WO2020233803A1

WO2020233803A1 - Modular cascaded charge-pump converter

Info

Publication number: WO2020233803A1
Application number: PCT/EP2019/063134
Authority: WO
Inventors: Mikael Davidsson; Nan Chen; Kalle ILVES
Original assignee: Abb Schweiz Ag
Priority date: 2019-05-21
Filing date: 2019-05-21
Publication date: 2020-11-26
Also published as: EP3973626A1

Abstract

A converter (10) converting between a first voltage (V1) and a second voltage (V2) has a first connection port (CP1) for the first voltage (V1), a second connection port (CP2) for the second voltage and a stack of cells (CE1, CE2, CE3, CE4), where each cell comprises an energy storage element (C1, C2, C3, C4). In the converting the cells are configured to sequentially move a charge between the energy storage elements from one end of the stack to another end of the stack, the sequential moving of a charge comprising a cell being configured to move the charge between itself and a neighbour cell.

Description

MODULAR CASCADED CHARGE-PUMP CONVERTER

FIELD OF INVENTION The present invention generally relates to converters. More particularly the present invention relates to a converter for converting between a first and a second voltage.

BACKGROUND

A conventional high voltage direct current (HVDC) converter station has high-voltage transformers that steps-up/down an AC voltage to/from the voltage of a DC-bus. A converter valve is then used to generate a DC voltage in the transmitting end and an AC voltage on the receiving end.

For solar power plants an inverter is used to generate an AC voltage which is then stepped up to a higher voltage using a conventional transformer.

Transforming a low voltage to high voltage is usually done with a transformer. For good efficiency the coils of the transformer must be tightly coupled to one another via a magnetic circuit in the form of a magnetic core and use large core and coil wires. It is also not possible to convert a DC voltage using a transformer. The requirement of tight coupling gives challenging isolation problems when the transfer ratio is high (high output/input voltage).

One solution for transforming the voltage of e.g. solar plants is to use cascaded high-frequency DC/DC converters but the high frequency generates higher magnetic losses. However, the isolation of the top most transformers need to be dimensioned for the full DC voltage. It is therefore of interest to obtain a converter that does not require a transformer.

WO 93/20610 discloses a transformerless power conversion system that includes a plurality of series connected capacitors, a load circuit connected to a plurality of capacitors, where the load circuit charges a plurality of capacitors from a voltage source at a predetermined voltage. The system also includes a circuit that reverses the polarity of the accumulated charge in selected ones of the plurality of capacitors. The polarity reversal circuit comprises a plurality of inductance circuits that can be coupled by switching to a corresponding capacitor different from the selected capacitors, to form a resonant circuit for reversing the polarity of the charge accumulated in that capacitor. There is also a discharge circuit which extracts power from the plurality of capacitors to a transformed voltage.

WO 96/16468 discloses a DC-to-DC voltage convertor that is made up of a capacitor array having plural capacitor elements and a plurality of switches which are switchable between at least two states. When the switches are switched in the first state, the capacitor elements are connected in series, and when the switches are connected in the second state, the capacitor elements are connected in parallel. The DC-to-DC voltage convertor may be configured as a step-down converter or a step-up converter. Another type of converter is the Marx DC-to-DC converter, which is for instance described by Etienne Veilleux in a thesis named”DC Power Flow Controller and Marx DC-DC Converter for Multiterminal HVDC System”, Department of Electrical & Computer Engineering, McGill University, Montreal, Canada, 2013. However, there is still room for improvement in the realization of a converter that does not use a transformer, especially with regard to the voltage stress on a connection port of such a converter. The invention addresses at least some of the above-mentioned problems.

SUMMARY OF THE INVENTION

The present invention is directed towards solving at least some of the above-mentioned problems.

This is according to a first aspect achieved through a converter converting between a first voltage and a second voltage, the converter having a first connection port for the first voltage, a second connection port for the second voltage and a stack of cells, where each cell comprises an energy storage element. During the converting, the cells are configured to sequentially move a charge between the energy storage elements from one end of the stack to another end of the stack, where the sequential moving of a charge comprises a cell being configured to move the charge between itself and a neighbour cell.

This is according to a second aspect achieved through a method of converting between a first voltage on a first connection port and a second voltage on a second connection port of a converter comprising a stack of cells, each cell comprising an energy storage element, the method comprising

sequentially moving a charge between the energy storage elements from one end of the stack to another end of the stack, the sequential moving of a charge comprising a cell moving the charge between itself and a neighbour cell in the stack.

This is according to a third aspect achieved by a computer program product for controlling conversion between a first voltage on a first connection port and a second voltage on a second connection port of a converter comprising a stack of cells, each cell comprising an energy storage element, the computer program product comprising a data carrier with computer program code configured to cause a control unit to, when the computer program code is loaded into the control unit

control the converter to sequentially move a charge between the energy storage elements from one end of the stack to another end of the stack, the control of the converter to sequentially move the charge comprises controlling a cell to move the charge between itself and a neighbour cell in the stack.

Each cell may additionally comprise an inductor.

According to a first variation of the first aspect, a cell, when moving the charge between itself and a neighbour cell in the stack, is configured to form an LC circuit comprising the own inductor and own energy storage element together with the energy storage element of the neighbor cell

According to a corresponding variation of the second aspect, the moving of the charge from a cell to a neighbour cell comprises the cell forming an LC circuit comprising the own inductor and own energy storage element together with the energy storage element of the neighbour cell.

