WO2013075735A1

WO2013075735A1 - High voltage dc/dc converter with transformer driven by modular multilevel converters (mmc)

Info

Publication number: WO2013075735A1
Application number: PCT/EP2011/070629
Authority: WO
Inventors: Pierluigi Tenca; Konstantinos Papastergiou
Original assignee: Abb Technology Ag
Priority date: 2011-11-22
Filing date: 2011-11-22
Publication date: 2013-05-30

Abstract

It is presented a power converter (7) for converting power between a first high voltage direct current, DC, connection (DC1) and a second high voltage DC connection (DC2). The power converter assembly comprises: a first side assembly (1a) comprising at least two first side channels (2a, 5a, 2b, 5b), each arranged between two connectors (DC1+, DC1-) of the first high voltage DC connection (DC1), wherein each one of the first side channels comprises: a converter arm (2a) and a primary winding (5a), such that the converter arms of the first side channels are arranged to balance an alternating current component through the at least two first side channels to prevent the alternating current component from reaching the first high voltage DC connection; and a second side assembly of the same structure as the first side assembly but with a secondary winding (5b). At least two separate electromagnetic couplings are formed, each including a respective primary winding (5a) of a first side channel and a secondary winding (5b) of a second side channel.

Description

HIGH VOLTAGE DC/DC CONVERTER WITH TRANSFORMER DRIVEN BY MODULAR MULTILEVEL CONVERTERS (MMC)

TECHNICAL FIELD

The invention relates to a power converter for converting power between a first high voltage DC (Direct Current) connection and a second high voltage DC connection. BACKGROUND

High voltage power conversion between DC and DC is required for a variety of different applications. One such application is for links related to HVDC (high voltage DC) and DC Grids.

An article entitled "A Versatile DC-DC Converter for Energy Collection and

Distribution using the Modular Multilevel Converter" by Stephan Kenzelmann et al., presented at IET Renewable Power Generation Conference 2011, Edinburgh, UK, 6-8 September 2011, discusses power conversion using the MMC in a back to back configuration including a transformer.

An article entitled "A Methodology for Developing 'Chainlink' Converters" by Dr Colin Oates, Power Electronics and Applications, 2009, EPE '09 presents a chainlink topology in which a DC potential or low frequency AC potential is applied to the left hand side of the converter and two four quadrant chain circuits connect to a symmetrical AC connection such as the terminals of a transformer as shown in its Fig 11. The operation of the circuit is to circulate the AC current between the symmetrical AC source and the chain circuits. To ensure a power balance the DC current must divide equally between the top and bottom chain circuit. If a transformer is used this means that the symmetrical winding must be arranged so that the components of the magneto motive forces (MMF) created by the DC current cancels between the windings and do not bring the magnetic core into saturation An issue in all such converters is the component cost, the fault tolerancy as well as the size. It would be beneficial to find more fault tolerant solutions capable of reducing the converter footprint and volume while also reducing the number of components, which frequently leads to a life cycle cost reduction. SUMMARY

It is an objective of the present invention to alleviate a problem with the prior art discussed above.

According to a first aspect, it is presented a power converter for converting power between a first high voltage direct current, DC, connection, and a second high voltage DC connection. The power converter assembly comprises: a first side assembly comprising at least two first side channels, each arranged between two connectors of the first high voltage DC connection, wherein each one of the first side channels comprises: a converter arm and a primary winding, such that the converter arms of the first side channels are arranged to balance an alternating current component through the at least two first side channels to prevent the mentioned alternating current component from reaching the first high voltage DC connection; and a second side assembly comprising at least two second side channels, each arranged between two connectors of the second high voltage DC connection, wherein each one of the second side channels comprises: a converter arm and a secondary winding, such that the converter arms of the second side channels are arranged to balance an alternating current component through the at least two second side channels to prevent the mentioned alternating current component from reaching the second high voltage DC connection; wherein at least two separate electromagnetic couplings are formed, each including a respective primary winding of a first side channel and a secondary windings of a second side channel.

By providing separate electromagnetic couplings comprising primary and secondary windings, a modularity is achieved. For example, the electromagnetic couplings could be provided using a plurality of single transformers or a multiphase transformer. Moreover, this modularity simplifies construction, assembly and transport of components. By using the leakage inductance of the coupled windings in the side channels, the need for separate filtering inductors is reduced or even eliminated, thereby saving cost and complexity.

