DE202017007011U1 - n-Port-m-phase bridge converter Active - Google Patents

Abstract

wobei U_k die DC-Spannung an einem frei wählbaren Port k mit k ⊆ 1, 2, 3,..., n und k≠i ist, L_sk eine Leckinduktivität der Transformatorwicklung des Ports k referenziert auf Port i=1 ist und N_1k ein Transformatorwindungsverhältnis zwischen dem Port i=1 und dem Port k ist und φ_k1 der vorgegebene Relativ-Last-Winkel zwischen dem Port k mit k≠i und dem Port 1 ist.

Device (100) for operating an n-port m-phase active bridge converter (10), where k, i ⊆ 1, 2, .., n are a number of a freely selectable port (1, ..., n) where m ≥ 1 and n ≥ 2, where each port (1, ..., n) can be operated as an input or output port, each of the ports (1, ... n) being m phase branches (16) comprising a plurality of active switches (17, 19), wherein a phase shift (20) takes place between the phase branches (16) of each port, wherein the ports can convert a DC voltage (U) into an AC voltage (v) and vice versa, wherein the ports (1) are connected via an m-phase transformer (25) or via separate m transformers (25) respectively connected to each of the m phase legs (16) of each of the n ports (1, ..., n) , are connected to transfer power between the ports (1, ..., n), the switches (17, 19) of the phase branches (16) being operated by a control unit (30) at least one initial DC Offset (32) in a magnetization trom (I _M ), the control unit being designed to:
Dividing (110) a switching period T into m intervals (38) if m is an even number, and in 2 * m intervals (38) if m is an odd number, the intervals (38) being interval limits (40) Switching the active switches (17, 19) define
- Predetermining (120) a power to be transmitted of the n-port m-phase active bridge converter (10) corresponding to a predetermined relative load angle φ _i1 = φ _i -φ ₁ between a load angle φ _{i of} an arbitrary Ports i and an absolute load angle φ _{1 of} a reference port 1, and
Switching (130) of the active switches (17, 19) of the n ports (1, ..., n) such that each phase branch (16) of a freely selectable port i is offset by the absolute load angle φ _i relative to shifts the interval boundaries (40) of any of the intervals (38) in which the AC voltage (v) of the m phase branches (16) changes sign, where φ _{i is} calculated as $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

where U _{k is} the DC voltage at an arbitrary port k with k ⊆ 1, 2, 3, ..., n and k ≠ i, L _{sk is} a leakage inductance of the transformer winding of the port k referenced to port i = 1 and N _{1k is} a transformer _{turn ratio} between the port i = 1 and the port k, and φ _{k1 is} the predetermined relative load angle between the port k with k ≠ i and the port 1.

Description

Field of the invention

The invention relates to a device for operating an n-port m-phase active bridge converter for reducing at least an initial DC offset of a magnetizing current and for controlling the DC offset during a transient condition of a load change and a corresponding n- A port-to-phase active-bridge converter and a system comprising said n-port-m-phase active bridge converter and for carrying out a method of operating an n-port m-phase active bridge Converter is provided to reduce at least an initial DC offset of a magnetizing current and to control the DC offset during a transient condition of a load change, and a corresponding n-port m-phase active bridge converter.

Background of the invention

A DC-DC converter, also known as a power converter, refers to an electrical circuit that converts a DC or DC voltage supplied to the input side into a DC or DC voltage having a higher, lower or inverted voltage level. DC-DC converters can be found, for example, in switched-mode power supplies of PC power supplies, notebooks, mobile telephones, small motors and hi-fi devices. Their advantages over linear power supplies are their higher efficiency and lower heat generation. In a linear voltage regulator or a series resistance element, in contrast, the superfluous voltage is simply "burned".

DC-DC converters are also available as fully encapsulated converter modules, sometimes for direct insertion into printed circuit boards. Depending on the model, the output voltage (secondary voltage) may be lower than, equal to or greater than the input voltage (primary voltage). The best-known modules are those that transform a micro-voltage into a galvanically isolated micro-voltage. The encapsulated DC-DC converters are available, for example, for isolation voltages in the range of 1.5 kV to over 3 kV, and are used to deliver power to small consumers in DC networks, such as 24V in plant or 48V in telecommunications or in the electronic modules, such as 5 V for digital circuits or ± 15 V for the operation of operational amplifiers. DC-DC converters are also found in high power applications such as automotive and traction. In automotive applications, for example, they serve to charge batteries or supply power from the batteries or fuel cells to the DC link of the inverter. DC-DC converters are classified according to various criteria and subdivided into different topologies (types of a structure of a branched network on current paths). Unlike unidirectional converters, with respect to bidirectional or multidirectional multiport DC-DC converters, it is immaterial which port (s) are defined as input and which port (s) are defined as output. A bidirectional flow of energy allows power to flow from the defined input (primary side) towards the output (secondary side) and vice versa. In the case of multiport Active Bridge converters, the ports are not labeled primary or secondary, but are numbered instead, such as port 1 , Port 2 , Port 3 Etc.

In DC-DC converters, which are based on the principle of operation of a two-port active bridge converter, the so-called Dual Active Bridge (DAB) topology, the DC input voltage is converted in an input converter into an AC voltage, which is then fed to a transformer becomes. The output of the transformer is connected to an output transducer, which converts the AC voltage back to a DC output voltage for a load. These DC-DC converters may be implemented in single-phase or multi-phase configurations. Such DAB DC-DC converter topologies, such as those in the U.S. Patent No. 5,027,264 are high power transducer topologies that enable bidirectional power flow and isolation across the transformer and high voltage operation. This type of converter is particularly well suited for use in high power DC networks, eg medium voltage networks.

In the case of a DC-DC converter with a dual active bridge (DAB) topology, the transmitted power is determined by varying the load angle between the voltage on the primary side and the voltage on the secondary side. If the converter starts or the transmitted power needs to be changed abruptly during normal operation, unwanted DC offsets may occur in the magnetizing currents of the transformer. When DC offset currents occur in the magnetizing currents, the Transformer core, which leads to high currents through the primary-side and the secondary-side semiconductor devices and the transformer windings. The high currents cause overheating, which can lead to failures of the devices. Assuming a distribution factor of the leakage inductance and the winding resistance between primary and secondary windings of the transformer of k = 0.5, these DC offset currents will decay with a time constant of (2L _m + L _s ) / R _w , where L _m the magnetizing inductance is, Ls is the total leakage inductance relative to the primary side, R _w is the total winding leakage resistance with respect to the primary side and k is the ratio of the primary leakage inductance to the total leakage inductance Ls. A low winding resistance R _w means a higher efficiency and is therefore desirable. Because the magnetizing inductance L _m is normally large, the time constant (2L _m + L _s ) / R _{w has} a large value. Therefore, the DC offsets in the magnetizing currents only decay slowly. Therefore, it is desirable to provide means for a DC to DC converter to avoid the DC offsets in the magnetizing current of the transformer and to control the DC offset under transient conditions, for example load changes.

Overview of the invention

The invention has for its object to provide a device for a multidirectional multiport DC-DC converter to avoid at least DC offsets in the magnetizing current of the transformer and to control the DC offset under transient conditions.

• - Aufteilen einer Schaltperiode T in m Intervalle, wenn m eine gerade Zahl ist, und in 2*m Intervalle, wenn m eine ungerade Zahl ist, wobei die Intervalle Intervallgrenzen zum Schalten der aktiven Schalter definieren,
• - Vorbestimmen einer zu übertragenden Leistung des n-Port-m-Phasen-Active-Bridge-Wandlers entsprechend einem vorgegebenen Relativ-Last-Winkel φ_i1=φ_i-φ₁ zwischen einem Lastwinkel φ_i eines frei wählbaren Ports i und einem Absolut-Last-Winkel φ₁ eines Referenzports 1, und
• - Schalten (130) der aktiven Schalter (17, 19) der n Ports (1, ..., n) derart, dass sich jeder Phasenzweig (16) eines frei wählbaren Ports i um den Absolut-Last-Winkel φ_i relativ zu den Intervallgrenzen (40) eines beliebigen der Intervalle (38), in dem die AC-Spannung (v) der m Phasenzweige (16) ihr Vorzeichen ändert, verschiebt, wobei φ_i berechnet wird als

$${\phi }_i=-\frac{{\displaystyle \sum _{k=\mathrm{1,}k\ne i}^n\frac{{\phi }_{k1}{N}_{1\text{k}}{U}_k}{L_{sk}}}}{{\displaystyle \sum _{k=1}^n\frac{N_{1\text{k}}{U}_k}{L_{sk}}}}$$

wobei U_k die DC-Spannung an einem frei wählbaren Port k mit k ⊆ 1, 2, 3,..., n und k#i ist, L_sk eine Leckinduktivität der Transformatorwicklung des Ports k referenziert auf Port i=1 ist und N_1k ein Transformatorwindungsverhältnis zwischen dem Port i=1 und dem Port k ist und φ_k1 der vorgegebene Relativ-Last-Winkel zwischen dem Port k mit k#i und dem Port 1 ist. Im Folgenden wird das Verfahren als unmittelbare Flusssteuerung (Instantaneous Flux Control - IFC) bezeichnet.The object is achieved by a device for operating an n-port-m-phase active bridge converter, where k, i ⊆ 1, 2, .., n are a number of a freely selectable port with m ≥ 1 and n ≥ 2, wherein each port may be operated as an input or output port, each of the ports having m phase legs with multiple active switches, with a phase shift between the phase legs of each port, the ports forming a DC voltage into an AC voltage and vice versa, wherein the ports are connected via an m-phase transformer or via separate m transformers respectively connected to each of the m phase legs of each of the n ports, the switches of the phase legs being driven by a control unit at least reduce an initial DC offset in a magnetizing current, the control unit being provided for:

• - Splitting a switching period T in m intervals, if m is an even number, and in 2 * m intervals, if m is an odd number, the intervals defining interval limits for switching the active switches,
• Predetermining a power to be transmitted of the n-port m-phase active bridge converter in accordance with a predetermined relative load angle φ _i1 = φ _i -φ ₁ between a load angle φ _i a freely selectable port i and an absolute load angle φ _{1 of} a reference port 1 , and
• - switching ( 130 ) the active switch ( 17 . 19 ) of the n ports ( 1 . .. ., n ) such that each phase branch ( 16 ) of a freely selectable port i by the absolute load angle φ _i relative to the interval limits ( 40 ) of any of the intervals ( 38 ), in which the AC voltage ( v ) of the m phase branches ( 16 ) changes its sign, shifts, where φ _i is calculated as

$${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{k1}{N}_{1\text{k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{1\text{k}}{U}_k}{L_{sk}}}}$$

where U _{k is} the DC voltage at an arbitrary port k with k ⊆ 1, 2, 3, ..., n and k # i, L _{sk is} a leakage inductance of the transformer winding of port k referenced to port i = 1 and N _{1k is} a transformer _{turn ratio} between the port i = 1 and the port k, and φ _{k1 is} the predetermined relative load angle between the port k with k # i and the port 1 is. In the following, the method is referred to as Instantaneous Flux Control (IFC).