According to a second variation of the first aspect, a cell, when moving the charge between the cells, is configured to connect a first charging branch comprising the own energy storage element in parallel with a second charging branch comprising the energy storage element of the neighbour cell and to remove the parallel connection when the charge has been transferred from one of the energy storage elements to the other.

According to a corresponding variation of the second aspect, the moving of the charge from a cell to a neighbour cell comprises connecting a first charging branch comprising the own energy storage element in parallel with a second charging branch comprising the energy storage element of the neighbour cell and removing the parallel connection when the charge has been transferred. According to a third variation of the first aspect, the cell is further controllable to connect the own energy storage element in series with the energy storage element of the neighbour cell after the charge has been moved. According to a corresponding variation of the second aspect the method further comprises connecting the own energy storage element of the cell in series with the energy storage element of the neighbour cell after the charge has been moved. It is possible that the own energy storage element of the cell is connected in series with the energy storage element of the neighbour cell before the charge is transferred.

According to a fourth variation of the first aspect, the cell may be further controllable to remove the series connection when connecting the charging branches in parallel with each other.

According to a corresponding variation of the second aspect, the method further comprises removing the series connection when connecting the charging branches in parallel with each other.

The first converter connection port may be provided at one end of the stack, while the second converter connection port may be provided across the whole of the stack.

A cell may comprise a first and a second semiconductive element, where a first end of the own energy storage element is connected to a first end of the energy storage element of the neighbour cell via a first interconnecting branch comprising the first semiconductive element and a second end of the own energy storage element is connected to a second end of the energy storage element of the neighbour cell via a second interconnecting branch comprising the second semiconductive element in series with the own inductor. A least one of the first and second semiconductive element is furthermore a controllable semiconductive element, such as a controllable unidirectional conducting element or a controllable bidirectional conducting element. The first semiconductive element may be an uncontrollable unidirectional semiconductive element and the second semiconductive element may be a controllable semiconductive element or vice versa.

A cell may furthermore comprise a third interconnecting branch connected between the second end of the own energy storage element and the first end of the energy storage element of the neighbour cell, where the third interconnecting branch comprises a third semiconductive element, which may be a non-controllable semiconductive element such as a non- controllable unidirectional conducting element, or a controllable semiconductive element, such as controllable unidirectional conducting element or a controllable bidirectional conducting element.

A cell may additionally comprise a fourth interconnecting branch connected between the first end of the own energy storage element and second end of the energy storage element of the neighbour cell, where the fourth interconnecting branch may comprise a fourth semiconductive element that may be a controllable semiconductive element, such as a controlled unidirectional and bidirectional conducting element.

The energy storage elements may be capacitors.

For a cell at a position k from the top of the stack, where k ranges between one and n, the capacitance of the capacitor of this cell may be k times the capacitance C of the capacitor at the cell at the top of the stack. In case the cells comprise inductors, the inductance of the inductor of the cell at position k may then be the inductance of the inductor of the cell at the top divided by k.

The converter may additionally comprise a number of sections, where the cells in the stack are provided in different sections and the sections have a different number of parallel cells. The number of parallel cells in the sections may for instance vary by one cell from one section to the next from one end in the stack to the other, such as from the top of the stack towards the bottom.

The converter may further comprise a control unit controlling the cells. The invention has a number of advantages. Conversion is achieved in one step without the use of a transformer. The converter is also power and cost efficient and the voltage stress on the output is low.

It should be emphasized that the term“comprises/comprising” when used in this specification is taken to specify the presence of stated features, integers, steps or components, but does not preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will in the following be described with reference being made to the accompanying drawings, where fig. l schematically shows a first realization of a converter having a first and a second converter connection port and a stack of cells of a first type, fig. 2 schematically shows a second realization of a converter having a first and a second converter connection port and a stack of cells of a second type,

fig. 3 schematically shows a third realization of a converter having a first and a second converter connection port and a stack of cells of a third type, fig. 4 schematically shows the structure of the cell according to the first type comprising an energy storage element, an inductor as well as a first, second and third semiconductive element,

fig. 5 schematically shows an LC circuit created through the cell of fig. 4 and an energy storage element of a neighbour cell,

fig. 6 schematically shows a current and voltage of the cell in fig. 4 as well as a charge being transferred between this cell and a neighbour cell, fig. 7 shows a method step of operating the converter according to the first realization,

fig. 8 schematically shows a number of steps for moving a charge between a controlled cell and its neighbour,

fig. 9a schematically shows the charging of an energy storage element of a first cell in the stack of the first converter realization,

fig. 9b schematically shows moving of a charge from the first to the second cell in the first converter realization,

fig. 10a schematically shows currents circulating in the stack when moving the charge together with the output current from the converter of the first converter realization,

fig. 10b schematically shows a relationship between the values of the inductors and energy storage elements in the cells of the first type of converter realization,

fig. 11 schematically shows a variation of the first converter realization for providing negative output voltages and where the first and second converter connection ports are placed at the opposite ends of the stack compared with fig. 1,

fig. 12 shows a fourth type of converter realization comprising cells of a fourth type, and fig. 13 schematically shows a computer program product in the form of a CD ROM disc with a computer program code performing the functionality of the control units controlling the converter. DETAILED DESCRIPTION OF THE INVENTION