The first side assembly may be arranged such that there is an absence of any converter elements between the primary windings and one of the two connectors of the first high voltage DC connection, and the second side assembly is arranged such that there is an absence of any converter elements between the secondary windings and one of the two connectors of the second high voltage DC connection. This is rendered possible by controlling the AC component between the windings in each side assembly. In this way, converter elements on the other side of the respective windings are made redundant, saving complexity and const.

Each one of the converter arms may comprise a plurality of converter cells. The converter cells can be connected in series to increase voltage rating or in parallel to increase current rating. The serially connected converter cells may optionally be individually controlled to achieve a finer granularity in the conversion, e.g. to achieve a more sinusoidal (or square, saw tooth shaped, etc.) power conversion.

Each one of the converter cells may be capable of synthesizing at least two different voltages across its two main terminals.

Each one of the converter cells may be capable of synthesizing voltage values of both signs across its two main terminals.

Each one of the converter cells may be of a full H bridge structure or of a voltage- source multilevel structure.

Each one of the converter cells may be of a half bridge structure

The power converter may be arranged to minimise a value associated with

electromagnetic field distribution in the electromagnetic couplings.

The power converter may be arranged to minimise the value associated with

electromagnetic field distribution in the electromagnetic couplings in order to prevent magnetic saturation from occurring in the electromagnetic couplings.

The power converter may further comprise a controller arranged to determine the value associated with electromagnetic field distribution in the electromagnetic couplings.

The value associated with electromagnetic field distribution in the electromagnetic couplings may be a norm of a magneto-motive force or a norm of magnetic flux in the electromagnetic couplings. The value associated with electromagnetic field distribution in the electromagnetic couplings may be the magnitude of an average value of the magneto-motive force or the magnitude of the average value of the magnetic flux.

The power converter may be arranged to balance the non-zero DC currents through corresponding primary windings and secondary windings in order to prevent operating the at least two electromagnetic couplings in a state of magnetic saturation.

The power converter may further comprise an auxiliary side assembly, the auxiliary side assembly comprising at least two auxiliary side channels, each arranged between two connectors of an auxiliary high voltage DC connection, wherein each one of the auxiliary side channels comprises: an upper converter arm, a lower converter arm and an auxiliary winding arranged between the upper converter arm and the lower converter arm, such that the upper and lower converter arms of the auxiliary side channels are arranged to minimise a magnitude of a DC current through the respective auxiliary windings, and each auxiliary winding respectively forming part of the at least two electromagnetic couplings such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and an auxiliary winding.

The power converter may further comprise a third side assembly of the same structure as the second side assembly, the third side assembly comprising at least two third side channels, each comprising a tertiary winding forming part of the at least two electromagnetic couplings such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and a tertiary winding.

The power converter may further comprise an alternating current, AC, side assembly comprising a plurality of AC tertiary windings, corresponding in numbers with the number of electromagnetic couplings, the AC tertiary windings respectively forming part of the at least two electromagnetic couplings, such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and an AC tertiary winding.

The second side assembly may be used only to minimise a value associated with the electromagnetic field distribution in the electromagnetic couplings in order to prevent magnetic saturation from occurring in said electromagnetic couplings. The value associated with the electromagnetic field distribution in the electromagnetic couplings may be used to prevent magnetic saturation from occurring in said electromagnetic couplings.

The power converter may further comprise a third side assembly of the same structure as the second side assembly, wherein each one of the at least two electromagnetic couplings is a two sided electromagnetic coupling comprising a primary winding and a secondary winding selected from one of the second side assembly and the third side assembly. This provides a split of power in the power transfer to/ from the first side assembly between the second side assembly and the third side assembly. Each one of the electromagnetic couplings may form part of a transformer.

The first side assembly may comprise three first side channels, the second side assembly may comprise three second side channels and three separate electromagnetic couplings may be formed. By using three channels, the power converter can still operate even if one channel fails, albeit with less power capacity. The faulty channel can be isolated with breakers in case it is deemed necessary. The remaining functioning channels can then be controlled in accordance with the new number of channels, e.g. by shifting phase to compensate such that the AC component is still prevented from reaching the DC terminals. Moreover, conventional three-phase transformers can be used for the electromagnetic couplings which can be an advantage when implementing the power converter in terms of logistic cost and ease of implementation. More channels can be added to achieve greater failure handling capacity, if desired.

Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the element, apparatus, component, means, step, etc." are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is now described, by way of example, with reference to the accompanying drawings, in which: Fig 1 is a schematic diagram illustrating a two sided DC to DC power converter according to one embodiment with two channels on either side of the power converter;

Fig 2 is a schematic graph illustrating the time evolution of the voltage over the two converter arms of Fig 1; Figs 3A-B are schematic diagrams illustrating operating domains of non-linear magnetic materials in one example of the electromagnetic couplings, such as transformers;

Fig 4 is a schematic diagram illustrating a two sided DC to DC power converter according to one embodiment with three channels on either side of the power converter; Fig 5 is a schematic diagram illustrating a four sided DC to DC power converter according to one embodiment;

Fig 6 is schematic diagram illustrating a three sided power converter according to one embodiment with two DC sides and one AC side;

Fig 7 is a schematic diagram illustrating a three sided DC to DC power converter according to one embodiment where one side is not used for DC current balancing;

Fig 8 is a schematic diagram illustrating a three sided DC to DC power converter according to one embodiment where power on one side is split between two other sides;

Figs 9A-B are schematic diagrams illustrating possible converter cell arrangements of converter arms of Figs 1 and 4-8.

Figs lOA-C are schematic diagrams illustrating embodiments of converter cells of the converter arms of Figs 9A-B;

Fig 11 A is a schematic diagram illustrating an electromagnetic coupling between a primary and a secondary winding which can be used in the power converter of Figs 1, 4 and 8; Fig 1 IB is a schematic diagram illustrating an electromagnetic coupling between a primary, secondary and tertiary winding which can be used in the power converter of Figs 6-7; and

Fig l lC is a schematic diagram illustrating an electromagnetic coupling between a primary, secondary, tertiary and quaternary winding which can be used in the power converter of Fig 5.

DETAILED DESCRIPTION

The invention will now be described more fully hereinafter with reference to the accompanying drawings, in which certain embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided by way of example so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Like numbers refer to like elements throughout the description. Fig 1 is a schematic diagram illustrating a two sided DC to DC power converter 7 according to one embodiment with two channels on either side of the power converter. A first side assembly la comprises two connectors lOa-b of a first high voltage DC connection DQ. Throughout the description and claims, high voltage is to be interpreted as a voltage of at least several kilovolts. In one embodiment, high voltage is at least one hundred kilovolts. It is to be noted that any DC source connection can be either a constant voltage source connection or a constant current source connection. The first side assembly la comprises two first side channels 30a-b which are both arranged between the two connectors lOa-b of the first high voltage DC connection. Each one of the first side channels 30a-b comprises a converter arm 2a-b and a primary winding 5a-b connected serially between the connectors lOa-b. There are logical points 15a-b between the converter arms 2a-b and the primary windings 5a-b, which are used to explain Fig 2 below. As can be seen there is no converter arm between the windings 5a-b and the negative connector 10b.

On a second side is a second side assembly lb of the same structure as the first side assembly la. The second side assembly lb comprises two connectors l la-b of a second high voltage DC connection DC₂. The second side assembly lb further comprises two second side channels 31a-b which are both arranged between the two connectors l la-b of the second high voltage DC connection. Each one of the second side channels 31a-b comprises a converter arm 2'a-b and a secondary winding 5'a-b connected serially between the connectors l la-b. A controller 50 controls the operation of the converter arms 2a-b, 2'a-b.

A first electromagnetic coupling is formed comprising one of the primary windings 5a and one of the secondary windings 5'a, connected e.g. via an iron core 4a. Other suitable materials can also be used for the core. Analogously, a second electromagnetic coupling is formed comprising the other one 5b of the primary windings and the other one 5'b of the secondary windings, connected e.g. via an iron core 4b which is separated from the other iron core 4a. Each one of the electromagnetic couplings can form part of a transformer. The two transformers can be individual transformers or part of a multiphase transformer. Other suitable materials, apart from iron can equally well be used to replace the iron cores 4a-b.

The first side assembly la, the second side assembly lb and the electromagnetic couplings 4a-c can be contained in one physical unit, or distributed over several physical units.

