CONVERTER

Abstract

The present invention relates to a converter for transferring power between a first DC system of DC voltage V1 and a second DC system of DC voltage V2, the converter comprising: —a first AC/DC converter for transforming DC voltage V1 into a first single phase AC voltage V1ac, of frequency ω, root mean square line-neutral magnitude V1acm and angle α1; a second AC/DC converter for transforming DC voltage V2 into a second single phase AC voltage V2ac, of frequency ω, root mean square line-neutral magnitude V2acm and angle α2; and two inductors L1, L2 and a capacitor C, wherein the first terminals of the inductors and capacitor are connected together, the second terminal of inductor L1 and the second terminal of the capacitor C are connected to the first AC voltage V1ac, and the second terminal of inductor L2 and the second terminal of the capacitor C are connected to the second AC voltage V2ac; wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to enable required power transfer and to minimise current in inductor L1, and/or minimise current in inductor L2.

Description

The present invention relates to an electronic transformer for interfacing with DC transmission.

Historically, there has not been much use of DC-DC converters in the power range of tens and hundreds of MW, because of insufficient market need and the lack of suitable technology. However, the market demand for DC-DC connection has significantly increased in recent years as increasing numbers of power sources which generate DC are developed. DC power sources which approach power levels in the multiple MW range include: —fuel cells, photovoltaics, batteries and redox flow [1,2]. Additionally, variable speed machines such as permanent magnet wind generators or small hydro generators may be viewed as DC sources if the last converter stage is removed [3]. Furthermore, most electrical storage and load levelling devices use some form of DC storage media, such as, batteries, supercapacitors, capacitors, superconducting magnetic energy storage, etc). Many of these DC sources utilise very low voltage basic cell, or require wide variation of DC voltage. Consequently, their integration into the power grid has traditionally been difficult.

The rapid development of High Voltage DC (HVDC) transmission technologies is also driving demand for DC-DC converters. The recently developed HVDC light, ie HVDC with Voltage Source Converters (VSC HVDC), [4] has been implemented in many interconnections and shows many benefits compared with the traditional thyristor based HVDC. Virtually all existing HVDC schemes, both line commutated and VSC HVDC, operate as two-terminal systems. However, there is significant incentive to develop the technology to provide additional access points to the existing HVDC lines. A suitable MW size DC-DC converter would enable tapping on HVDC lines and aid development of multi-terminal HVDC. In particular, the recent development of offshore renewable sources creates the scenario of distributed DC sources with the requirement for submarine DC transmission and DC/DC voltage stepping at MW power levels.

A high-power DC transformer is needed to connect a DC transmission network with another DC network at a different DC voltage level, or to connect DC sources and loads.

DC-DC converters have been extensively utilized at low power levels, and many topologies exist. However, most low power converter technology is unsuitable for scaling up to MW power levels.

Conventional, unidirectional boost converters [5] can not achieve gains larger than 2-4, or higher powers, due to difficulties with the output diode and poor efficiency.

There have been attempts to develop converters with an internal AC transformer, for example flyback and forward converters [5-7], at higher power levels. However, some serious inherent limitations in terms of stepping ratios and power levels have been demonstrated. Reference [1] studies scaling up to 5 kW with a stepping ratio of 5, and [2] describes a 100 kW, 14 kV forward converter. However, these converters utilise MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistor) as switches with around 10 kHz frequency, which gives little prospect for further increasing to MW power levels, where IGBT (Insulated-Gate Bipolar Transistor) switches are required and lower frequency.

Parallel resonant converters can achieve high step-up gain [5]. The main limitations of these topologies are caused by increased switching losses, poor power quality, poor power factor and switch utilization, difficulties with power direction reversal and control difficulties.

Switched capacitor converters have been proposed as a method of achieving high DC boost without transformers or inductors [8]. However, each module only increases the output voltage by the value of the input voltage. Thus, to achieve a stepping ration of 10, for example, 9 modules are needed and over 18 switches. This results in significant losses and complexity.

A resonant LCLC circuit has been proposed for applications with high-frequency ballast lighting, which shows the capability of transformerless step-down operation [9]. However, such a converter is only suitable for driving a passive capacitive load (two LC circuits), and uses frequency control.

LCL-T resonant converters give improved performance at high power. However, an internal AC transformer is required to achieve voltage stepping [10]. The internal transformer increases weight and losses, and creates difficulties in the case of faults. Moreover, these converter topologies are normally used with one active bridge and one diode bridge, which precludes power reversal.

An LCL resonant converter which does not require a transformer has been proposed for an induction heating application [11]. However, this converter operates as an LCLR circuit and is only suitable for supplying a passive resistive load.

The topologies of [9,10,11] use a single active bridge and must employ a transformer for voltage stepping.

Most of the above topologies have destructive currents under DC fault conditions. DC faults are very important with high power systems because of difficulties in interrupting high fault currents.

