminal of the circuit arrangement to which the output power is transmitted, wherein f) the first and the second converter branches each consist of a switch and at least one or no inductance, wherein g) the third converter branch is designed as a multilevel converter branch with several individual modules connected in series and at least one or no inductance, h) the circuit arrangement has at least one control device that is configured to control the power semiconductors of the converter branches, wherein the control device is configured to carry out a method according to one of claims 1 to 11. Hybrid multilevel converter,which delivers an electrical output power comprising an output voltage and an output current, wherein the output power is transmitted by means of a circuit arrangement, in particular a hybrid multilevel converter according to claim 13, which a) has at least a first, a second and a third converter branch, wherein b) the first converter branch is connected to a first input potential, c) the second converter branch is connected to a second input potential that differs from the first input potential, and d) the third converter branch is connected to a third input potential that lies between the first and the second input potential, wherein e) the first, the second and the third converter branch are connected to one another at an output terminal of the circuit arrangement, to which the output power is transmitted,wherein f) the first and the second converter branch each consist of a switch and at least one or no inductance, wherein, g) the third converter branch is designed as a multilevel converter branch with several individual modules connected in series and at least one or no inductance, h) characterized in that the circuit arrangement has one or more further converter branches, wherein a respective further converter branch is connected on one side to a further input potential which differs from the first, second and third input potential as well as from all other further input potentials, and on the other side directly or indirectly via further switches to the output terminal of the circuit arrangement to which the output power is transmitted.

15. Hybrid multilevel converter according to claim 14, characterized in that a respective further converter branch consists of a switch and at least one or no inductance.

16. Hybrid multilevel converter according to claim 15, characterized in that the bidirectional switch has an arrangement of power semiconductor switching elements which can symmetrically absorb blocking voltage and can symmetrically carry current.

17. Hybrid multilevel converter according to one of claims 13 to 16, characterized in that the third input potential is provided by a circuit arrangement which has a first capacitor Ci , which is connected with a first terminal to the first input potential and a second terminal to the third input potential, and a second capacitor C2, which is connected with a first terminal to the second input potential and a second terminal to the third input potential.

18. Hybrid multilevel converter according to claim 17, characterized in that the respective further input potential to which a respective further converter branch is connected is provided by a voltage divider circuit arrangement formed from series-connected capacitors.

Images