The operation of the power converter 7 will be explained, also with reference to Fig 2. In this scenario, power flows from DQ to DC₂ but power could equally well flow in the opposite direction. A DC current flows from DQ⁺ to DQ through the two first side channels 30a-b. The converter arm 2a, as explained in more detail below, comprises a number of converter cells comprising energy storage such that each one can be controlled to supply different voltages. The controller 50 thus makes the converter arm 2a generate voltages between its terminals, connector 10a and the logical point 15a in order to provide an alternating voltage (AC) to the primary winding 5a. A first curve 16a corresponds to the voltage of the connector 10a measured with respect to the logical point 15a. The curve 16a has a maximum of two times the nominal DC high -voltage at the primary side, i.e. 2*_VDC1 when all converter cells operate to supply a maximum negative voltage to the winding 5a. Conversely, the curve 16a has a minimum of zero when all converter cells operate to supply a maximum positive voltage to the winding 5a.

Although the first curve 16a is here sinusoidal, the converter arms could produce any type of alternative voltage waveform on the winding 5a, e.g. square, saw tooth, etc. In order to prevent the AC component of the current in the winding 5a from flowing into the DC terminals the converter arm 2b of the other first side channel is controlled to produce a corresponding, but complementary, voltage of the connector 10a measured with respect to the logical point 15b, as seen by a second curve 16b. The excursion of the voltage of the connector 10a with respect to the logical point 15b is symmetrical around the nominal voltage of the primary side DC , i.e .V_DC1+, ranging between 2* V_DC1+ and 0. In other words, the second curve 16b is a mirror image of the first curve 16a around the nominal voltage V_DC1+. This has the effect that the AC current flowing through the left one 30a of the first side channels is compensated by the AC current flowing through the right one 30b of the first side channels. This arrangement prevents the AC components of the winding currents from reaching either one of the first high voltage DC connectors lOa-b. The net result is that essentially only DC current flows from DQ⁺ to DQ^~.

Through the electromagnetic couplings and cores 4a, 4b, the AC power is transferred from the respective primary windings 5a-b to the respective secondary windings 5'a, 5'b of the second side assembly lb. The converter arms 2'a-b of the second side channels 31a-b are controlled in a corresponding way to the converter arms 2a-b of the first side assembly la to provide a DC output only to DC₂ ⁺ and DC₂ .

The frequency of operation of the converter arms can be selected freely based on arm capacitor voltage ripple, losses etc. However, if an AC side is present, as in Fig 6 below, the frequency is typically selected to be the grid frequency, e.g. often 50 or 60 Hz.

With this structure, a simpler construction is provided, where converter arms are only needed on one side of the windings as seen between the DC terminals. Also, expensive capacitors to allow AC to flow between channels are made redundant. This reduces the number of components that are required compared to the prior art. Moreover, no separate inductor is required for filtering as the leakage inductances of the coupled windings can perform this function. Figs 3A and 3B are schematic diagrams illustrating operating domain for the modulus of the B and H vectors in the B-H static characteristics of non-linear magnetic materials commonly used in electromagnetic couplings, e.g. magnetic iron in transformers. B represents the magnitude of magnetic field and H represents the magnitude of the magnetising field The levelling off of the B-H static characteristic above a certain value of the magnitude of H indicates regions of magnetic saturation. Fig 3A shows a desired operating domain range 60, in a B-H plane. The conventional operating domain 60 is symmetrical around the origin of coordinates..

In Fig 3B, the working range is shifted. This corresponds to an operating condition caused by non-zero average value of the magneto-motive force (MMF) acting on the magnetic material which, in turn, often originates from a DC component in at least one of the currents flowing through the primary and secondary windings. Owing to the action of the controller 50, that simultaneously controls both sides of the

electromagnetic couplings, the working range 60' is displaced, while being prevented from drifting into the saturated region. Whenever there are more than two sides of the electromagnetic coupling, such as illustrated in Figs 5-7 and explained in more detail below, it is sufficient to have two sides whose DC current components are controlled prevent drifting into the saturated region. This control works can for example work by minimising a magnitude of the average value of the the MMF and/ or the magnitude of the average value of the flux. In other words, the magnitude of the average value of the MMF and/ or of the flux is reduced or even nullified.

If there is a fault on one side, there may be an undesired large energy flow. Such a scenario can be mitigated by actively controlling the operating range into saturation. This essentially creates a very low impedance on the path of the undesired energy flow that prevents it from reaching the other sides. By this use, the converter can perform as an emergency decoupler among networks. The power still needs to be taken care of on the failing side, but a propagation of the problem to other networks is stopped in this way. Such an event can for example occur when operating a circuit breaker.