In the following, a detailed description of preferred embodiments of the invention will be given. Fig. 1 shows a first realization of a converter 10 for converting between a first voltage Vi and a second voltage V2. The converter 10 comprises a stack of n cells CEi, CE2, CE3 and CE4 here exemplified by four cells, i.e. where n = 4. Each cell comprises an energy storage element Cl, C2, C3, C4. Each cell may additionally comprise an inductor Li, L2, L3 and L4. The cells are connected in cascade. The converter 10 comprises a first converter connection port CPi having a first and a second converter connection terminal Ti and T2 and a second converter connection port CP2 having a third and a fourth converter connection terminal T3 and T4. The first converter connection port CPi has a first voltage Vi between the first and second converter connection terminals Ti and T2 while the second converter connection port CP2 has a second voltage V2 between the third and fourth converter connection terminals T3 and T4. The voltages may with advantage be direct current (DC) voltages. The first converter connection port CPi may additionally be provided at one end of the stack, while the second converter connection port CP may be provided across the whole of the stack.

In fig. 1 there is a first connection port inductor LCPi and first connection port capacitor CCPi, where the first connection port inductor LCPi may be a smoothing inductor and the first connection port capacitor CCPi is also a first connection port energy storage element, i.e. an energy storage element associated with and connected to the first converter connection port CPi. A first positive end of the first converter connection port capacitor CCPi is connected to the first converter connection terminal Ti via the first converter connection port inductor LCPi, while a second negative end of the first converter connection port capacitor CCPi is directly connected to the second converter terminal T2 of the first converter connection port CPi. Thereby the first converter connection port capacitor CCPi is connected in series with the first converter connection port inductor LCPi between the first and second converter connection terminals Ti and T2 of the first converter connection port CPi. The fourth converter connection terminal T4 of the second converter connection port CP2 is in this converter realization joined with the second converter connection terminal T2 of the first converter connection port CP2. Thereby the second end of the first converter connection port capacitor CCTi is also connected to the fourth converter connection terminal T4. As was mentioned above, the converter 10 comprises a stack of cascaded cells, where each cell is according to a first type, which is generally depicted in fig. 4. A cell Cn which may be an nth cell in the stack may comprise a first cell connection port CEPi and a second cell connection port CEP2, where the first cell connection port CEPi has a first cell connection terminal TCEi and a second cell connection terminal TCE2 and the second cell connection port CEP2 has a third cell connection terminal TCE3 and a fourth cell connection terminal TCE4. Between the first cell connection terminal TCEi and the third cell connection terminal TCE3 there is connected a first interconnecting branch IBi comprising a first semiconductive element. There is also a second interconnecting branch

IB2 connected between the second cell connection terminal TCE2 and the fourth cell connection terminal TCE4. The second interconnecting branch IB2 comprises a cell inductor Ln and a second semiconductive element. At least one of the first and second semiconductive element is a controllable semiconductive element. Here the first semiconductive element Dn is an uncontrollable unidirectional semiconductive element in the form of a diode and the second semiconductive element Sni is a first switch Sni that may be a controllable semiconductive element, such as a controllable unidirectional conducting element like a thyristor or a controllable bidirectional conducting element like a transistor. In this specific realization the first semiconductive element Dn is a diode having an anode connected to the first cell connection terminal TCEi and a cathode connected to the third cell connection terminal TCE3, while the first switch Sni is realized as a thyristor. The anode of the thyristor is in this case connected to the fourth cell connection terminal TCE4, while the cathode is connected to the second cell connection terminal TCE2. The connection of the first switch Sni to one of the cell connection terminals is also made via the inductor Ln and in the example given here the first switch Sni is connected to the second cell connection terminal TCE2 via the cell inductor Ln.

There is also an energy storage element in the cell, in this case realized as a cell capacitor Cn, where a first positive end of the cell capacitor Cn is connected to the third cell connection terminal TCE3 and a second negative end of the cell capacitor Cn is connected to the fourth cell connection terminal TCE4. Finally, there is third interconnecting branch IB3 interconnecting the first and fourth cell connection terminals TCEi and TCE4, which third interconnecting branch IB3 also comprises a third semiconductive element, which may be a non-controllable unidirectional conducting element or a controllable semiconductive element, such as a controllable unidirectional conducting element like a thyristor or a controllable bidirectional conducting element like a transistor. In this first realization it is realized as a controllable semiconductor device in the form of a second switch Sn2, for instance using a transistor such as an

Integrated-Gate Bipolar Transistor (IGBT).

Furthermore, a similar neighbour cell may be connected with its third cell connection terminal to the first cell connection terminal TCEi of the cell

CEn and with its fourth cell connection terminal to the second cell connection terminal TCE2 of the cell CEn. Thereby the first end of the own energy storage element Cn of the cell is connected to a first end of the energy storage element of the neighbour cell via the first interconnecting branch IBi that comprises the first semiconductive element Dn and the second end of the own energy storage element Cn is connected to a second end of the energy storage element of the neighbour cell via the second interconnecting branch IB2 that comprises the second semiconductive element Sni in series with the own inductor Ln. In a similar manner the third interconnecting branch IB3 is connected between the second end of the own energy storage element Cn and the first end of the energy storage element of the neighbour cell.

If again looking at fig. 1 it can be seen that the first cell connection terminal of a first cell CEi is connected to a junction between the first converter connection port capacitor CCPi and the first converter connection port inductor LCPi, while the second cell connection terminal of the first cell CEi is connected to the first and fourth converter connection terminals T2 and T4.