On n Port m-phase active bridge converter has a number n of ports and a number m of phases, where n, m ⊆ N. The converter is well known in its two-port m-phase form, which is also referred to as a dual-active bridge converter, DAB converter for short. For an n-port active bridge converter where n> 2, the terms "multiport active bridge DC-DC converter", "MAB converter" for short, and "multiport DAB converter" are used interchangeably. When transmitting power in a MAB topology, at least one of the ports converts a DC input voltage to an AC input voltage as an input port (s) and converts at least one of the other ports an AC output voltage a DC output voltage as the output port (s). Generally speaking, however, each port may convert a DC voltage to an AC voltage, and each port may have a bidirectional power flow. Under standby conditions, when the transmitted power between ports is zero, all ports convert a DC voltage to an AC voltage that produces only reactive power in the transformer.

In DAB converters, the primary side typically refers to the input port of the DC-DC converter. Accordingly, the secondary side refers to the other side of the transformer connected to the output port. When the primary and secondary sides are isolated from each other, their respective ports are connected through a transformer. Non-isolated variants can be realized using inductors or autotransformers between primary and secondary ports. If the DAB converter is designed to be bidirectional, the power may be transmitted either from the primary side to the secondary side or vice versa. The DC-DC converter according to the present invention may be configured as a m-phase multiphase DAB converter that may be configured as a single-phase, three-phase, or five-phase, etc., DAB or multiport active bridge (MAB) converter.

The number of phases m determines the minimum number of voltage bridges in a port, which are the so-called phase branches. A three-phase DAB has at least three phase branches in one port, and in the case of more than three phase branches, some phase branches are connected in parallel to provide three effective phases. Each phase branch has active switches in a two-stage or multi-stage configuration. By switching the active switches (switching operation with a switching pattern), a DC input voltage can be converted into an AC output voltage to the transformer. Conversely, an AC output voltage into a DC input voltage. Normally, the power semiconductor switches are fully on or off. Therefore, in the case of two-stage converter phase branches, AC voltages approximately in the form of a square wave (square-wave voltage) are produced at the transformer. In the case of multistage transducers, multi-level waveforms are generated. Consequently, the voltage across the transformer windings is stepped. Optionally, due to so-called snubbers, the edges of the square-wave voltage are not infinitely steep, that is, the shape deviates from that of a square-wave voltage (step shape on the transformer windings). Snubbers are used to ensure dynamic voltage compensation during switching operations (snubber network). The term "snubber net" refers to an electrical circuit having snubber elements which, in the event of an abrupt interruption of the current flow, are intended to neutralize, for example, interfering high frequencies or voltage spikes which usually occur when inductive loads are switched , Snubber elements limit the rate of voltage increase or the rate of current increase on semiconductors. For example, capacitive snubbers cause zero crossing switching during turn off of the power semiconductors, thereby reducing switching losses.

Depending on the configuration, two or more active switches can be used per phase branch. Suitable switches for the phase legs are active semiconductors (power semiconductors), such as turn-off thyristors, transistors, MOS FETs, Insulated Gate Bipolar Transistors (IGBTs), or integrated gate-commutated thyristors (ICGTs) with intelligent gate drivers. Alternatively, one skilled in the art may also use other suitable active switches.

The term "transformer" as used herein refers to the magnetic circuit - usually a ferrite or iron core - with the associated windings of the primary and secondary phase branches around the magnetic core. When each of the phase legs has a separate transformer, only the conductors of a primary phase leg and the associated other secondary phase leg are wound around the transformer associated with that phase leg. The magnetic cores of the transformers of the individual phase branches are then physically separated. By contrast, the term "multiphase transformer" or "m-phase transformer" refers to a transformer having a shared magnetic core for all of the phase legs, the windings of the phase branches of a first, second, and nth ports in different areas of the magnetic core are arranged. The function of a DC-DC converter is to cause a systematic voltage drop across the AC voltages on the transformer through the leakage inductance of the transformer and thus control the power flow. Active-connected phase arms make it possible to independently control the load angle between the primary and secondary AC voltages present on the transformer and thus to systematically control the power flow.

Once the transformer core is saturated due to large DC offsets in the magnetizing currents, high currents flow through the transformer windings (magnetizing currents) and the Devices in the actively connected bridges or phase branches. This can cause overheating and lead to failure of the device (DC-DC converter) and should therefore be avoided.

The control unit need not be part of the n-port m-phase active bridge converter, but may also be an external component used to control the active switches with the n-port m-phase active bridge converter connected is. This is advantageous if the control must be flexibly adapted to the given application. Such a control unit, which is not an integral part of an n-port m-phase active bridge converter, may be incorporated in an interconnect network with, for example, a separately created microcontroller, a digital signal processor (DSP), or programmable logic devices, e.g. FPGA, be installed. These or similar devices may be connected to the switches via a power semiconductor gate driver, which is conventionally implemented as part of the power electronics converters. The control unit can also be integrated via printed conductors in the same circuit board of the semiconductor switches or integrated into external printed circuit boards and then connected via cables. The connection (e.g., cable) can be realized in several ways, e.g. via a printed circuit board, standard cable, coaxial cable, ribbon cable, RJ-45 cable, optical signals (eg optical waveguide) etc. As long as the DC-DC converter provides a connection port for a gate driver, the control unit can also be realized as a control frame / control platform / which is connected to the DC-DC converter.

The split into switching intervals is defined by the number of phases that the MAB converter has. When the number of phases for a MAB converter is m, it must be considered whether m is an even or odd number. If m is an odd number, the number of intervals is twice m because the turn-on and turn-off times of each phase must be taken into account. If the number of phases m is an even number, the number of intervals is reduced to m because the turn-on times of half a number of phase legs coincide with the turn-off times of the other half of phase legs. Therefore, when m is three, for example, the number of intervals is six. If m is two, the number of intervals remains two. If m is five, the number of intervals is ten. If m is four, the number of intervals is four.

The interval limits are defined by the division of the intervals. The number of interval limits is defined by the number of intervals plus one. If the number of intervals is equal to six in three phases, there are seven interval limits. The interval limits are set at a start time to and end at a time to + T, where T is a period length. The intervening time is equal divided by the remaining number of interval boundaries. If in the above example (three phases) the period T is divided into six parts, the interval limits can be found at the times to, to + T / 6, to + T / 3, to + T / 2, to + 2T / 3, to + 5T / 6, to + T. Here, the time interval to + T / 6 denotes the interval in the range from to to T / 6 with the times to and T / 6 as limits of this interval. The subsequent time interval t ₀ + T / 3 denotes the interval in the range of T / 6 to T / 3 With T / 6 and T / 3 as limits of this interval. The time interval t ₀ + T / 2 denotes the interval in the range of T / 3 to T / 2 With T / 3 and T / 2 as limits of this interval. The time interval t ₀ + 2T / 3 denotes the interval in the range of T / 2 to 2T / 3 With T / 2 and 2T / 3 as limits of this interval. The time interval t ₀ + 5T / 6 denotes the interval in the range of 2T / 3 to 5T / 6 With 2T / 3 and 5T / 6 as limits of this interval. The time interval to + T denotes the interval in the range of 5T / 6 to T With 5T / 6 and T as limits of this interval.

A predetermined transmitted power of the n-port m-phase active bridge converter is needed to power a load on at least one arbitrary port selected as the output port relative to at least one input port. To transfer this power, a relative load angle is established between the input and output ports (s). Therefore, a given relative load angle φ _i1 = φ _i -φ _{1 is} between the absolute load angle φ _{i of} a freely selectable port i and the absolute load angle φ _{1 of} a reference port 1 defined, where i ≠ 1. The reference port is a freely selectable port 1 because the DC-DC converter always have at least one port. The absolute load angle φ _{i of} a port i denotes the angle between the rising edge of the AC voltage of port i and the corresponding interval limit.

wobei U₁ , U₂ , U₃ , ..., U_n die DC-Zwischenkreis-Spannungen an Port 1, 2, 3, ... n sind und L_s2 , L_s3 , ..., L_sn die Leckinduktivität der Transformatorwicklung von Port 1, 2, 3, ..., n bezogen auf die Seite von Port 1 sind.Based on the relative load angle φ _i1 , it is possible to calculate φ _i for any given i. After determining φ _i , it is possible to determine φ ₁ with φ ₁ = φ _i1 + φ _i . For determining φ _i , a difference in the flux of force in the magnetizing inductance ΔΦ _m when performing steps of φ ₂₁ , φ ₃₁ , ..., φ _n1 = 0 to arbitrary angles φ ₂₁ , φ ₃₁ , ..., φ _n1 , calculated. For example, for a triport-active-bridge converter, ie n = 3, ΔΦ _{m is} calculated according to: $$\begin{array}{l}\Delta {\mathrm{\Phi }}_{\text{m}}\alpha \frac{L_{\text{s2}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{1\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}1}+\frac{L_{\text{s1}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{2\mathrm{<figure-callout\; id="2"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}2}\\ +\frac{L_{\text{s1}}{L}_{\text{s2}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{13}{\mathrm{<figure-callout\; id="3"\; label="U"\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">U</figure-callout>}}_{3\mathrm{<figure-callout\; id="3"\; label="\phi "\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">\phi </figure-callout>}3}\stackrel{!}{=}0\end{array}$$

in which U ₁ . U ₂ . U ₃ , ..., U _n the DC link voltages at port 1 . 2 . 3 . .. , n are and L _s2 . L _s3 , ..., L _sn the leakage inductance of the transformer winding of Port 1 . 2 . 3 . .. ., n relative to the side of port 1 are.

was umgeschrieben werden kann zu: $${\phi }_1=-\frac{\frac{{\phi }_{21}{N}_{12}{U}_2}{L_{\text{s2}}}+\frac{{\phi }_{31}{N}_{13}{U}_3}{L_{\text{s3}}}}{\frac{U_1}{L_{\text{s1}}}+\frac{N_{12}{U}_2}{L_{\text{s2}}}+\frac{N_{13}{U}_3}{L_{\text{s3}}}}$$

Therefore, the equivalent equation for φ ₁ resolution of a triport-active-bridge converter is: $$\begin{array}{l}\frac{L_{\text{s2}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{1\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}1}+\frac{L_{\text{s1}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{12}{U}_{2\left({\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_{21}+{\phi }_1\right)}\\ +\frac{L_{\text{s1}}{L}_{\text{s2}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{13}{U}_{3\left({\phi }_{31}+{\phi }_1\right)}=0\\ \iff {\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_1=-\frac{L_{\text{s1}}{L}_{\text{s3}}{N}_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{2\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}21}+L_{\text{s1}}{L}_{\text{s2}}{N}_{13}{\mathrm{<figure-callout\; id="3"\; label="U"\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">U</figure-callout>}}_{3\phi 31}}{L_{\text{s2}}{\mathrm{<figure-callout\; id="1"\; label="L\; s3\; U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">L\; </figure-callout>}}_{\text{s3}}{U}_{}{}_1+L_{\text{s1}}{L}_{\text{s3}}{N}_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_2+L_{\text{s1}}{L}_{\text{s2}}{N}_{13}{\mathrm{<figure-callout\; id="3"\; label="U"\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">U</figure-callout>}}_3}\end{array}$$

what can be rewritten to: $${\phi }_1=-\frac{\frac{{\phi }_{21}{N}_{12}{U}_2}{L_{\text{s2}}}+\frac{{\phi }_{31}{N}_{13}{U}_3}{L_{\text{s3}}}}{\frac{U_1}{L_{\text{s1}}}+\frac{N_{12}{U}_2}{L_{\text{s2}}}+\frac{N_{13}{U}_3}{L_{\text{s3}}}}$$

was ebenfalls umgeschrieben werden kann in die Form: $${\phi }_i=-\frac{{\displaystyle \sum _{k=\mathrm{1,}k\ne i}^n\frac{{\phi }_{\text{k1}}{N}_{1\text{k}}{U}_k}{L_{sk}}}}{{\displaystyle \sum _{k=1}^n\frac{N_{1\text{k}}{U}_k}{L_{sk}}}}$$

For an n-port Active Bridge converter, the generalized equation for the absolute load angle of Port i with i = 1, 2, ...., n can be derived from: $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\left(N_{1\text{k}}{U}_{\text{k}}\cdot {\displaystyle \underset{j=\mathrm{1,}j\ne k}{\overset{n}{\Pi }}{L}_{\text{sj}}\cdot {\phi }_{\text{ki}}}\right)}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\left(N_{1\text{k}}{U}_{\text{k}}\cdot {\displaystyle \underset{j=\mathrm{1,}j\ne k}{\overset{n}{\Pi }}{L}_{\text{j}}}\right)}}$$

which can also be rewritten in the form: $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{\text{k1}}{N}_{1\text{k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{1\text{k}}{U}_k}{L_{sk}}}}$$

For example, in a DAB (two port) converter, at the primary side as the reference port 1 and the relative load angle between port 2 and port 1 from φ ₂₁ = φ ₂ -φ ₁ the absolute load angle at port 1 and 2 from φ ₁ and φ ₂ can each be calculated using the above equation.