It is to be noted that any hysteresis effects are not shown here in order not to obscure the ideas of the embodiments presented herein. Fig 4 is a schematic diagram illustrating a two sided DC to DC power converter 7 according to one embodiment with three channels on either side of the power converter. There are thus three converter arms 2a-c and three primary windings in the first side assembly la and three converter arms 2'a.c and three secondary windings 5'a-c in the second side assembly lb. A third core 4c is used to connect the third primary winding 5c with the third secondary winding 5'c. The three channels on either side are controlled such that the sum of the AC components in the winding currents is zero. By using three channels, the power converter can still operate even if one channel fails, albeit with less power capacity. The faulty channels can be isolated with breakers in case it is deemed necessary. The remaining functioning channels are controlled in accordance with the new number of channels, e.g. by shifting phase to compensate such that the AC component is still prevented from reaching the DC terminals. Moreover, conventional three-phase transformers can be used for the electromagnetic couplings which can be an advantage when implementing the power converter. More channels can be added to achieve greater failure handling capacity, if desired.

Fig 5 is a schematic diagram illustrating a four sided DC to DC power converter 7 according to one embodiment. Here, the power converter 7 also comprises a third side assembly lc for a third DC connection DC₃ and a fourth side assembly Id for a fourth DC connection DC₄. The third side assembly lc is of a similar structure to the first and second side assemblies la-b and comprises its own converter arms 2"a-c and its own tertiary windings 5"a (the remaining not numbered). The fourth side assembly Id is also of a similar structure to the first and second side assemblies and comprises its own converter arms 2"'a-c and its own quaternary windings 5"'a (the remaining not numbered). The core 4a is used in an electromagnetic coupling comprising respective primary winding 5a, secondary winding 5'a, tertiary winding 5"a and quaternary winding 5"'a. The other electromagnetic couplings between the other channels work in a similar way but are not shown in order not to obscure the diagram. Such a multi-side structure is robust in the way that any side can help to minimise the e.g. the magnitude of an electromagnetic field in the coupling, for example by compensating for a an unbalanced DC current component if one of the side assemblies la-d fails. In one embodiment, one of the side assemblies, such as the third side assembly lc, could be completely devoted towards minimising the magnitude of the electromagnetic field in the coupling by compensating for a still unbalanced DC current component. In such a case, the third side assembly lc would thus not be used for any power transfer.

Fig 6 is schematic diagram illustrating a three sided power converter 7 according to one embodiment with two DC sides and one AC side. On the AC side, an AC side assembly 9 comprises a plurality of AC tertiary windings 8a-c. The number of AC tertiary windings 8a-c is equal in numbers compared to the number of electromagnetic couplings, in this example three. In this way, the AC tertiary windings 8a-c respectively form part of the electromagnetic couplings, such that each one of the electromagnetic couplings comprises a primary winding of the first side assembly la, a secondary winding of the second side assembly lb and an AC tertiary winding of the AC side assembly 9. Since the average value of the MMF magnitude in the electromagnetic couplings is minimised by using the controlled DC current components in the first side assembly la and the second side assembly lb, the AC side assembly 9 does not need to contribute with a DC current components to the minimisation of the average value of the MMF magnitude. This power converter thus implements a DC/DC/AC converter with the advantage of the reduced complexity of the DC sides as explained above.