The third cell connection terminal of the first cell CEi is further connected to the first cell connection terminal of a second cell CE2, while the fourth cell connection terminal of the first cell CEi is connected to the second cell connection terminal of the second cell CE2.

The third and fourth cell connection terminals of the second cell CE2 are in the same way connected to the first and second cell connection terminals of a third cell CE3, which in turn has third and fourth cell connection terminals connected to first and second cell connection terminals of a fourth cell CE4.

In this example the fourth cell is the last cell in the stack and the third cell connection terminal of this fourth cell CE4 is connected to the third converter connection terminal T3 of the second converter connection port CP2 via a second converter connection port inductor LCP2, which may be a smoothing inductor. Through the above-described cell realization, the first, second and third semiconductive elements of the first cell CEi are a diode Di and a first and second switch Sn and S12, the first, second and third semiconductive elements of the second cell CE2 are a diode D2 and a first and second switch S21 and S22, the first, second and third semiconductive elements of the third cell are a diode D3 and a first and second switch S31 and S32 and the first, second and third semiconductive elements of the fourth cell are a diode D4 and a first and second switch S41 and S42.

In fig. 1 also gate control units used for turning on the switches of the cells are shown.

The converter in fig. 1 may be functioning as step up converter in which the first voltage Vi is input to the first converter connection port CPi and the second voltage V2 is obtained at the second converter connection port CP2, where the second voltage V2 is higher than the first voltage Vi.

Fig. 2 schematically shows a second converter realization with a similar structure that may be used as a unidirectional DC/DC step-down converter, where an input voltage V2 is applied at the second converter connection port CP2 and an output voltage Vi is obtained at the first converter connection port CPi. In this converter, which as an example comprises five cells, the second converter connection terminal T2 is used as a connection terminal of both the first and second converter connection ports.

It can also be seen that the first semiconductive element in the first interconnecting branch IBi in this case is realized as a switch, here in the form of a controllable unidirectional conducting element, in this case exemplified by a thyristor with the anode pointing upwards in the stack and the cathode pointing downwards in the stack towards the first converter connection port CPi. The second semiconductive element in the second interconnecting branch is in turn a first non-controllable unidirectional conducting element, here in the form of a diode with the cathode connected to the second end of the cell capacitor and the anode connected to the cell inductor. It can also be seen that the third

semiconductive element in the third interconnecting branch has the same realization as a second non-controllable unidirectional conducting element, here in the form of a diode, with the anode connected to the fourth cell connection terminal and the cathode connected to the first cell connection terminal.

Thereby the first cell has a switch Si and a first and second diode Dn and Di2 as the first, second and third semiconductive element, the second cell has a switch S2 and a first and second diode D21 and D22 and as the first, second and third semiconductive element, the third cell has a switch S3 and a first and second diode D31 and D32 as the first, second and third semiconductive element, the fourth cell has a switch S4 and a first and second diode D41 and D42 as the first, second and third semiconductive element and the fifth cell has a switch S5 and a first and second diode D51 and D52 as the first, second and third semiconductive element.

There is also a control unit (CU) 12 controlling the cells. The control unit 12 provides control signals to the switches Si, S2, S3, S4 and S5 via their gate units. These control signals are in the gate units adapted to a voltage level required by the switches.

Fig. 3 schematically shows a third realization of the converter 10 of similar structure as both the two previous converter realizations. This third converter realization may be used both as a step up and a step-down converter, where in each case each semiconductive element is realized as a switch. The first and second semiconductive elements may in this case both be realized as first and second switches, each in the form of an anti parallel thyristor pair, while the third semiconductive switch may be realized as a transistor with anti-parallel diode. It is possible that also the third semiconductive element is realized as an anti-parallel thyristor pair. The first and second semiconductive elements may also be realized as transistors with anti-parallel diodes. Thereby the first cell has a first, second and third switch Sn, S12 and S13, the second cell has a first, second and third switch S21, S22 and S23, the third cell has a first, second and third switch S31, S32 and S33, the fourth cell has a first, second and third switch S41, S42 and S43 and the fifth cell has a first, second and third switch S51, S52 and S53 as the first, second and third semiconductive element.

In the examples given above switches were realized as single or pairs of anti-parallel thyristors or as IGBTs with or without anti-parallel diodes. It should be realized that other types of switches are possible, such as Bimode Insulated-Gate Transistors (BiGTs) and Reverse Conducting (RC)- IGBTs.

Moreover, although only the second converter realization in fig. 2 is shown as having a control unit, it should be realized that also the other converter realizations described herein may be equipped with such a control unit.

The basic idea behind the operation of the converter will now be described with reference being made to fig. 4, 5 and 6, where fig. 5 schematically shows an LC circuit formed by a controlled cell, i.e. a cell having its switches being operated, and the energy storage element of a neighboring cell and fig. 6 shows the current circulating in the LC circuit and a charge being transmitted to the energy storage element of the controlled cell from the energy storage element of the neighbour cell. A voltage is applied on one of the converter connection ports at one end of the stack and the resulting current is used to charge the cell at this end of the stack. Thereafter a sequential moving of a charge is made between the energy storage elements of the cells from one end of the stack to another end of the stack. There is thus a sequential moving of a charge from the above-mentioned end of the stack to the other end of the stack in order to provide an output voltage on the other converter connection port. When moving a charge from one cell to another, a cell CEn is controlled to cause the move to be made in relation to a neighbour cell. In the sequential moving of a charge a cell is thereby configured to move the charge between itself and a neighbour cell. The move may more particularly involve connecting a first charging branch CBi in parallel with a second charging branch CB2, where the first charging branch CBi comprises the energy storage element or capacitor Cn of the controlled cell CEn and the second charging branch comprises the energy storage element or capacitor Cn-i of the neighbour cell. One of the charging branches may also comprise the inductor Ln of the controlled cell. In this way the two branches may form an LC circuit. In the example in fig.5, the capacitor Cn and diode Dn of the controlled cell CEn are placed in the first charging branch CBi, while the inductor Ln and first switch Sni of the controlled cell CEn and capacitor Cn-i of the neighbour cello are placed in the second charging branch CB2.