In the example of a three-port Active Bridge converter with the reference port 1 and the given load angles between port 2 and port 1 from φ ₂₁ = φ ₂ -φ ₁ and port 3 and port 1 of φ ₃₁ = φ ₃ -φ ₁ can be the absolute load angle at port 1 . 2 and 3 from φ ₁ . φ ₂ and φ ₃ are each calculated using the equations mentioned above.

Therefore, the device enables a multidirectional multiport DC-DC converter to avoid DC offsets in the magnetizing current of the transformer and to control the DC offset during load transients during transient conditions.

In one embodiment of the invention, the device becomes one duty cycle of the switching period T implemented by 50%. For normal steady state operations, the 50% duty cycle enables simple power control with symmetrical voltage and current waveforms.

In another embodiment, the phase shifts between the phase branches of any port are 360 ° / m. Therefore, in a three-phase dual-active-bridge converter, the phase shift between the phase branches of each port is 120 °. It is the normal mode of operation of the DAB converter to maintain a soft-switching operation to reduce switching losses and EMI.

In a further preferred embodiment, the switching is reversed or cyclically reversed. The switching can be reversed or cyclically reversed due to the symmetry of the switching states.

In one embodiment, the n-port m-phase active bridge converter is an n-port three-phase active bridge converter. Three-phase MAB converters can offer many advantages. Among other things, they require much smaller DC capacitors for a given voltage ripple. Furthermore, the currents in the semiconductors at turn-off are lower than in single-phase MABs, which can significantly reduce circuit losses.

In another embodiment, the n-port m-phase active bridge converter is a two-port m-phase active bridge converter, i. an m-phase DAB converter. The multi-phase DAB converter enables a nested operation, thereby reducing the filter size and the turn-off currents of the semiconductor devices.

wobei U₁ und U₂ die DC-Spannungen an Port 1 und Port 2 bezeichnen, L_s1 und L_s2 die Leckinduktivitäten von Port 1 und Port 2 bezogen auf die Seite von Port 1 bezeichnen, φ₁ den Absolut-Last-Winkel von Port 1 berechnet gemäß dem Verfahren nach Anspruch 1 bezeichnet, φ₂₁ den vorgegebenen Relativ-Last-Winkel zwischen Port 1 und Port 2 bezeichnet und N₁₂ das Windungsverhältnis des Transformators zwischen Port 1 und Port 2 bezeichnet und zum Einstellen des Schaltens der Phasenzweige mit einem entsprechenden Lastwinkel verwendet wird. Port 1 und Port 2 können auch als „Primärseite“ und „Sekundärseite“ bezeichnet werden. Wenn der Laststrom auf herkömmliche Weise verändert wird, werden entweder die primärseitigen oder die sekundärseitigen Phasenzweige in festen Intervallen geschaltet. Daher wird der Lastwinkel durch Modifizieren der Phasenverschiebungen der jeweiligen anderen Phasenzweige eingestellt, d.h. der Lastwinkel wird durch Modifizieren der Phasenverschiebung entweder an den primärseitigen oder den sekundärseitigen Phasenzweigen eingestellt. Bei einer bevorzugten Ausführungsform ändern sich jedoch die Lastwinkel sämtlicher Phasenzweige innerhalb der Primär- und der Sekundärseite, um ein DC-Offset im Magnetisierungsstrom zu vermeiden. Bei einem Beispiel eines DAB-Wandlers mit einem vorgegebenen Lastwinkel φ und x=x_DAB kann die Verteilung auf eine Primärphasenverschiebung φ_p und die sekundärseitige Phasenverschiebung φ_s ebenfalls wie folgt geschrieben werden: $${\mathrm{\phi }}_{\text{p}}=-{\text{x}}_{\text{DAB}}*\mathrm{\phi }\text{ und }{\mathrm{\phi }}_{\text{s}}=\left(1-{\text{x}}_{\text{DAB}}\right)*\mathrm{\phi }$$

mit x_DAB als Phasenverschiebungsfaktor x für den DAB-Wandler, wobei x_DAB entsprechend der folgenden Gleichung ausgewählt wird: $${\text{x}}_{\text{DAB}}={\text{U}}_{\text{s}}*{\text{L}}_{\text{sp}}/\left({\text{U}}_{\text{p}}*{\text{N}}_{\text{tr}}*{\text{L}}_{\text{ss}}+{\text{U}}_{\text{s}}*{\text{L}}_{\text{sp}}\right),$$

wobei U_s die DC-Spannung auf der Sekundärseite ist, U_p die DC-Spannung auf der Primärseite ist, L_sp die primärseitige Leckinduktivität ist, Lss die sekundärseitige Leckinduktivität (bezogen auf die Sekundärseite) ist, N_tr das Windungsverhältnis des Transformators ist, d.h. die Anzahl von primärseitigen Wicklungen dividiert durch die Anzahl von sekundärseitigen Wicklungen. Daher ermöglicht das Verfahren, dass bei einem multidirektionalen Zweiport-m-Phasen-Active-Bridge-Wandler DC-Offsets im Magnetisierungsstrom des Transformators vermieden werden und der DC-Offset bei transienten Bedingungen von Laständerungen gesteuert wird.In an alternative embodiment, considering a two-port m-phase active bridge converter, a phase shift factor x is given by the following equation: $$x=-{\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_1/{\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_{21}={\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}*L_{\text{s1}}*{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_2/\left(L_{\text{s2}}*{\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_1+{\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}*L_{\text{s1}}*U_2\right)$$

in which U ₁ and U ₂ the DC voltages at port 1 and port 2 describe, L _s1 and L _s2 the leakage inductances of Port 1 and port 2 relative to the side of port 1 describe, φ ₁ the absolute load angle of port 1 calculated according to the method of claim 1, φ ₂₁ the given relative load angle between port 1 and port 2 designated and N ₁₂ the turns ratio of the transformer between port 1 and port 2 is designated and used to adjust the switching of the phase branches with a corresponding load angle. port 1 and port 2 can also be referred to as "primary page" and "secondary page". When the load current is changed in a conventional manner, either the primary-side or the secondary-side phase branches are switched at fixed intervals. Therefore, the load angle is adjusted by modifying the phase shifts of the respective other phase branches, ie, the load angle is adjusted by modifying the phase shift on either the primary-side or the secondary-side phase branches. However, in a preferred embodiment, the load angles of all phase branches within the primary and secondary sides change to avoid a DC offset in the magnetizing current. In an example of a DAB converter with a given load angle φ and x = x _DAB can change the distribution to a primary phase shift φ _p and the secondary-side phase shift φ _s also be written as follows: $${\mathrm{\phi }}_{\text{p}}=-{\text{x}}_{\text{DAB}}*\mathrm{\phi }\text{and}{\mathrm{\phi }}_{\text{s}}=\left(1-{\text{x}}_{\text{DAB}}\right)*\mathrm{\phi }$$

with x _DAB as phase shift factor x for the DAB converter, where x _{DAB is selected} according to the following equation: $${\text{x}}_{\text{DAB}}={\text{U}}_{\text{s}}*{\text{L}}_{\text{sp}}/\left({\text{U}}_{\text{p}}*{\text{N}}_{\text{tr}}*{\text{L}}_{\text{ss}}+{\text{U}}_{\text{s}}*{\text{L}}_{\text{sp}}\right).$$

in which U _s the DC voltage is on the secondary side, U _p the DC voltage is on the primary side, L _sp the primary-side leakage inductance is, Lss is the secondary-side leakage inductance (relative to the secondary side), N _tr is the turns ratio of the transformer, that is, the number of primary-side windings divided by the number of secondary-side windings. Therefore, the method enables DC offsets in the magnetizing current of the transformer to be avoided in a multidirectional two-port m-phase active bridge converter and the DC offset to be controlled by load changes in transient conditions.

In another embodiment, the two-port m-phase active bridge converter is a two-port three-phase active bridge converter, ie, a three-phase DAB converter. The two-port three-phase active bridge converter is the most common topology for m-phase DAB converters. In addition to the above advantages, the control methods and the power electronics components for three-phase systems are already well developed.

• - beim Referenzzeitpunkt to: Schalten von Phasenzweig A1 mit einem Lastwinkel -xφ_21,a und Schalten von Phasenzweig B1 mit einem Lastwinkel (1-x)φ_21,a;
• - beim Referenzzeitpunkt t₀+T/6: Schalten von Phasenzweig A3 mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B3 mit einem Lastwinkel (1-X)φ_21,t;
• - beim Referenzzeitpunkt t₀+T/3: Schalten von Phasenzweig A2 mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B2 mit einem Lastwinkel (1-X)φ_21,t;
• - beim Referenzzeitpunkt t₀+T/2: Schalten von Phasenzweig A1 mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B1 mit einem Lastwinkel (1-X)φ_21,t;
• - beim Referenzzeitpunkt t₀+2T/3: Schalten von Phasenzweig A3 mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B3 mit einem Lastwinkel (1-X)φ_21,b;
• - beim Referenzzeitpunkt t₀+5T/6: Schalten von Phasenzweig A2 mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B2 mit einem Lastwinkel (1-x)φ_21,b; und
• - beim Referenzzeitpunkt t₀+T: Schalten von Phasenzweig A1 mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B1 mit einem Lastwinkel (1-X)φ_21,b,

wobei x der Phasenverschiebungsfaktor x ist. Unter Anwendung der oben genannten Schaltsequenz muss der Kern des Transformators nicht überdimensioniert werden, um ein DC-Offset zu vermeiden.An embodiment used in two-port three-phase active bridge converters, direct flow control (IFC) may be combined with a current control (CC) for a three-phase DAB resulting in an instantaneous flow control (Instantaneous) process Flux Current Control Method - IFCC). This allows a highly dynamic load angle change of the predetermined relative load angle of a first load angle φ _{21, a} to a second load angle φ _{21, b} , wherein a mean load angle φ _{21, t} as φ _{21, t} = (exp (-1 / (2 * f * τ)) * (φ _{21, a} + φ _{21, b} ) / (exp (-1 / (2 * f * τ)) + 1) with τ = (L _s1 + N ₁₂ ^{2 *} L _s2 ) / R _w , where R _w is a total winding resistance of the transformer comprising six intervals to + T / 6, to + T / 3, to + T / 2, to + 2T / 3, to + 5T / 6, to + T from a starting point to; full:

• - at reference time to: switching phase branch A1 with a load angle -xφ _{21, a} and switching of phase leg B1 with a load angle ( 1 -x) φ _{21, a} ;
• - at reference time t ₀ + T / 6: switching of phase branch A3 with a load angle -xφ _{21, t} and switching of phase leg B3 with a load angle (1-X) φ _{21, t} ;
• - at reference time t ₀ + T / 3: switching of phase branch A2 with a load angle -xφ _{21, t} and switching of phase leg B2 with a load angle (1-X) φ _{21, t} ;
• - at reference time t ₀ + T / 2: switching of phase branch A1 with a load angle -xφ _{21, t} and switching of phase leg B1 with a load angle (1-X) φ _{21, t} ;
• - at reference time t ₀ + 2T / 3: switching of phase branch A3 with a load angle -xφ _{21, b} and switching of phase leg B3 with a load angle (1-X) φ _{21, b} ;
• - at reference time t ₀ + 5T / 6: switching of phase branch A2 with a load angle -xφ _{21, b} and switching of phase leg B2 with a load angle (1-x) φ _{21, b} ; and
• - at reference time t ₀ + T: switching of phase branch A1 with a load angle -xφ _{21, b} and switching of phase leg B1 with a load angle (1-X) φ _{21, b} ,

where x is the phase shift factor x. Using the above switching sequence, the core of the transformer need not be oversized to avoid DC offset.