In one embodiment, the second side assembly lb is only used to minimise the magnitude of average value of the MMF and/ or the magnitude of the average value of the flux in the coupling, not for energy transfer. In order to achieve this purpose, the controller operates the second side assembly by instantaneously regulating its current in such a way that it comprises a proper DC current component, and optionally the rapid variations (ripple) superimposed to it, typically of reduced amplitude. This use essentially creates a power converter 7 which is an AC/DC converter. With such an arrangement, it is possible to exchange AC power directly with a DC grid, also known as DC tapping. In one embodiment, the AC side can be used to connect to an AC grid and the DC sides are short circuited. This implies that no average power is exchanged with the AC grid since the DC sides have zero instantaneous power (being short-circuited). Such a special use of the converter renders it an AC STATCOM (static synchronous compensator), i.e. a device capable of generating desired reactive power in the AC grid. Fig 7 is a schematic diagram illustrating three sided DC to DC power converter 7 according to one embodiment where one side is not used for minimising the average of the electromagnetic field in the couplings by providing a proper DC current component. The power converter 7 here comprises an auxiliary side assembly 27 comprising auxiliary side channels 33a-c, each arranged between two connectors 12a-b of an auxiliary high voltage DC connection DC. Each one of the auxiliary side channels 33a-c comprises an upper converter arm 3a-c, a lower converter arm 3d-f and an auxiliary winding 24a-c arranged in the middle, between the upper converter arm 3a-c and the lower converter arm 3d-f. This structure operates such that the upper and lower converter arms 3a-f of the auxiliary side channels 30a-b are arranged to minimise the magnitude of a DC current component through the respective auxiliary windings. Each auxiliary winding respectively forms part of the electromagnetic couplings, such that each one of the electromagnetic couplings comprises a primary winding, a secondary winding and an auxiliary winding. In this way, conventional DC converter channels can be used without modification, coupled to the electromagnetic coupling, as long as there are at least two sides directed to minimising the average of the electromagnetic field via proper control of their DC components current components, through the

electromagnetic couplings in order to avoid magnetic saturation.

In one embodiment, the second side assembly lb is only used to minimise the magnitude of the average value of the of the electromagnetic field, such as MMF and/ or magnetic flux in the coupling, and not for energy transfer. In order to achieve this purpose, the controller operates the second side assembly by instantaneously regulating its current in such a way that it comprises a proper DC current component, and optionally the rapid variations (ripple) superimposed to it, typically of reduced amplitude.

Fig 8 is a schematic diagram illustrating a three sided DC to DC power converter 7 according to one embodiment where power on one side is split between two other sides. Here, the power converter comprises a first side assembly la, a second side assembly lb' and a third side assembly lc'. The number of windings of the second side assembly lb' plus the number of windings of the third side assembly lc' is equal to the number of windings of the first side assembly la. Each primary winding 5a-d is electromagnetically coupled to either a secondary winding of the second side assembly lb' or a secondary winding of the third side assembly lc'. In this way, the power (in or out) of the first side assembly la is the sum of the power (in or out) of the second side assembly lb' and the third side assembly lc'. The number of channels of the second side assembly lb' can be equal to the number of channels of the third side assembly lc' or they can differ, as long as their sum is equal to the number of channels of the first side assembly la. Figs 9A-B are schematic diagrams illustrating possible converter cell arrangements of converter arms of Figs 1 and 4-8. Fig 9A illustrates the structure of any one of the converter arms 2a-d, 2'a-c, 2"a-c, 2"'a-c, here represented by a single converter arm 2. The converter arm comprises a plurality of converter cells 32a-d, wherein each converter cell 32a-d is controlled by the controller 50. The converter cells 32a-d can be connected in series to increase the voltage rating or in parallel to increase the current rating. The serially connected converter cells 32a-d can optionally be individually controlled to achieve a finer granularity in the conversion, e.g. to achieve a more sinusoidal (or square, saw tooth shaped, etc.) power conversion. While the converter arm is here illustrated to have four converter cells 32a-d, any number of converter cells is possible, including one, two, three or more. In one embodiment, the number of converter cells in each converter arm is in the range from 30 to 1000 converter cells.

Fig 9B illustrates the structure of any one of the converter arms 3a-f of Fig 7, here represented by a single converter arm 3. The converter cells of this converter arm 4 have the same possible configurations as the converter arm 2 of Fig 9A, but can be of a different actual configuration than the converter arm 2 of Fig 9A.

Figs lOA-C are schematic diagrams illustrating embodiments of converter cells 32a-d of the converter arms of Figs 9A-B. Any one of the converter cells 32a-d is here represented as a single converter cell 32. A converter cell 32 is a combination of semiconductor switches, such as transistors, and energy storing elements, such as capacitors, supercapacitors, inductors, batteries, flywheels, fuel cells, etc. Optionally, a converter cell can be a multilevel converter structure such as a flying capacitor or MPC (Multi-Point-Clamped) or ANPC ( Active - Neutral-Point-Clamped ) multilevel structure.