When moving a charge between itself and a neighbour cell the controlled cell may thus be configured to form an LC circuit comprising the own inductor Ln and own energy storage element Cn together with the energy storage element Cn-i of the neighbour cell. It can additionally be seen that the controlled cell may be configured to connect a first charging branch comprising the own energy storage element in parallel with a second charging branch comprising the energy storage element Cn-i of the neighbour cell. The first or the second charging branch may additionally comprise the own inductor Ln.

The capacitor delivering the charge has a voltage that is higher than the voltage of the capacitor receiving the charge. If for instance a capacitor in the cell at a lower end of the stack is to transfer a charge to the capacitor of a controlled cell immediately above the cell with the charge delivering capacitor, then the first switch Sni of the controlled cell is turned on, which causes the charge delivering capacitor Cn-i of the neighboring cell and the inductor Ln of the controlled cell CEn to be connected in parallel with the capacitor Cn of the controlled cell CEn. There is thereby also formed an LC circuit comprising the two capacitors Cn and Cn-i and the inductor Ln. Due to the voltage Un-i of the charge delivering cell capacitor Cn-i being higher than the voltage Un of the charge receiving cell capacitor Cn, the diode Dn in the controlled cell CEn will be conducting. A current I is thereby delivered to the charge receiving capacitor Cn and thereby the voltage Un of this capacitor Cn increases. At the same time the voltage across the charge delivering capacitor will sink. However, the rate of change of the voltage may differ based on the capacitor sizes. The current I stops increasing when the voltages have become equal and the inductor Ln is then charged to its maximum. The current continues to flow until it reaches zero by which the diode Dn blocks any current return. Thereby a charge Q is moved from the charge delivering capacitor Cn-i to the charge receiving capacitor Cn. Through the use of the diode Dn, the controlled cell CEn may additionally be configured to remove the parallel connection when the charge has been transferred from one of the capacitors to the other. After the move of the charge the cell may further be controllable, via the second switch Sn2, to connect the own energy storage element Cn in series with the energy storage element of the neighbour cell.

The above described operation is from the bottom of the stack to the top. It should be realized that as an alternative it is possible with a transfer in the opposite direction. In this case a charge is transferred from the capacitor of the controlled cell to the capacitor of its neighbor below it. The capacitor of the controlled cell will in this case be a charge delivering capacitor, while the capacitor in the lower neighbor cell will be a charge receiving capacitor. How operation of the converter 10 according to the first converter realization may be carried out will now be described with reference also being made to fig. 7, which shows a general method step of converter operation involving sequential movement of a charge from the bottom to the top of the stack of cells, to fig. 8 which shows method steps performed when moving a charge between a controlled cell and its neighbor and to fig. 9a and 9b that schematically show the charging of the first cell and the movement of a charge Q from the first to the second cell in the converter according to the first realization.

In the first converter realization, the first converter connection port CPi is an input port with an applied input voltage Vi, which as an example is a DC voltage, and the second converter connection port CP2 is an output port providing an output voltage V2 that may also be a DC voltage.

Thereby the first and second converter connection terminals Ti and T2 are input terminals of the converter 10 and the third and fourth converter connection terminals T3 and T4 are output terminals of the converter 10. Thereby the converter 10 is also a step-up converter. In this case the first converter connection port capacitor CCPi may be larger than the capacitor Cl of the first cell CEi, the capacitor Cl of the first cell CEi be larger than the capacitor C2 of the second cell CE2, the capacitor C2 of the second cell CE2 be larger than the capacitor C3 of the third cell CE3 and the capacitor C3 of the third cell may be larger than the capacitor C4 of the fourth cell CE4.

Before the operation is started it is possible that the capacitors are series connected. The second switch S12 of the first cell CEi may thus

interconnect the second end of the capacitor Cl in the first cell with the first end of the first converter connection port capacitor CCPi, the second switch S22 of the second cell CE2 may interconnect the second end of the second cell capacitor C2 with the first end of the first cell capacitor Cl, the second switch S32 of the third cell CE3 may interconnect the second end of the third cell capacitor C3 with the first end of the second cell capacitor C2 and the second switch S42 of the fourth cell CE4 may interconnect the second end of the fourth cell capacitor C4 with the first end of the third cell capacitor C3. Moreover, the first converter connection port capacitor CCPi is charged by the input voltage Vi, which input voltage can thereby be seen as being provided by a DC voltage source.

The control of the converter involves sequentially moving a charge Q from one end of the stack to the other, step 14.