• - Aufteilen einer Schaltperiode T in sechs Intervalle t₀+T/6, t₀+T/3, t₀+T/2, t₀+2T/3, t₀+5T/6, t₀+T;
• - zur Zeit t=t₀ Einschalten der oberen aktiven Schalter der ersten und der zweiten Phasenzweige und Einschalten der unteren aktiven Schalter der dritten Phasenzweige sämtlicher n Ports;
• - zur Zeit t₀+T/6 Ausschalten der oberen aktiven Schalter der ersten Phasenzweige und der unteren aktiven Schalter der dritten Phasenzweige und Einschalten der unteren aktiven Schalter der ersten Phasenzweige und der oberen aktiven Schalter der dritten Phasenzweige sämtlicher n Ports;
• - zur Zeit t₀+T/3 Ausschalten der oberen aktiven Schalter der zweiten Phasenzweige und Einschalten der unteren aktiven Schalter der zweiten Phasenzweige sämtlicher n Ports;
• - zur Zeit t₀+T/2 Ausschalten der unteren aktiven Schalter der ersten Phasenzweige und Einschalten der oberen aktiven Schalter der ersten Phasenzweige sämtlicher n Ports;
• - zur Zeit t₀+2T/3 Ausschalten der oberen aktiven Schalter der dritten Phasenzweige und Einschalten der unteren aktiven Schalter der dritten Phasenzweige sämtlicher n Ports; und
• - zur Zeit t₀+5T/6 Ausschalten der unteren aktiven Schalter der zweiten Phasenzweige und Einschalten der oberen aktiven Schalter der zweiten Phasenzweige sämtlicher n Ports.

When starting a two-port three-phase active bridge converter, when the DC-DC converter drives the transformer without transferring active power from the input DC bus (input port) to the output DC bus (output port) Applying a special switching sequence of the active switches, an immediate magnetization correction of the transformer core can be activated and thus DC offsets in the magnetizing currents of the transformer can be eliminated. To avoid saturation as in the prior art, the core must be oversized to avoid this disadvantage. In one embodiment, the steps for starting an n-port three-phase active bridge converter (10) at time t = to, wherein all active switches for t <to are off, each port having a first, a second and a third phase leg corresponding respectively to the first, the second and the third phase leg of the other n-1 ports, and wherein each phase leg has an upper active switch and a lower active switch:

• - Splitting a switching period T in six intervals t ₀ + T / 6, t ₀ + T / 3, t ₀ + T / 2, t ₀ + 2T / 3, t ₀ + 5T / 6, t ₀ + T;
• at time t = t _0, turning on the upper active switches of the first and second phase legs and turning on the lower active switches of the third phase branches of all n ports;
• turning off the upper active switches of the first phase legs and the lower active switches of the third phase legs and turning on the lower active switches of the first phase branches and the upper active switches of the third phase branches of all the n ports; at time t ₀ + T / 6;
• turning off the upper active switches of the second phase legs at the time t ₀ + T / 3 and turning on the lower active switches of the second phase branches of all the n ports;
• turning off the lower active switches of the first phase branches at the time t ₀ + T / 2 and turning on the upper active switches of the first phase branches of all the n ports;
• - at time t ₀ + 2T / 3 turn off the upper active switches of the third phase branches and turn on the lower active switches of the third phase branches of all the n ports; and
• at time t ₀ + 5T / 6, turn off the lower active switches of the second phase legs and turn on the upper active switches of the second phase branches of all the n ports.

Each phase leg of any port has at least one corresponding phase leg in the other n-1 ports. The phase leg and its corresponding phase leg are directly connected via the transformer. For example, in the following embodiment, the phase branches A1 and B1 corresponding phase branches as well A2 and B2 , and A3 corresponds to B3 , If a three-port three-phase active bridge converter is present, the phase legs would be A1 . B1 and C1 corresponding phase branches, and A2 . B2 , C2 etc. This method is also called a soft-start method.

In another embodiment, the number of ports for the soft-start method is n = 2, resulting in a three-phase dual-active-bridge converter, where the first port is a primary side and the second port is a secondary side.

The invention further relates to a system comprising an n-port m-phase active bridge converter and for reducing at least one initial DC offset in a magnetizing current I _M is suitable, wherein in the n-port m-phase active bridge converter k, i ⊆ 1, 2, .., n is a number of a freely selectable port with m ≥ 1 and n ≥ 2, each port can be operated as an input or output port, wherein each of the ports m phase shifter having multiple active switches, wherein a phase shift between the phase branches of each port, wherein the ports can convert a DC voltage into an AC voltage and vice versa, wherein the Ports are connected via an m-phase transformer or via separate m transformers, each connected to each of the m phase legs of each of the n ports, for transferring power between the ports, and a control unit intended to operate the switches of the m phase branches according to the method of claim 1. The n Port-Active-Bridge converter can have one phase, two phases, three phases, four phases, five phases or more phases.

In one embodiment, the system's n-port m-phase active bridge converter is an n-port three-phase active bridge converter. As mentioned above, the three-phase converter reduces the size of the filter capacitors and the turn-off currents of the semiconductor devices. In addition, the control methods and power electronics components for three-phase systems are well developed.

In an alternative embodiment, the control unit is provided to operate the active switches according to the soft-start method for an n-port three-phase active bridge converter.

In another embodiment, the system's n-port m-phase active bridge converter is a two-port three-phase active bridge converter, i. a three-phase DAB converter.

In a further embodiment, the control unit of the system is arranged to operate the switches of the three phase legs according to the said methods applied for operation in a two-port three-phase active bridge converter.

In a further embodiment, the invention relates to a system in which the control unit is arranged to operate the switches of the three phase legs according to the method for starting the two-port three-phase active bridge converter.

In one embodiment, the system's n-port m-phase active bridge converter is a two port m-phase active bridge converter, i. an m-phase DAB converter.

In an alternative embodiment, the control unit of the system is arranged to switch the active switches in accordance with the direct flow control for two port m-phase active bridge converters.

In one embodiment, the control unit of the system is part of the n-port m-phase active bridge converter, thereby forming a compact unit without requiring the dual-active-bridge converter in an additional step with an external control unit connect to.

list of figures

• 1 zeigt eine schematische Ansicht eines Zweiport-Dreiphasen-Active-Bridge-Wandlers, d.h. DAB;
• 2 zeigt eine Einphasen-Ersatzschaltung der Transformatoren für a) einen Zweiport-m-Phasen-Active-Bridge-Wandler; b) einen Dreiportm-Phasen-Active-Bridge-Wandler;
• 3 zeigt eine unmittelbare Flusssteuerung zum Schalten eines Zweiport-Dreiphasen-Active-Bridge-Wandlers zum Vermeiden eines DC-Offsets im Magnetisierungsstrom;
• 4 zeigt Spannungswellenformen einer AC-Eingangsspannung (durchgezogene Linien) und einer AC-Ausgangsspannung (gestrichelte Linien) der unmittelbaren Flusssteuerung mit einer dynamischen Lastwinkeländerung gemäß der vorliegenden Erfindung als Funktion der Zeit t;
• 5 zeigt Spannungswellenformen einer AC-Eingangsspannung (durchgezogene Linien) und einer AC-Ausgangsspannung (gestrichelte Linien) des Soft-Start-Verfahrens gemäß der vorliegenden Erfindung für einen Zweiport-Dreiphasen-Active-Bridge-Wandler als Funktion der Zeit t;
• 6 zeigt Spannungswellenformen von Relativ-Last-Winkel φ₂₁(°), Dreiphasenströmen I_PH(A) und Magnetisierungsstrom IM(A) für Start und Lastwinkeländerung mit einer dem Stand der Technik entsprechenden unmittelbaren Stromsteuerung als Funktion der Zeit t;
• 7 zeigt Stromwellenformen von Relativ-Last-Winkel φ₂₁(°), Dreiphasenströmen I_PH(A) und Magnetisierungsstrom IM(A) für Start und Lastwinkeländerung gemäß dem Soft-Start-Verfahren der vorliegenden Erfindung als Funktion der Zeit t;
• 8 zeigt a) Heranzoomen der Magnetisierungsströme von 6 ohne Start-Verfahren; b) Heranzoomen der Magnetisierungsströme von 7 mit dem Soft-Start-Verfahren;
• 9 zeigt Stromwellenformen von φ₂₁(°),I_PH(A) und IM(A) für Start und Lastwinkeländerung gemäß dem Soft-Start-Verfahren und dem Verfahren zur unmittelbaren Flussstromsteuerung der vorliegenden Erfindung als Funktion der Zeit t;
• 10 zeigt a) Heranzoomen der Magnetisierungsströme während der Leistungsumkehr in 7 nur mit dem Soft-Start-Verfahren; b) Heranzoomen der Magnetisierungsströme während der Leistungsumkehr in 9 mit dem Soft-Start-Verfahren und der unmittelbaren Flusssteuerung;
• 11 zeigt ein Verfahren zur unmittelbaren Flusssteuerung für einen n-Port-m-Phasen-Active-Bridge-Wandler gemäß der vorliegenden Erfindung; und
• 12 zeigt ein System gemäß der vorliegenden Erfindung, wobei (a) die Steuereinheit separat von dem n-Port-m-Phasen-Active-Bridge-Wandler angeordnet ist und (b) die Steuereinheit Teil des n-Port-m-Phasen-Active-Bridge-Wandlers ist.