Fig 10A illustrates a converter cell comprising an active component in the form of a switch 40 and an energy storage component 41 in the form of a capacitor. The switch 40 can for example be implemented using a combination of one or more insulated gate bipolar transistors (IGBT), Integrated Gate-Commutated Thyristors (IGCT), a Gate Turn-Off thyristors (GTO), or any other group of suitable high power semiconductor components.

Fig 10B illustrates a converter cell 32 implementing a half bridge structure. The converter cell 32 here comprises a leg of two serially connected active components in the form of switches 40a-b, e.g. IGBTs, IGCTs, GTOs, etc. A leg of two serially connected diodes 42a-b is connected with the leg of serially connected switches 40a-b as shown in the figure. An energy storage component 41 is also provided in parallel with the leg of transistors 40a-b and with the leg of diodes 32a-b. The output voltage synthesized by the converter cell can thus either be zero or the voltage of the energy storage component 41.

Fig IOC illustrates a converter cell 32 implementing a full bridge structure. The converter cell 32 here comprises four switches 40a-d, e.g. IGBTs, IGCTs, GTOs, etc. An energy storage component 41 is also provided in parallel across a first leg of two transistors 40a-b and a second leg of two transistors 40c-d. Compared to the half bridge of Fig 10B, the full bridge structure allows the synthesis of an output voltage capable of assuming both signs, whereby the output voltage of the converter cell can either be zero, the voltage of the energy storage component 41, or the opposite value of the voltage of the energy storage component 41. Fig 11 A is a schematic diagram illustrating an electromagnetic coupling which can be used in the power converter of Figs 1, 4 and 8. The electromagnetic coupling is here implemented as a two sided transformer 18 with a primary winding 5, a secondary winding 5' and an optional suitable core 4. Several two sided transformers 18 can be utilised individually or as part of a multiphase transformer. Fig 1 IB is a schematic diagram illustrating an electromagnetic coupling which can be used in the power converter of Figs 6-7. The electromagnetic coupling is here implemented as a three sided transformer 18' with a primary winding 5, a secondary winding 5', a tertiary winding 5" and an optional suitable core 4.

Fig l lC is a schematic diagram illustrating an electromagnetic coupling which can be used in the power converter of Fig 5. The electromagnetic coupling is here implemented as a four sided transformer 18" with a primary winding 5, a secondary winding 5', a tertiary winding 5", a quaternary winding 5"' and an optional suitable core 4.

The invention has mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended claims.

Claims

1. A power converter (7) for converting power between a first high voltage direct current, DC, connection, and a second high voltage DC connection, the power converter assembly comprising:

a first side assembly (la) comprising at least two first side channels (30a-b), each arranged between two connectors (lOa-b) of the first high voltage DC connection, wherein each one of the first side channels (30a-b) comprises: a converter arm (2a-b) and a primary winding (5a-b), such that the converter arms (2a-b) of the first side channels (30a-b) are arranged to balance an alternating current component through the at least two first side channels (30a-b) to prevent the mentioned alternating current component from reaching the first high voltage DC connection; and

a second side assembly (lb, lb') comprising at least two second side channels (31a-b), each arranged between two connectors (l la-b) of the second high voltage DC connection, wherein each one of the second side channels (31a-b) comprises: a converter arm (2'a-b) and a secondary winding (5'a-b), such that the converter arms (2'a -b) of the second side channels (31a-b) are arranged to balance an alternating component current through the at least two second side channels (31a-b) to prevent the mentioned alternating current component from reaching the second high voltage DC connection;

wherein at least two separate electromagnetic couplings (18, 18', 18") are formed, each including a respective primary winding (5a-b) of a first side channel and a secondary windings (5'a-b) of a second side channel.

2. The power converter (7) according to claim 1, wherein the first side assembly (la) is arranged such that there is an absence of any converter elements between the primary windings and one of the two connectors (1 Oa-b) of the first high voltage DC connection, and the second side assembly (lb) is arranged such that there is an absence of any converter elements between the secondary windings and one of the two connectors (l la-b) of the second high voltage DC connection.

3. The power converter (7) according to claim 1 or 2, wherein each one of the converter arms (2a-b, 2'a-b) comprises a plurality of converter cells (32a-d).

4. The power converter (7) according to claim 3, wherein each one of the converter cells (32a-d) is capable of synthesizing at least two different voltages across its two main terminals.

5. The power converter (7) according to claim 3, wherein each one of the converter cells (32a-d) is capable of synthesizing voltage values of both signs across its two main terminals.