The movement may start with charging of the capacitor Cl of the first cell CEi. This may be done through the control unit turning off the second switch S12 and turning on the first switch S11 of the first cell CEi. As can be seen in fig. 9a, this leads to the DC source forming a charging loop with the capacitor Cl and inductor Li of the first cell CEi. Another way that this could be looked at is as if a charge is moved from the first converter connection port capacitor CCPi to the first cell capacitor Cl. The DC source applies a voltage across the first converter connection port capacitor CCPi, which capacitor is then connected in parallel with the first cell capacitor Cl and inductor Li, thereby the first connection port capacitor CCPi forms an LC circuit together with the inductor Li and capacitor Cl of the first cell CEi used to move a charge from the first converter connection port capacitor CCPi to the first cell capacitor Cl. When the capacitor Cl of the first cell CEi has been charged with a charge Q, the second switch S12 of the first cell CEi is then turned on so that the capacitor Cl is connected in series with the first converter connection port capacitor CCPi.

The sequential movement of the charge Q from one end of the stack to the other involves moving the charge between a controlled cell and its neighbor. When the charge is to be moved from the first cell CEi, the second cell CE2 is the controlled cell. Before the charge is transferred from the first cell to the controlled second cell CE2, the own capacitor C2 of the second controlled cell CE2 may be connected in series with the capacitor of the neighbour cell, i.e. with the capacitor Cl of the first cell CEi. The cell CE2 is further controllable to remove the series connection when connecting the capacitors in parallel with each other. A charge Q is thus moved between a cell and its neighbor, step 16-. For the movement from the first cell CEi to the second cell CE2, this may more particularly involve the control unit removing the series connection between the capacitor C2 of the controlled second cell CE2 and the capacitor Cl of its neighbor CEi, step 16a, and forming an LC circuit through connecting the first charging branch comprising the capacitor C2 of the controlled second cell CE2 in parallel with the second charging branch comprising the capacitor Cl of the first neighbor cell CEi, step 16b. As one of the branches comprises the inductor L2 there is thus formed an LC circuit comprising the two capacitors Cl and C2 and the inductor Li. Both the removing of the series connection and the forming of the LC circuit may be achieved through the control unit turning on the first switch S21 of the second cell CE2. The turning on of the first switch S21 will achieve the connecting of the first charging branch in parallel with the second charging branch, step 16b, and thereby a current will run from the first end of the first cell capacitor Cl towards the first end of the second cell capacitor C2. Thereby a charge Q is moved from the capacitor Cl of the first cell CEi to the capacitor C2 of the second cell CE2. Once the charge Q has been moved, the parallel connection of the two charging branches is then removed, step 16c, which may be automatically achieved by the diode

D2 as well as the first switch S21 stopping to conduct.

This may then be followed by connecting the capacitor C2 of the second cell CE2 in series with the capacitor Cl of the first cell CEi, step i6d, which may be achieved through the control unit turning on the second switch S22 of the second cell CE2. This procedure may then be followed sequentially for each cell in the stack until the capacitor C4 of the last cell has been charged, which leads to a stepped -p output voltage V2 being delivered on the output port CP2. This can also be expressed in the following way. For a turned on first switching element in a current cell k, where k ranges from 1 to n-i, a charge is transferred between the capacitor of the current cell and the capacitor of the neighboring lower cell, the charge is thereby transferred to or from the capacitor of the lower cell.

With this type of control one cell is always charging and the output voltage is hence the sum of the voltage of at least some of the other cell capacitors. The output voltage is thus a sum of voltages of cells that are not controlled to move the charge.

This type of operation has a number of advantages. Conversion is achieved in one step without the use of a transformer. The converter is also power and cost efficient and the voltage stress on the output is low. Through using an inductor in the cells, the switching losses and Electromagnetic interference (EMI) may also be lowered.

The invention has a number of further advantages. It is possible to convert a DC voltage with very high conversion ratio and provide mega-high- voltage DC transmissions. The converter may also become short-circuit proof. It further eliminates the need for expensive, and difficult to design, high-voltage transformers. It is also a good fit for connecting solar power plants to future MV/HV transmission grids. The converter can also be used for high-ratio voltage conversion of Off-shore wind power and can thereby interconnect an off-shore wind farm with an HVDC collection grid. The converter further allows parallel and interleaved operation for higher, smoother and redundant output power. The resonant type of switching employed has a number of further advantages: It may provide a Sinusoidal current with a limited dl/dt.

Fig. 10a and 10b schematically show the current sizes and capacitor and inductor sizes of the first converter realization.

Because of conservation of energy, the average input current is n x Iout,avg were n is the number of levels in the stack and Iout,avg is the average output current. The average current circulating between the capacitor of a cell and the capacitor of its neighbor will therefore be very high, where the average current at the highest level, the fourth level, may be two times the average output current, the average circulating current at the second highest level, the third level, is three times the average output current, the average circulating current at the second lowest level, the second level is four times the average output current and the average circulating current at the lowest level, the first level is five times the average output current. Note that the second switch in each cell is hard switched but only see the average output current Iout,avg. Put differently, the average output current (discharge current) of the top cell capacitor C4 is equal to the converter output current.

In order to keep the top capacitor voltage constant, the third cell CE3 must charge this capacitor C4 with an average current of the same magnitude. It will also need to supply the average converter output current since this current always flow.

The average output current of the cell capacitor of the third cell CE3 is hence twice the converter output current. This reasoning is true also for the succeeding lower cells in the cascaded converter.

Since one cell is always charging the input current to the converter is the product of the output current and the number of cells. The current capability (both conduction and switching) of a cell hence increases by a factor of one for each cell as going down from the top to the bottom of the converter.