These and other aspects of the invention are more fully disclosed in the following drawings:

• 1 shows a schematic view of a two-port three-phase active bridge converter, ie DAB;
• 2 shows a single-phase equivalent circuit of the transformers for a) a two-port m-phase active bridge converter; b) a three port phase-to-active bridge converter;
• 3 shows an immediate flow control for switching a two-port three-phase active bridge converter to avoid a DC offset in the magnetizing current;
• 4 FIG. 12 shows voltage waveforms of an AC input voltage (solid lines) and an AC output voltage (dashed lines) of the direct flow control with a dynamic load angle change according to the present invention as a function of time t ;
• 5 FIG. 12 shows voltage waveforms of an AC input voltage (solid lines) and an AC output voltage (broken lines) of the soft-start method according to the present invention for a two-port three-phase active bridge converter as a function of time t ;
• 6 shows voltage waveforms of relative load angle φ ₂₁ (°) , Three-phase currents I _PH (A) and magnetizing current IM (A) for start and load angle change with prior art instantaneous current control as a function of time t ;
• 7 shows current waveforms of relative load angle φ ₂₁ (°) , Three-phase currents I _PH (A) and magnetizing current IM (A) for start and load angle change according to the soft-start method of the present invention as a function of time t ;
• 8th shows a) Zoom in of the magnetization currents of 6 without start procedure; b) Zooming in of the magnetization currents of 7 with the soft-start procedure;
• 9 shows current waveforms of φ ₂₁ (°) . I _PH (A) and IM (A) for start and load angle change according to the soft-start method and method for immediate flow-current control of the present invention as a function of time t ;
• 10 shows a) zooming of the magnetization currents during power reversal in 7 only with the soft-start procedure; b) Zooming in of magnetization currents during power inversion 9 with the soft-start method and the immediate flow control;
• 11 shows a method for immediate flow control for an n-port m-phase active bridge converter according to the present invention; and
• 12 shows a system according to the present invention, wherein (a) the control unit is arranged separately from the n-port m-phase active bridge converter and (b) the control unit is part of the n-port m-phase active Bridge converter is.

Detailed description of the embodiments

1 shows a schematic view of an n-port m-phase active bridge converter with n = 2 and m = 3, which is also referred to as a three-phase dual active bridge (DAB) converter. The n-port m-phase active bridge converter 10 has a number of freely selectable ports labeled k, i ⊆ 1, 2, .., n and where m ≥ 1 and n ≥ 2. Every port 1 . 2 . 3 . .. ., n can be operated as an input or output port. Each of the ports 1 . 2 . 3 . .. ., n also has at least m phase branches 16 with several active switches 17 . 19 on, in the present example, three phase branches. A phase shift 20 is between the phase branches 16 every port 1 . 2 . 3 . .. ., n Are defined. While the term "phase shift" means a difference in the phases between AC voltages of the phase legs within the same port, as in 5 2, the verb "phase shift" or "phase shift" may be a modification of an absolute load angle to be shifted φ ₁ . φ ₁ , ..., φ _i describe. This can only affect one phase branch. The latter may also be described as "changing a load angle γ to a load angle z" to avoid confusion. The ports can have a DC voltage U in an AC voltage v and vice versa, where, in standard operation, the input port (s) is a DC input voltage U _in into an AC output voltage, and the output port (s) converts an AC output voltage to a DC output voltage U _out converts (convert). The ports 1 . 2 . 3 . .. ., n are over an m-phase transformer 25 or via separate m transformers 25 , each with each of the m phase branches 16 each of the n ports 1 . 2 . 3 . .. ., n are connected to transmit power to the ports. The switches 17 . 19 the phase branches 16 be through a control unit 30 according to the direct flow control 100 operated to at least an initial DC offset 32 in a magnetizing current I _M to reduce. The output port of the two-port Active Bridge (DAB) converter can be used as the primary side 14 be designated, the three phase branches A1 . A2 . A3 16 with an upper active switch 17 and a lower active switch 19 per phase branch 16 having.

The output port of the DAB converter can be used as a secondary side 15 be designated, the three phase branches B1 . B2 . B3 16 with an upper active switch 17 and a lower active switch 19 per phase branch 16 having. The potential nodes 11 . 12 . 13 on the phase branches of the primary side and 21 . 22 . 23 on the phase branches of the secondary side are connected to the transformer 25 connected.

2 shows a single-phase (m = 1) equivalent circuit for a) a two-port transformer (n = 2) and b) a three-port transformer (n = 3). v ₁ . v ₂ and v ₃ denote an immediate phase voltage of port 1 . 2 and 3 the single-phase equivalent circuit, and L _s1 . L _s2 . L _s3 denote the leakage inductance of the transformer winding of port 1 . 2 . 3 relative to the side of port 1 in the single-phase T Replacement circuit and L _m the magnetizing inductance, which is usually significantly higher than the leakage inductances. 2 B) For example, to derive the equation for calculating a load angle φ _i with _i ≤ 3 and then the extrapolation to i = n, where n is the number of ports, may be used. For calculating the DC link voltage U ₁ . U ₂ . U ₃ at the port 1 . 2 . 3 becomes the transformer winding ratio N ₁₁ . N ₁₂ . N ₁₃ between the port 1 and port 1 . 2 . 3 where N11 = 1, further defined. The default relative load angle φ ₂₁ . φ ₃₁ from port 2 and 3 relative to port 1 and the difference in the flux of force in the magnetizing inductance ΔΦ _m in making a step of φ ₂₁ , φ ₃₁ = 0 to arbitrary one φ ₂₁ . φ ₃₁ are also defined. To calculate the desired load angle φ ₁ . φ ₂ . φ ₃ for direct flow control of a freely chosen phase of port 1 . 2 . 3 relative to a fixed referenced waveform, the following is indicated: $$\begin{array}{llll}\hfill & {\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_{21}={\phi }_2-{\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_1\hfill & {\phi }_{31}={\phi }_3-{\phi }_1\hfill & \left(1\right)\hfill \\ \iff \hfill & {\mathrm{<figure-callout\; id="2"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_2={\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_{21}+{\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_1\hfill & {\mathrm{<figure-callout\; id="3"\; label="\phi "\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_3={\phi }_{31}+{\phi }_1\hfill & \left(2\right)\hfill \end{array}$$

The equation for the magnetization AC voltage v _m in the equivalent circuit of 2 B is: $$\begin{array}{ll}{\mathrm{\nu }}_{\text{m}}=\frac{L_{\text{s2}}//L_{\text{s3}}}{L_{\text{s1}}+L_{\text{s2}}//L_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="\nu "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\nu </figure-callout>}}_1+\frac{L_{\text{s1}}//L_{\text{s3}}}{L_{\text{s2}}+L_{\text{s1}}//L_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\nu </figure-callout>}}_2+\frac{L_{\text{s1}}//L_{\text{s2}}}{L_{\text{s3}}+L_{\text{s1}}//L_{\text{s2}}}\cdot {\mathrm{<figure-callout\; id="13"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{13}{\mathrm{\nu }}_3\hfill & \left(3\right)\hfill \\ \iff {\mathrm{\nu }}_{\text{m}}=\frac{L_{\text{s2}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="\nu "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\nu </figure-callout>}}_1+\frac{L_{\text{s1}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\nu </figure-callout>}}_2\hfill & \hfill \\ +\frac{L_{\text{s1}}{L}_{\text{s2}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="13"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{13}{\mathrm{\nu }}_3\hfill & \left(4\right)\hfill \end{array}$$

If all given relative load angles are equal to zero, the difference in magnetizing force flux is: $$\begin{array}{l}\Delta {\mathrm{\Phi }}_{\text{m}}\alpha \frac{L_{\text{s2}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{1\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}1}+\frac{L_{\text{s1}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot n_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{2\mathrm{<figure-callout\; id="2"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}2}\\ +\frac{L_{\text{s1}}{L}_{\text{s2}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{13}{\mathrm{<figure-callout\; id="3"\; label="U"\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">U</figure-callout>}}_{3\mathrm{<figure-callout\; id="3"\; label="\phi "\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">\phi </figure-callout>}3}\stackrel{!}{=}0\end{array}$$

wobei φ₁ auch in der folgenden Form geschrieben werden kann: $${\phi }_1=-\frac{\frac{{\phi }_{21}{N}_{12}{U}_2}{L_{\text{s2}}}+\frac{{\phi }_{31}{N}_{13}{U}_3}{L_{\text{s3}}}}{\frac{U_1}{L_{\text{s1}}}+\frac{N_{12}{U}_2}{L_{\text{s2}}}+\frac{N_{13}{U}_3}{L_{\text{s3}}}}$$

und φ₂ sowie φ₃ durch Einsetzen von φ₁ aus Gleichung (7) in (2) abgeleitet werden. Die generalisierte Gleichung für einen beliebigen Lastwinkel φ_i von Port i mit i ⊆ 1, 2,.., n: $${\phi }_i=-\frac{{\displaystyle \sum _{k=\mathrm{1,}k\ne i}^n\left(N_{\text{1k}}{U}_k\cdot {\displaystyle \prod _{j=\mathrm{1,}j\ne k}^n{L}_{\text{sj}}\cdot {\phi }_{\text{ki}}}\right)}}{{\displaystyle \sum _{k=1}^n\left(N_{\text{1k}}{U}_{\text{k}}\cdot {\displaystyle \prod _{j=\mathrm{1,}j\ne k}^n{L}_{\text{j}}}\right)}}$$

die auch geschrieben werden kann als: $${\phi }_i=-\frac{{\displaystyle \sum _{k=\mathrm{1,}k\ne i}^n\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \sum _{k=1}^n\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

When substituting equation (2) and resolving for φ ₁ $$\begin{array}{ll}\iff \frac{L_{\text{s2}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot {\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{1\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}1}+\frac{L_{\text{s1}}{L}_{\text{s3}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{12}{U}_{2\left(\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}21+\phi 1\right)}\hfill & \hfill \\ +\frac{L_{\text{s1}}{L}_{\text{s2}}}{L_{\text{s1}}{L}_{\text{s2}}+L_{\text{s2}}{L}_{\text{s3}}+L_{\text{s1}}{L}_{\text{s3}}}\cdot N_{13}{U}_{3\left(\phi 31+\phi 1\right)}=0\hfill & \left(6\right)\hfill \\ \iff \mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}1=-\frac{L_{\text{s1}}{L}_{\text{s3}}{N}_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_{2\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}21}+L_{\text{s1}}{L}_{\text{s2}}{N}_{13}{\mathrm{<figure-callout\; id="3"\; label="U"\; filenames="DE202017007011U1\_0023.png,DE202017007011U1\_0024.png"\; state="\{\{state\}\}">U</figure-callout>}}_{3\phi 31}}{L_{\text{s2}}{\mathrm{<figure-callout\; id="1"\; label="L\; s3\; U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">L\; </figure-callout>}}_{\text{s3}}{U}_{}{}_1+L_{\text{s1}}{L}_{\text{s3}}{N}_{12}{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_2+L_{\text{s1}}{L}_{\text{s2}}{N}_{13}{U}_3}\hfill & \left(7\right)\hfill \end{array}$$

where φ _{1 can} also be written in the following form: $${\phi }_1=-\frac{\frac{{\phi }_{21}{N}_{12}{U}_2}{L_{\text{s2}}}+\frac{{\phi }_{31}{N}_{13}{U}_3}{L_{\text{s3}}}}{\frac{U_1}{L_{\text{s1}}}+\frac{N_{12}{U}_2}{L_{\text{s2}}}+\frac{N_{13}{U}_3}{L_{\text{s3}}}}$$

and φ ₂ such as φ ₃ by inserting φ ₁ from equation (7) in (2). The generalized equation for any load angle φ _i of port i with i ⊆ 1, 2, .., n: $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\left(N_{\text{1k}}{U}_k\cdot {\displaystyle \underset{j=\mathrm{1,}j\ne k}{\overset{n}{\Pi }}{L}_{\text{sj}}\cdot {\phi }_{\text{ki}}}\right)}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\left(N_{\text{1k}}{U}_{\text{k}}\cdot {\displaystyle \underset{j=\mathrm{1,}j\ne k}{\overset{n}{\Pi }}{L}_{\text{j}}}\right)}}$$

which can also be written as: $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