6. The power converter (7) according to any one of claims 3 to 5, wherein each one of the converter cells (32a-d) is of a full H bridge structure.

7. The power converter (7) according to any one of claims 3 to 5, wherein each one of the converter cells (32a-d) is of a half bridge structure

8. The power converter (7) according to any one of the preceding claims, wherein the power converter is arranged to minimise a value associated with electromagnetic field distribution in the electromagnetic couplings (18, 18', 18").

9. The power converter (7) according to any one of the preceding claims, wherein the power converter is arranged to minimise the value associated with electromagnetic field distribution in the electromagnetic couplings (18, 18', 18") in order to prevent magnetic saturation from occurring in the electromagnetic couplings (18, 18', 18").

10. The power converter (7) according to claim 8 or 9, any one of the preceding claims, further comprising a controller arranged to determine the value associated with electromagnetic field distribution in the electromagnetic couplings.

11. The power converter (7) according to any one of claims 8 to 10, wherein the value associated with electromagnetic field distribution in the electromagnetic couplings is a norm of a magneto -motive force or a norm of magnetic flux in the electromagnetic couplings (18, 18', 18").

12. The power converter (7) according to claim 11, wherein the value associated with electromagnetic field distribution in the electromagnetic couplings is the magnitude of an average value of the magneto-motive force or the magnitude of an average value of the magnetic flux.

13. The power converter (7) according to any one of the preceding claims, wherein the power converter is arranged to balance non-zero DC components of the currents through corresponding primary windings and secondary windings in order to prevent operating the electromagnetic couplings (18, 18', 18") in a state of magnetic saturation.

14. The power converter (7) according to any one of the preceding claims, further comprising an auxiliary side assembly (27), the auxiliary side assembly (lc) comprising at least two auxiliary side channels (33a-c), each arranged between two connectors (12a-b) of an auxiliary high voltage DC connection, wherein each one of the auxiliary side channels (33a-c) comprises: an upper converter arm (3a-c), a lower converter arm (3d-f) and an auxiliary winding (24a-c) arranged between the upper converter arm (3a-c) and the lower converter arm (3d-f), such that the upper and lower converter arms (3a-f) of the auxiliary side channels (30a-b) are arranged to minimise a magnitude of a DC current through the respective auxiliary windings, and each auxiliary winding respectively forming part of the at least two electromagnetic couplings (18, 18', 18") such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and an auxiliary winding.

15. The power converter (7) according to any one of the preceding claims, further comprising a third side assembly (lc) of the same structure as the second side assembly, the third side assembly (lc) comprising at least two third side channels, each comprising a tertiary winding forming part of the at least two electromagnetic couplings (18', 18") such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and a tertiary winding.

16. The power converter (7) according to any one of the preceding claims, further comprising an alternating current, AC, side assembly (9) comprising a plurality of AC tertiary windings (8a-b), corresponding in numbers with the number of electromagnetic couplings, the AC tertiary windings (8a-b) respectively forming part of the at least two electromagnetic couplings, such that each one of the at least two electromagnetic couplings comprises a primary winding, a secondary winding and an AC tertiary winding.

17. The power converter (7) according to any one of claims 14 to 16, the preceding claims wherein the second side assembly is used only to minimise a value associated with the electromagnetic field distribution in the electromagnetic couplings (18, 18', 18") in order to prevent magnetic saturation from occurring in said electromagnetic couplings.

18. The power converter (7) according to any one of claims 17, the preceding claims wherein the value associated with the electromagnetic field distribution in the electromagnetic couplings is used to prevent magnetic saturation from occurring in said electromagnetic couplings.

19. The power converter (7) according to any one of claims 1 to 13, further comprising a third side assembly (lc) of the same structure as the second side assembly, wherein each one of the at least two electromagnetic couplings (18) is a two sided electromagnetic coupling comprising a primary winding and a secondary winding selected from one of the second side assembly (lb) and the third side assembly (lc').

20. The power converter (7) according to any one of the preceding claims, wherein each one of the electromagnetic couplings forms part of a transformer.

21. The power converter (7) according to any one of the preceding claims, wherein the first side assembly (la) comprises three first side channels (30a-c), the second side assembly comprises three second side channels (31a-c) and wherein three separate electromagnetic couplings (18, 18', 18") are formed.