In order to handle these larger currents, it is possible to connect several cells in parallel on a level or to provide larger sized components. This is schematically indicated in fig. 10b. As was mentioned earlier, the first converter connection port capacitor

CCPi may be larger than the capacitor Cl of the first cell CEi, the capacitor Ci of the first cell CEi be larger than the capacitor C2 of the second cell CE2, the capacitor C2 of the second cell CE2 may be larger than the capacitor C3 of the third cell CE3 and the capacitor C3 of the third cell may be larger than the capacitor C4 of the fourth cell CE4. The fourth cell capacitor C4 may for instance have a capacitance of C, the third cell capacitor C3 a capacitance of 2*C, the second cell capacitor C2 a capacitance of 3*C and the first cell capacitor Ci a capacitance of 4*C and the first connection port capacitor CCPi a capacitance of 5*C. It is at same time possible that the inductance of the fourth cell inductor L4 is L, that the inductance of the third cell inductor L3 is L/2, that the inductance of the second cell inductor L2 is L/3 and that the inductance of the first cell inductor Li is L/4. Put differently, for a cell at a position k starting from the top of the stack, where k ranges between 1 and n, the capacitance of the capacitor is k times the capacitance C of the capacitor in the top cell and the inductance of the inductor in the cell is the inductance of the inductor of the top cell divided by k.

This is in essence the same as connecting four cells in parallel on the lowest level, three cells in parallel on the second lowest level, two cells in parallel on the next highest level and only one cell, the fourth cell CE4 on the highest level, where the components of the cells have equal size.

It hence makes sense to connect cells in parallel to achieve higher capability. This also gives a constant resonance frequency for all pair of capacitors. Due to the fact that one cell (and only one) charges at a time, the peak charge current must be much higher than the average charge current depicted in fig. 10a. An example: For a converter with 10 times voltage transformation the peak current of the bottom cell will be 10*10=100 times the output current.

At least one of the cells in the stack, and possibly also more cells, may thereby be connected in parallel with at least one further cell for paralleling and cascading purposes.

Thereby the converter may be seen as comprising a number of sections, where the cells in the stack are provided in different sections and the sections have a different number of parallel cells. The number of parallel cells in the sections may additionally vary by one cell from one section to the next from one end in the stack to the other, such as from the top of the stack towards the bottom

The first converter realization may be a unidirectional DC/DC step up converter, where the charging current is delivered from the first end of the capacitor in a neighbour cell to the first end of the capacitor of a controlled cell above it.

As can be seen in fig. 2 the second converter realization is a unidirectional step-down converter and in this case the charging current in a cell has the opposite direction compared with in the first converter realization. A charge is also moved from the capacitor of a controlled cell to the capacitor of the neighbour cell below it. The converter realization in fig. 3 is a bidirectional DC/DC step up/step down converter in which the charging current of an LC circuit may have any of the two described directions.

Fig. 11 shows a variation of the first converter realization to be used for negative voltages. In this case the first converter connection port may be placed at the top of the stack of cells, which is here to a fifth cell in the stack. The second converter connection port is here provided between the first connection terminal of the first cell and the fourth connection terminal of the fifth cell. A DC source may because of this be connected between the first and second converter port terminals, where the second converter port terminal is grounded and the first converter port terminal is placed at the first converter port inductor. This converter realization may be combined with the converter of the first realization in order to be connected to a symmetric HVDC link.

Fig. 12 shows a further converter realization comprising a stack of five cells, which converter 10 may be used for step down and step up operation for both positive and negative voltages. In this case each cell additionally comprises a fourth interconnecting branch, where the fourth

interconnecting branch is connected between the third and second cell connection terminals and the third and fourth interconnecting branches comprise third and fourth semiconducting elements that in this case are transistors, while the first and second interconnecting branches comprise first and second semiconducting elements realized as anti-parallel thyristor pairs. Current can flow in both directions through the converter.

In the first cell there is in this case a first thyristor switch S11 in the first interconnecting branch, a second thyristor switch S12 in the second interconnecting branch, a third transistor switch S13 in the third interconnecting branch and a fourth transistor switch S14 in the fourth interconnecting branch. In the second cell there is a first thyristor switch S21 in the first interconnecting branch, a second thyristor switch S22 in the second interconnecting branch, a third transistor switch S23 in the third interconnecting branch and a fourth transistor switch S24 in the fourth interconnecting branch. In the third cell there a first thyristor switch S31 in the first interconnecting branch, a second thyristor switch S32 in the second interconnecting branch, a third transistor switch S33 in the third interconnecting branch and a fourth transistor switch S34 in the fourth interconnecting branch. In the fourth cell there is a first thyristor switch S41 in the first interconnecting branch, a second thyristor switch S42 in the second interconnecting branch, a third transistor switch S43 in the third interconnecting branch and a fourth transistor switch S14 in the fourth interconnecting branch. In the fifth cell there is finally a first thyristor switch S51 in the first interconnecting branch, a second thyristor switch S52 in the second interconnecting branch, a third transistor switch S53 in the third interconnecting branch and a fourth transistor switch S54 in the fourth interconnecting branch.

There are a number of variations that are possible to make of the converter realizations. For instance, in the previously described control, the capacitors of the controlled cell and its neighbor where connected in series with each other after the transfer of the charge. It should be realized that it is possible to omit this step for a cell and instead keep the cell capacitor bypassed after having delivered or received a charge. This improves the capability of controlling the output voltage.