For the immediate flow control to be implemented for switching an n-port m-phase active bridge converter to avoid a DC offset in the magnetizing current, the switching period should T in m intervals 38 divided up 110 if m is an even number and in 2 * m intervals 38 if m is an odd number. The intervals 38 even define interval limits 40 for switching the active switches 17 . 19 , where each interval defines two boundaries (left, right or beginning, end). The power of the n-port m-phase active bridge converter to be transmitted 10 corresponding to a given relative load angle φ _i1 = φ _i -φ ₁ between a load angle φ _i a freely selectable port i and an absolute load angle φ ₁ a reference port 1 corresponds, should be predetermined 120 become. The active switches 17 . 19 the n ports 1 should be switched like this 130 be that every phase branch 16 a freely selectable port i around the absolute load angle φ _i relative to the interval limits 40 any of the intervals 38 in which the AC voltage v the m phase branches 16 their sign changes, shifts, where φ _i as calculated in Equation 10, where U _k the DC voltage is at a freely selectable port k with k ⊆ 1, 2, 3, ..., n and k # i, L _sk a leakage inductance of the transformer winding of port k referenced to port i = 1 is and N _1k a transformer turn ratio between the port i = 1 and port k is and _φk1 the given relative load angle between port k with k ≠ i and port 1 is.

3 Figure 11 shows an immediate flow control for switching an n-port m-phase active bridge converter to avoid a DC offset in the magnetizing current in a simplified two-port three-phase active bridge converter topology for one period T , The primary load angle φ _p corresponds to a general load angle φ _i a port 1 , and the secondary load angle φ _s corresponds to a general load angle φ ₂ a port 2 , Since the number of phases is three, the number of intervals is 38 six. The six intervals define the seven interval limits 40 at times to, to + T / 6, to + T / 3, to + T / 2, to + 2T / 3, t ₀ + 5T / 6 and to + T. An AC voltage of the first phase branch on the primary side 111 is therefore at t ₀ to the load angle φ _p shifted, and accordingly, an AC voltage of the first phase branch on the secondary side 121 therefore at t ₀ around the load angle φ _s postponed. The same AC voltages 111 and 121 are transformed at t ₀ + T / 2 and t ₀ + T φ _p respectively φ _s postponed. The time of the shift always corresponds to the voltage wave when it makes a zero crossing and therefore changes its sign. The reason for the sign change is switching from on to off or vice versa. In the same way, an AC voltage of the second phase leg on the primary side becomes 112 also at t ₀ + T / 3 around the load angle φ _p accordingly, and accordingly, an AC voltage of the second phase leg on the secondary side becomes 122 at t ₀ + T / 3 around the load angle φ _s postponed. The same AC voltages 112 and 122 become um at t ₀ + 5T / 6 φ _p respectively φ _s postponed. An AC voltage of the third phase branch on the primary side 113 then becomes t ₀ + T / 6 around the load angle φ _p shifted, and accordingly, an AC voltage of the third phase branch on the secondary side 123 also at t ₀ + T / 6 around the load angle φ _s postponed. The same AC voltages 113 and 123 become um at t ₀ + 2T / 3 φ _p respectively φ _s postponed. This method is called direct flow control and avoids DC offsets of magnetization currents.

wobei U₁ und U₂ die DC-Spannungen an Port 1 und Port 2 bezeichnen, L_s1 und L_s2 die Leckinduktivitäten von Port 1 und Port 2 bezogen auf die Seite von Port 1 bezeichnen, φ₁ den Absolut-Last-Winkel von Port 1 berechnet gemäß dem Verfahren 100 nach Anspruch 1 bezeichnet, φ₂₁ den vorgegebenen Relativ-Last-Winkel zwischen Port 2 und Port 1 bezeichnet und N₁₂ das Windungsverhältnis des Transformators 25 zwischen Port 1 und Port 2 bezeichnet und zum Einstellen des Schaltens der Phasenzweige mit einem entsprechenden Relativ-Last-Winkel verwendet wird.For a dynamic load angle change, direct flux control can be used to avoid DC offset in the magnetizing current. For a two port m-phase active bridge converter, the method defines a phase shift factor x resulting from the following equation: $$x=-{\mathrm{<figure-callout\; id="1"\; label="\phi "\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_1/{\mathrm{<figure-callout\; id="21"\; label="\phi "\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">\phi </figure-callout>}}_{21}={\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}*L_{s1}*{\mathrm{<figure-callout\; id="2"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_2/\left(L_{\text{s2}}*{\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="DE202017007011U1\_0022.png,DE202017007011U1\_0023.png"\; state="\{\{state\}\}">U</figure-callout>}}_1+{\mathrm{<figure-callout\; id="12"\; label="N"\; filenames="DE202017007011U1\_0022.png"\; state="\{\{state\}\}">N</figure-callout>}}_{12}*L_{\text{s1}}*U_2\right)$$

in which U ₁ and U ₂ the DC voltages at port 1 and port 2 describe, L _s1 and L _s2 the leakage inductances of Port 1 and port 2 relative to the side of port 1 describe, φ ₁ the absolute load angle of port 1 calculated according to the procedure 100 according to claim 1, φ ₂₁ the given relative load angle between port 2 and port 1 designated and N ₁₂ the turns ratio of the transformer 25 between port 1 and port 2 and used to adjust the switching of the phase legs with a corresponding relative load angle.

• - beim Referenzzeitpunkt to: Schalten von Phasenzweig A1, das die Phasenverschiebung einer AC-Spannung 111 ist, mit einem Lastwinkel -xφ_21,a und Schalten von Phasenzweig B1, das die Phasenverschiebung einer AC-Spannung 121 ist, mit einem Lastwinkel (1-x)φ_21,a;
• - beim Referenzzeitpunkt t₀+T/6: Schalten von Phasenzweig A3, das die Phasenverschiebung einer AC-Spannung 113 ist, mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B3, das die Phasenverschiebung einer AC-Spannung 123 ist, mit einem Lastwinkel (1-x)φ_21,t;
• - beim Referenzzeitpunkt t₀+T/3: Schalten von Phasenzweig A2, das die Phasenverschiebung einer AC-Spannung 112 ist, mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B2, das die Phasenverschiebung einer AC-Spannung 122 ist, mit einem Lastwinkel (1-x)φ_21,t;
• - beim Referenzzeitpunkt t₀+T/2: Schalten von Phasenzweig A1, das die Phasenverschiebung der AC-Spannung 111 ist, mit einem Lastwinkel -xφ_21,t und Schalten von Phasenzweig B1, das die Phasenverschiebung der AC-Spannung 121 ist, mit einem Lastwinkel (1-x)φ_21,t;
• - beim Referenzzeitpunkt t₀+2T/3: Schalten von Phasenzweig A3, das die Phasenverschiebung der AC-Spannung 113 ist, mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B3, das die Phasenverschiebung der AC-Spannung 123 ist, mit einem Lastwinkel (1-x)φ_21,b;
• - beim Referenzzeitpunkt t₀+5T/6: Schalten von Phasenzweig A2, das die Phasenverschiebung der AC-Spannung 112 ist, mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B2, das die Phasenverschiebung der AC-Spannung 122 ist, mit einem Lastwinkel (1-x)φ_21,b; und
• - beim Referenzzeitpunkt t₀+T: Schalten von Phasenzweig A1, das die Phasenverschiebung der AC-Spannung 111 ist, mit einem Lastwinkel -xφ_21,b und Schalten von Phasenzweig B1, das die Phasenverschiebung der AC-Spannung 111 ist, mit einem Lastwinkel (1-x)φ_21,b,

wobei x der Phasenverschiebungsfaktor x ist. 4 Figure 11 shows the immediate flux control method used in a two-port three-phase Active Bridge topology. In the circuit diagram of 4 changes the given relative load angle φ _i1 dynamically from a first load angle φ _{21, a} to a second load angle φ _{21, b} where an intermediate load angle φ _{21, t} as φ _{21, t} = (exp (-1 / (2 * f * T)) * (φ _{21, a} + φ _{21, b} ) / (exp (-1 / (2 * f * τ)) + 1) with τ = ( L _s1 + N ₁₂ ^{2 *} L _s2 ) / R _{w is} defined, where R _w is a total winding resistance of the transformer. For the three phases, the switching method comprises six intervals t ₀ + T / 6, t ₀ + T / 3, t ₀ + T / 2, t ₀ + 2T / 3, t ₀ + 5T / 6, t ₀ + T from one starting point t ₀ , and the switching is done as follows:

• - at reference time to: switching phase branch A1 , which is the phase shift of an AC voltage 111 is, with a load angle -xφ _{21, a} and switching phase branch B1 , which is the phase shift of an AC voltage 121 is, with a load angle (1-x) φ _{21, a} ;
• - at reference time t ₀ + T / 6: switching of phase branch A3 , which is the phase shift of an AC voltage 113 is, with a load angle -xφ _{21, t} and switching phase branch B3 , which is the phase shift of an AC voltage 123 is, with a load angle (1-x) φ _{21, t} ;
• - at reference time t ₀ + T / 3: switching of phase branch A2 , which is the phase shift of an AC voltage 112 is, with a load angle -xφ _{21, t} and switching phase branch B2 , which is the phase shift of an AC voltage 122 is, with a load angle (1-x) φ _{21, t} ;
• - at reference time t ₀ + T / 2: switching of phase branch A1 , which is the phase shift of the AC voltage 111 is, with a load angle -xφ _{21, t} and switching phase branch B1 , which is the phase shift of the AC voltage 121 is, with a load angle (1-x) φ _{21, t} ;
• - at reference time t ₀ + 2T / 3: switching of phase branch A3 , which is the phase shift of the AC voltage 113 is, with a load angle -xφ _{21, b} and switching phase branch B3 , which is the phase shift of the AC voltage 123 is, with a load angle (1-x) φ _{21, b} ;
• - at reference time t ₀ + 5T / 6: switching of phase branch A2 , which is the phase shift of the AC voltage 112 is, with a load angle -xφ _{21, b} and switching phase branch B2 , which is the phase shift of the AC voltage 122 is, with a load angle (1-x) φ _{21, b} ; and
• - at reference time t ₀ + T: switching of phase branch A1 , which is the phase shift of the AC voltage 111 is, with a load angle -xφ _{21, b} and switching phase branch B1 , which is the phase shift of the AC voltage 111 is, with a load angle (1-x) φ _{21, b} ,

where x is the phase shift factor x.