The control unit may be realized as a processor acting on computer program code. This computer program code may be provided with the processor in a dedicated circuit, such as a Field-programmable Gate Array (FPGA) or an Application Specific Integrated Circuit (ASIC). However, the processor may also act on computer program instructions in a program memory. The computer program instructions may additionally be provided on a data carrier, which performs the functionality of the control unit when loaded into such a memory. Fig. 13 schematically shows such a data carrier 18 in the form a CD ROM disk with such code 20. Other possible types of data carriers are memory sticks.

From the foregoing discussion it is evident that the present invention can be varied in a multitude of ways. It shall consequently be realized that the present invention is only to be limited by the following claims.

Claims

1. A converter (10) converting between a first voltage (V 1) and a second voltage (V2), the converter having a first connection port (CPi) for the first voltage (V 1), a second connection port (CP2) for the second voltage and a stack of k cells (CEi, CE2, CE3, CE4, CE5), where each cell comprises an energy storage element (Cl, C2, C3, C4, C5) and in the converting the cells are configured to sequentially move a charge (Q) between the energy storage elements from one end of the stack to another end of the stack, the sequential moving of a charge comprising a cell (Cen) being configured to move the charge (Q) between itself and a neighbour cell.

2. The converter (10) according to claim 1, wherein each cell comprises an inductor (Li, L2, L3, L4, L5) and a cell (CEn), when moving the charge between itself and a neighbor cell in the stack, is configured to form an LC circuit comprising the own inductor (Ln) and own energy storage element (Cn) together with the energy storage element (Cn-i) of the neighbor cell.

3. The converter according to claim 1 or 2, wherein a cell (Cn), when moving the charge between the cells, is configured to connect a first charging branch (CBi) comprising the own energy storage element in parallel with a second charging branch (CB2) comprising the energy storage element (Cn-i) of the neighbour cell and to remove the parallel connection when the charge (Q) has been transferred from one of the energy storage elements to the other.

4. The converter (10) according to claim 2, wherein the cell comprises a first and a second semiconductive element (Dn, Sni), where a first end of the own energy storage element (Cn) is connected to a first end of the energy storage element of the neighbour cell via a first

interconnecting branch (IBi) comprising the first semiconductive element (Dn) and a second end of the own energy storage element (Cn) is connected to a second end of the energy storage element of the neighbour cell via a second interconnecting branch (IB2) comprising the second semiconductive element (Sni) in series with the own inductor (Ln), where a least one of the first and second semiconductive element is a controllable semiconductive element.

5. The converter (10) according to any previous claim, wherein the first semiconductive element (Dn) is an uncontrollable unidirectional semiconductive element and the second semiconductive element (Sni) is a controllable semiconductive element or vice versa.

6. The converter (10) according to any previous claim, the cell (Cen) being further controllable to connect the own energy storage element (Cn) in series with the energy storage element of the neighbour cell after the charge has been moved.

7. The converter (10) according to claim 6, wherein the own energy storage element of the cell is connected in series with the energy storage element of the neighbour cell before the charge is transferred and the cell is further controllable to remove the series connection when connecting the charging branches in parallel with each other.

8. The converter (10) according to claim 7, the cell further comprising a third interconnecting branch (IB3) connected between the second end of the own energy storage element (Cn) and the first end of the energy storage element of the neighbour cell, the third interconnecting branch comprising a third semiconductive element (Sn2).

9. The converter (10) according to claim 7 or 8, the cell further comprising a fourth interconnecting branch connected between the first end of the own energy storage element and second end of the energy storage element of the neighbour cell, the fourth interconnecting branch comprising a fourth semiconductive element.

10. The converter (10) according to any previous claim, further comprising a control unit (12) controlling the cells.

11. The converter (10) according to any previous claim, wherein the converter comprises a number of sections, where the cells in the stack are provided in different sections, said sections having a different number of parallel cells.

12. A method of converting between a first voltage (Vi) on a first connection port (CPi) and a second voltage (V2) on a second connection port (CP2) of a converter comprising a stack of cells (CEi, CE2, CE3, CE4, CE5), each cell comprising an energy storage element (Cl, C2, C3, C5, C5) the method comprising

sequentially moving (14) a charge (Q) between the energy storage elements from one end of the stack to another end of the stack, the sequential moving of a charge comprising a cell (Cen) moving (16) the charge (Q) between itself and a neighbour cell in the stack.

13. The method according to claim 12, wherein each cell comprises an inductor and the moving of the charge from a cell to a neighbour cell comprises the cell forming an LC circuit comprising the own inductor (Ln) and own energy storage element (Cn) together with the energy storage element (Cn-i) of the neighbour cell.

14. The method according to claim 12 or 13, wherein the moving of the charge from a cell to a neighbour cell comprises connecting a first charging branch comprising the own energy storage element in parallel with a second charging branch (CB) comprising the energy storage element (Cn-i) of the neighbour cell and removing the parallel connection when the charge (Q) has been transferred.

15. A computer program product for controlling conversion between a first voltage (Vi) on a first connection port (CPi) and a second voltage (V2) on a second connection port (CP2) of a converter comprising a stack of cells (CEi, CE2, CE3, CE4, CE5), each cell comprising an energy storage element (Cl, C2, C3, C5, C5), said computer program product comprising a data carrier (18) with computer program code (20) configured to cause a control unit (12) to, when said computer program code is loaded into said control unit (12).