• - zur Zeit t=to Einschalten der oberen aktiven Schalter 17 von Phasenzweigen A1, A2, B1 und B2 und Einschalten der unteren aktiven Schalter 19 von Phasenzweigen A3 und B3;
• - zur Zeit t₀+T/6 Ausschalten der oberen aktiven Schalter 17 von Phasenzweigen A1 und B1 und der unteren aktiven Schalter 19 von Phasenzweigen A3 und B3 und Einschalten der unteren aktiven Schalter 19 von Phasenzweigen A1 und B1 und der oberen aktiven Schalter 17 von Phasenzweigen A3 und B3;
• - zur Zeit t₀+T/3 Ausschalten der oberen aktiven Schalter 17 von Phasenzweigen A2 und B2 und Einschalten der unteren aktiven Schalter 19 von Phasenzweigen A2 und B2;
• - zur Zeit t₀+T/2 Ausschalten der unteren aktiven Schalter 19 von Phasenzweigen A1 und B1 und Einschalten der oberen aktiven Schalter 17 von Phasenzweigen A1 und B1;
• - zur Zeit t₀+2T/3 Ausschalten der oberen aktiven Schalter 17 von Phasenzweigen A3 und B3 und Einschalten der unteren aktiven Schalter 19 von Phasenzweigen A3 und B3;
• - zur Zeit t₀+5T/6 Ausschalten der unteren aktiven Schalter 19 von Phasenzweigen A2, B2 und Einschalten der oberen aktiven Schalter 17 von Phasenzweigen A2 und B2.

5 shows the voltage waveform of a so-called "soft start" method for starting the operation of a two-port three-phase active bridge converter, wherein DC offsets are avoided in the magnetizing current. The soft start method comprises the steps of starting the two port three phase active bridge converter 10 at time t = to, where all active switches 17 . 19 for t <to are off. The three-phase DAB converter has a primary side 14 , the three phase branches A1 . A2 . A3 16 with an upper active switch 17 and a lower active switch 19 per phase branch 16 having. The three-phase DAB converter further has a secondary side 15 on, the three phase branches B1 . B2 . B3 16 with an upper active switch 17 and a lower active switch 19 per phase branch 16 having. The voltage 111 is as Potential difference between the potential node 11 of the first phase branch on the primary side and the middle point 18 of the DC link at the input port of 1 Are defined. The voltage 112 is as a potential difference between the potential node 12 of the second phase branch on the primary side and the middle point 18 of the DC link at the input port of 1 Are defined. The voltage 113 is as a potential difference between the potential node 13 of the third phase branch on the primary side and the middle point 18 of the DC link at the input port of 1 Are defined. The tension is corresponding 121 as potential difference between the potential node 21 of the first phase branch on the secondary side and the middle point 28 of the DC link at the output port of 1 Are defined. The voltage 122 is as a potential difference between the potential node 22 of the second phase branch on the secondary side and the middle point 28 of the DC link at the output port of 1 Are defined. The voltage 123 is as a potential difference between the potential node 23 of the third phase branch on the secondary side and the middle point 28 of the DC link at the output port of 1 Are defined. Further shows 5 rectangular voltage waveforms with a duty cycle of 50% and a frequency of f = 1 / T. The switching period T is then divided into six relevant intervals: to + T / 6, to + T / 3, t ₀ + T / 2, t ₀ + 2T / 3, t ₀ + 5T / 6, t ₀ + T, with the interval limits at t0 T / 6 . T / 3 . T / 2 . 2T / 3 . 5T / 6 and T lie. The period T to 2T will not be considered for the time being. In the first switching period, the predetermined relative load angle between the primary-side bridge and the secondary-side bridge 0 , where the switches are operated as follows:

• - at the time t = to turn on the upper active switch 17 from phase branches A1 . A2 . B1 and B2 and turn on the lower active switch 19 from phase branches A3 and B3 ;
• - at time t ₀ + T / 6 turn off the upper active switch 17 from phase branches A1 and B1 and the lower active switch 19 from phase branches A3 and B3 and turn on the lower active switch 19 from phase branches A1 and B1 and the upper active switch 17 from phase branches A3 and B3 ;
• - At time t ₀ + T / 3 turn off the upper active switch 17 from phase branches A2 and B2 and turn on the lower active switch 19 from phase branches A2 and B2 ;
• - At time t ₀ + T / 2 turn off the lower active switch 19 from phase branches A1 and B1 and turning on the upper active switches 17 from phase branches A1 and B1 ;
• - at time t ₀ + 2T / 3 turn off the upper active switch 17 from phase branches A3 and B3 and turn on the lower active switch 19 from phase branches A3 and B3 ;
• - at time t ₀ + 5T / 6 turn off the lower active switch 19 from phase branches A2 . B2 and turning on the upper active switches 17 from phase branches A2 and B2 ,

Consequently, the tension rushes 112 at the phase branch A2 the tension 111 at the phase branch A1 at 120 °. The voltage 113 at the phase branch A3 rushes the tension 111 at the phase branch A1 240 ° after. The voltage 121 at the phase branch B1 rushes the tension 111 at the phase branch A1 0 after. The voltage 122 at the phase branch B2 rushes the tension 112 at the phase branch A2 0 after. And the tension 123 at the phase branch B3 rushes the tension 113 at the phase branch A3 0 after.

6 shows the current transients in the case of starting a three-phase DAB converter and a rapid change in the relative load angle φ ₂₁ by means of the current control according to the prior art. 6 shows that if, then φ ₂₁ is zero, the magnetizing current I _M already has a DC offset. The magnetizing current I _M and the phase current I _PH become a big overshoot 32 excited, which should be avoided if φ ₂₁ changes from negative to positive. The resulting DC offsets 32 in the magnetizing currents are clearly visible and it takes a long time for them to decay.

7 shows the current waveforms of the converter when the switching pattern according to the soft-start method is used. Using this method, the DC offsets become in the magnetizing currents I _M not excited during the start compared to the diagram in 6 is shown. The amplitude of I _M in 7 is much smaller than the amplitude of I _M in 6 , However, when the load angle changes abruptly, the magnetizing current shows I _M still one, albeit smaller than in 6 without soft-start procedure, large overshoot and a visible DC offset 32 ,

8th shows a) a zoom in of magnetization currents of 6 without soft-start procedure; b) Zooming in of magnetizing currents of 7 with soft-start procedure. The initial magnetizing current I _M is in T / 6 stabilized when the soft start method is used. The effect is clear to see that the DC offset 32 is eliminated.

9 FIG. 12 shows the current waveforms of a three-phase DAB converter using the switching pattern according to the soft start method and the switching pattern according to the direct flow control. Compared to 6 and 7 are the DC offsets 32 during launch and during a quick change in the relative load angle smaller, almost invisible. Therefore, it can be said that DC offsets are eliminated. What can be seen during abrupt load changes is just overshoot, which can only occur for half a switching cycle (after which DC offsets 0 are) much smaller. Thus, with the methods of the present invention, excitation of the DC offset during start up and under transient conditions of load changes is avoided.

10 shows a) a zoom in the magnetization currents I _M during a power reversal in 7 only with soft-start procedures; b) Zoom in of the magnetization currents I _M during a power reversal in 9 with soft-start procedure and direct flow control. In 10b) is the magnetizing current in T / 2 during the power reversal stabilized and is the overshoot (DC offset 32 ) much smaller compared to the situation in 10 without immediate flow control.

• - Aufteilen 110 einer Schaltperiode T in m Intervalle 38, wenn m eine gerade Zahl ist, und in 2*m Intervalle 38, wenn m eine ungerade Zahl ist, wobei die Intervalle 38 Intervallgrenzen 40 zum Schalten der aktiven Schalter 17, 19 definieren,
• - Vorbestimmen 120 einer zu übertragenden Leistung des n-Port-m-Phasen-Active-Bridge-Wandlers 10 entsprechend einem vorgegebenen Relativ-Last-Winkel φ_i1=φ_i-φ₁ zwischen einem Lastwinkel φ_i eines frei wählbaren Ports i und einem Absolut-Last-Winkel φ₁ eines Referenzports 1, und
• - Schalten 130 der aktiven Schalter 17, 19 der n Ports 1 derart, dass sich jeder Phasenzweig 16 eines frei wählbaren Ports i um den Absolut-Last-Winkel φ_i relativ zu den Intervallgrenzen 40 eines beliebigen der Intervalle 38, in denen die AC-Spannung v der m Phasenzweige 16 ihr Vorzeichen ändert, verschiebt, wobei φ_i berechnet wird als

$${\phi }_i=-\frac{{\displaystyle \sum _{k=\mathrm{1,}k\ne i}^n\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \sum _{k=1}^n\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

wobei U_k die DC-Spannung an einem frei wählbaren Port k mit k ⊆ 1, 2, 3,..., n und k≠i ist, L_sk eine Leckinduktivität der Transformatorwicklung des Ports k referenziert auf Port i=1 ist und N_1k ein Transformatorwindungsverhältnis zwischen dem Port i=1 und dem Port k ist und φ_k1 der vorgegebene Relativ-Last-Winkel zwischen dem Port k mit k≠i und dem Port 1 ist. 11 shows the procedure 100 for operating an n-port m-phase active bridge converter 10, where k, i ⊆ 1, 2, .., n are a number of arbitrary ports 1 . .. ., n with m ≥ 1 and n ≥ 2, each port 1 . .. ., n can be operated as an input port or output port, with each of the ports 1 . .. ., nm phase branches 16 with several active switches 17 . 19 having a phase shift 20 between the phase branches 16 each port is made, with the ports being a DC voltage U in an AC voltage v and vice versa, where the ports 1 . .. ., n via a m-phase 201 transformer 25 or via separate m transformers 25 , each with each of the m phase branches 16 each of the n ports 1 . .. ., n are connected to a power between the ports 1 . .. , n, the switches 17 . 19 the phase branches 16 via a control unit 30 according to the method 100 be operated to at least an initial DC offset 32 in a magnetizing current I _M to reduce, comprising:

• - split up 110 a switching period T in m intervals 38 if m is an even number and in 2 * m intervals 38 if m is an odd number, with the intervals 38 interval limits 40 for switching the active switches 17 . 19 define,
• - Predict 120 a power to be transmitted of the n-port m-phase active bridge converter 10 according to a predetermined relative load angle φ _i1 = φ _i -φ ₁ between a load angle φ _i a freely selectable port i and an absolute load angle φ ₁ a reference port 1 , and
• - Switch 130 the active switch 17 . 19 the n ports 1 such that each phase branch 16 a freely selectable port i around the absolute load angle φ _i relative to the interval limits 40 any of the intervals 38 in which the AC voltage v the m phase branches 16 their sign changes, shifts, where φ _i is calculated as

$${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

in which U _k the DC voltage is at a freely selectable port k with k ⊆ 1, 2, 3, ..., n and k ≠ i, L _sk a leakage inductance of the transformer winding of port k referenced to port i = 1 is and N _1k a transformer winding ratio between the port i = 1 and the port k is and _φk1 the given relative load angle between the port k with k ≠ i and the port 1 is.

12 Figure 4 shows a system according to the present invention, in which (a) the control unit is arranged separately from the n-port m-phase active bridge converter and (b) the control unit is part of the n-port m-phase active Bridge converter is.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description should be considered illustrative only or exemplary, but not limiting.

Other modifications will become apparent to those skilled in the art from the present disclosure. Such modifications may include other features that are already in the art are known and can be used instead of or in addition to the features already described here.

Variations of the disclosed embodiments may be understood and made practical to those skilled in the art by studying the drawings, the disclosure, and the appended claims. In the claims, the term "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality of elements or steps. The mere fact that certain measures are listed in the different dependent claims does not indicate that a combination of these measures can not be used to advantage.

Reference signs in the claims should not be construed as limiting the scope thereof.

LIST OF REFERENCE NUMBERS

1, 2, 3, ..., n

any port of a DC-DC converter

10

n-Port-m-phase bridge converter Active

11

first potential node of the first phase leg on the primary side, which is connected to the transformer

12

second potential node of the second phase leg on the primary side, which is connected to the transformer

13

third potential node of the third phase branch on the primary side, which is connected to the transformer

14

Primary side of a DAB

15

Secondary side of a DAB

U

DC voltage

v

AC voltage

U _in

DC input voltage

v _in

AC input voltage

U _out

DC output voltage

v _out

AC output voltage

v ₁

AC voltage from port 1

v ₂

AC voltage from port 2

v ₃

AC voltage from port 3

16

phase arms

17

upper active switch

18

middle point of the DC link on the primary side

19

lower active switch

20

phase shift

21

first potential node of the first phase leg on the secondary side, which is connected to the transformer

22

second potential node of the second phase leg on the secondary side, which is connected to the transformer

23

third potential node of the third phase branch on the secondary side, which is connected to the transformer

25

transformer

28

middle point of the DC link on the secondary side

30

control unit

32

DC offset

I _M

magnetizing

L _m

magnetizing

L _s1

Leakage inductance of port 1

L _s2

Leakage inductance of port 2

Ls ₃

Leakage inductance of port 3

T

switching period

38

intervals

40

interval limits

φ

load angle

φ _i1

Relative load angle between a port i referenced to port 1

.phi..sub.i

Absolute load angle of port i

φ _p

Absolute load angle of the primary side of DAB (Port 1 )

φ _s

Absolute load angle of the secondary side of DAB (Port 2 )

A1

first phase branch on the primary side

A2

second phase branch on the primary side

A3

third phase branch on the primary side

B1

first phase branch on the secondary side (corresponding to A1 )

B2

second phase branch on the secondary side (corresponding to A2 )

B3

third phase branch on the secondary side (corresponding to A3 )

111

AC voltage on a primary side of the first phase leg

112

AC voltage on a primary side of the second phase leg

113

AC voltage on a primary side of the third phase branch

121

AC voltage on a secondary side of the first phase leg

122

AC voltage on a secondary side of the second phase leg

123

AC voltage on a secondary side of the third phase leg

200

System according to the present invention

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 5027264 [0004]

Claims (20)

Device (100) for operating an n-port m-phase active bridge converter (10), where k, i ⊆ 1, 2, .., n are a number of a freely selectable port (1, ..., n) where m ≥ 1 and n ≥ 2, where each port (1, ..., n) can be operated as an input or output port, each of the ports (1, ... n) being m phase branches (16) comprising a plurality of active switches (17, 19), wherein a phase shift (20) takes place between the phase branches (16) of each port, wherein the ports can convert a DC voltage (U) into an AC voltage (v) and vice versa, wherein the ports (1) are connected via an m-phase transformer (25) or via separate m transformers (25) respectively connected to each of the m phase legs (16) of each of the n ports (1, ..., n) , are connected to transfer power between the ports (1, ..., n), the switches (17, 19) of the phase branches (16) being operated by a control unit (30) at least one initial DC Offset (32) in a magnetization trom (I _M ), wherein the control unit is configured to: - divide (110) a switching period T into m intervals (38) if m is an even number and in 2 * m intervals (38) if m a odd number, where the intervals (38) define interval limits (40) for switching the active switches (17, 19), - predetermining (120) a power to be transmitted of the n-port m-phase active bridge converter ( 10) corresponding to a predetermined relative load angle φ _i1 = φ _i -φ ₁ between a load angle φ _{i of} a freely selectable port i and an absolute load angle φ _{1 of} a reference port 1, and - switching (130) the active switch (17, 19) of the n ports (1, ..., n) such that each phase branch (16) of a freely selectable port i is offset by the absolute load angle φ _i relative to the interval limits (40) of any one of Intervals (38), in which the AC voltage (v) of the m phase branches (16) changes sign, shifts, where φ _{i is} calculated as $${\phi }_i=-\frac{{\displaystyle \underset{k=\mathrm{1,}k\ne i}{\overset{n}{\Sigma }}\frac{{\phi }_{k1}{N}_{\text{1k}}{U}_k}{L_{sk}}}}{{\displaystyle \underset{k=1}{\overset{n}{\Sigma }}\frac{N_{\text{1k}}{U}_k}{L_{sk}}}}$$

where U _{k is} the DC voltage at an arbitrary port k with k ⊆ 1, 2, 3, ..., n and k ≠ i, L _{sk is} a leakage inductance of the transformer winding of the port k referenced to port i = 1 and N _{1k is} a transformer _{turn ratio} between the port i = 1 and the port k, and φ _{k1 is} the predetermined relative load angle between the port k with k ≠ i and the port 1. Device (100) according to Claim 1 in which one duty cycle of the switching period is 50%. Device (100) according to one of the preceding claims, in which the phase shifts (20) between the phase branches (16) of any one port (1) are 360 ° / m. Device (100) according to one of the preceding claims, in which the switching is reversed or cyclically reversed. The device (100) of any one of the preceding claims, wherein the n-port m-phase active bridge converter (10) is an n-port three-phase active bridge converter. Device after Claim 5 , further configured to start an n-port three-phase active bridge converter (10) at time t = to, all active switches for t <to being off, each port having a first, a second, and a third one Phase branch corresponding respectively to the first, the second and the third phase leg of the other n-1 ports, and wherein each phase leg has an upper active switch and a lower active switch, and which is arranged to: - divide a switching period T into six Intervals t ₀ + T / 6, t ₀ + T / 3, t ₀ + T / 2, t ₀ + 2T / 3, t ₀ + 5T / 6, t ₀ + T; at the time t = to switch on the upper active switches of all the first and second phase branches and switch on the lower active switches (19) of all the third phase branches of all the n ports; turning off the upper active switches of all the first phase legs and the lower active switches of all the third phase legs and turning on the lower active switches of all the first phase legs and the upper active switches of the third phase branches of all the n ports, at time t ₀ + T / 6; turning off the upper active switches of all the second phase legs at the time t ₀ + T / 3 and turning on the lower active switches of all the second phase branches of all the n ports; at time t ₀ + T / 2, turn off the lower active switches of all the first phase legs and turn on the upper active switches of all the first phase branches of all the n ports; - at time t ₀ + 2T / 3 turn off the upper active switches of all third phase legs and turn on the lower active switches of all third phase branches of all n ports; and - at time t ₀ + 5T / 6 turning off the lower active switches of all the second phase legs and turning on the upper active switches of all the second phase legs of all the n ports. Device (100) according to one of Claims 1 to 4 in which the n-port m-phase active bridge converter (10) is a two-port m-phase active bridge converter. Device (100) according to Claim 7 in which a phase shift factor x results from the following equation: $$x=-{\phi }_1/{\phi }_{21}=N_{12}*L_{\text{s1}}*U_2/\left(L_{\text{s2}}*U_1+N_{12}*L_{\text{s1}}*U_2\right)$$

where U ₁ and U ₂ denote the DC voltages at Port 1 and Port 2, L _s1 and L _s2 denote the leakage inductances of Port 1 and Port 2 relative to the port 1 side, φ ₁ denotes the calculated absolute load angle of Port 1, φ ₂₁ denotes the predetermined relative load angle between Port 1 and Port 2, and N ₁₂ denotes the turns ratio of the transformer (25) between Port 1 and Port 2 and is used for adjusting the switching of the phase legs with a corresponding load angle , Device (100) according to Claim 5 or 7 in which the two-port m-phase active bridge converter (10) is a two-port three-phase active bridge converter. Device (100) according to Claim 9 in which the first port is a primary side (14) and the second port is a secondary side (15). Device (100) according to Claim 9 in which the predetermined relative load angle is changed dynamically from a first load angle φ _{21, a} to a second load angle φ _{21, b} , wherein an intermediate load angle φ _{21, t} as φ _{21, t} = (exp (-1 / (2 * f * τ)) * φ _{21, a} + φ _{21, b} ) / (exp (-1 / (2 * f * τ)) + 1) with τ = (L _s1 + N ₁₂ ^{2 *} L _s2 ) / R _w , where R _{w is} a total winding resistance of the transformer, the six intervals to + T / 6, to + T / 3, to + T / 2, to + 2T / 3, to + 5T / 6, to + T from a starting point to, and which is designed for: at the reference time to: switching of phase branch A1 with a load angle -xφ _{21, a} and switching of phase branch B1 with a load angle (1-x) φ _{21, a} ; at the reference time t ₀ + T / 6: switching of phase branch A3 with a load angle -xφ _{21, t} and switching of phase branch B3 with a load angle (1-x) φ _{21, t} ; - at the reference time t ₀ + T / 3: switching of phase arm A2 with a load angle -xφ _{21, t} and switching of phase leg B2 with a load angle (1-x) φ _{21, t} ; at the reference time t ₀ + T / 2: switching of phase branch A1 with a load angle -xφ _{21, t} and switching of phase branch B1 with a load angle (1-x) φ _{21, t} ; at the reference time t ₀ + 2T / 3: switching of phase branch A3 with a load angle -xφ _{21, b} and switching of phase branch B3 with a load angle (1-x) φ _{21, b} ; at reference time t ₀ + 5T / 6: switching phase branch A2 with a load angle -xφ _{21, b} and switching phase branch B2 with a load angle (1-x) φ _{21, b} ; and - at the reference time t ₀ + T: switching phase branch A1 with a load angle -xφ _{21, b} and switching phase branch B1 with a load angle (1-x) φ _{21, b} , where x is the phase shift factor x. A system (200) comprising an n-port m-phase active bridge converter (10) and capable of reducing at least one initial DC offset (32) in a magnetizing current I _M , wherein at the n-port Port-m-phase active bridge converter k, i ⊆ 1, 2, .., n is a number of a freely selectable port with m ≥ 1 and n ≥ 2, wherein each port can be operated as an input or output port wherein each of the ports has m phase branches (16) with a plurality of active switches (17, 19), with a phase shift (20) between the phase branches of each port, the ports being able to convert a DC voltage to an AC voltage and vice versa wherein the ports are connected via an m-phase transformer (25) or via separate m transformers (25) respectively connected to each of the m phase legs (16) of each of the n ports to provide power between the ports and a control unit (30) provided for switching the switches (17, 19) of the m phases branches (16) according to Claim 1 to operate. System (200) after Claim 12 in which the n-port m-phase active bridge converter (10) is an n-port three-phase active bridge converter. System (200) after Claim 13 in which the control unit is provided to switch (17, 19) according to Claim 6 to operate. System (200) after Claim 13 in which the n-port m-phase active bridge converter (10) is a two-port three-phase active bridge converter. System (200) after Claim 15 in which the control unit (30) is provided for switching the switches (17, 19) of the three phase branches (16) according to Claim 9 to operate. System (200) after Claim 15 or 16 in that the control unit (30) is provided for switching on the switches (17, 19) of the three phase branches (16) for starting the two-port three-phase active bridge converter according to Claim 11 to operate. System (200) after Claim 12 in which the n-port m-phase active bridge converter (10) is a two-port m-phase active bridge converter. System (200) after Claim 18 in which the control unit is intended to operate the active switches (17, 19) in accordance with Claim 8 to operate. System (200) according to one of Claims 12 to 19 in which the control unit (30) is part of the n-port m-phase active bridge converter (10).

Images