A step-up DC-DC converter without internal voltage transformers has been proposed recently [12-15]. This converter can achieve very high step-up gains with a MW range test system, and bidirectional operation is possible. The switches on the low voltage side of this converter should be rated for P₁₂*n₁₂, where P₁₂ is the power transfer and n₁₂ is the stepping ratio (n₁₂=V₁/V₂). This represents a low switch utilisation factor with consequent disadvantages in terms of cost and increased losses. A further limiting factor of this converter is the low switching frequency caused by the use of thyristor switches, which have a long turn off time and introduce reverse recovery issues.

According to one aspect of the present invention there is provided a converter for transferring power between a first DC system of DC voltage V₁ and a second DC system of DC voltage V₂, the converter comprising: —

a first AC/DC converter for transforming DC voltage V₁ into a first single phase AC voltage V_1ac, of frequency ω, RMS line-neutral magnitude V_1acm and angle α₁;

a second AC/DC converter for transforming DC voltage V₂ into a second single phase AC voltage V_2ac, of frequency ω, RMS line-neutral magnitude V_2acm and angle α₂; and

two inductors L₁,L₂ and a capacitor C, wherein the first terminals of the inductors and capacitor are connected together, the second terminal of inductor L₁ and the second terminal of the capacitor C are connected to the first AC voltage V_1ac, and the second terminal of inductor L₂ and the second terminal of the capacitor C are connected to the second AC voltage V_2ac;

wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to enable required power transfer and to

minimise current in inductor L1,

and/or minimise current in inductor L2,

Preferably, for maximum power transfer in one direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially

$\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}$

and, for maximum power transfer in the opposite direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially

$180-\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}.$

Thus, with the present invention, power can be transferred between DC systems without the need for an internal transformer. Moreover, any stepping ratio at low losses can be achieved, unlike other LCL converters such as those disclosed in [11]. Moreover, the converter of the invention can use faster switches such as IGBT, with turn-off capability and no reverse blocking, which represents an improvement over thyristor switches as used in previous DC/DC transformerless converters such as those disclosed in [12, 13]. Furthermore, the converter of the invention is based on a voltage-source approach, whereas previous DC/DC transformerless converters such as those disclosed in [12, 13] are based on a current source approach. Additionally, the converter of the invention achieves excellent DC fault response, which is important for high-power applications. These advantages apply equally to the other aspects of the invention defined below.

According to a second aspect of the present invention there is provided a converter for transferring power between a first DC system of DC voltage V₁ and a second DC system of DC voltage V₂, the converter comprising: —

a first AC/DC converter for transforming DC voltage V₁ into a first p phase AC voltage V_1ac of frequency ω, RMS line-neutral magnitude V_1acm and angle α₁;

a second AC/DC converter for transforming DC voltage V₂ into a second p phase AC voltage V_2ac, of frequency ω, RMS line-neutral magnitude V_2acm and angle α₂; and

a resonant circuit for each phase p, wherein each resonant circuit comprises two inductors L₁,L₂ and a capacitor C connected together at their first terminals, the second terminal of inductor L₁ being connected to the respective phase of the first AC voltage V_1ac, and the second terminal of inductor L2 being connected to the respective phase of the second AC voltage V_2ac;

wherein the value of the capacitor C, inductor L₁ and inductor L₂ are selected to satisfy:

k ₁=1−ω²L₂C

k ₂=1−ω²L₁C

k ₃ =L ₁ +L ₂−ω²L₁L₂C

and where k₁ is selected in the region:

$-\frac{V_2}{V_1}<k_1<1$

and where

$k_2=k_1s^2$ $k_3=\frac{-p8V_1V_2\sqrt{1-{\left(V_1/V_2\right)}^2k_1^2}}{P_{1ac}\omega {\pi }^2}$

and where P₁₂ represents maximum power transfer by the converter.

Preferably, for maximum power transfer in one direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially

$\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}$

and, for maximum power transfer in the opposite direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially

$180-\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}.$

With the second aspect of the invention, the principles of the invention are applied to a p phase converter. The number of phases p may be any positive integer, preferably 1, 2 or 3. It will be appreciated that the case where p=1 corresponds to a single phase converter, and is equivalent to the first aspect of the invention.

According to a third aspect of the present invention there is provided a converter for transferring power between n first DC systems respectively of DC voltage V_1i (i=1, 2, . . . n) and m second DC systems respectively of DC voltage V_2j (j=1, 2, . . . m), where n and m are each any positive integer, the converter comprising: —

n AC/DC converters for respectively transforming DC voltages V_1i into respective p phase AC voltages V_1iac of frequency ω, RMS line-neutral magnitude V_1iaCm and angle α_1i;

• • m AC/DC converters for respectively transforming DC voltages V_2j into respective p phase AC voltages V_2jac, of frequency ω, RMS line-neutral magnitude V_2jacm and angle α_2j; and

a LCL circuit for each phase p, wherein each LCL circuit comprises n inductors L_1i, m inductors L_2j and a capacitor C connected together at their first terminals, the second terminal of each inductor L_1i being connected to the respective phase of the respective AC voltage V_1iac, and the second terminal of each inductor L_2j being connected to the respective phase of the respective AC voltage V_2jac;

wherein

wherein the value of the capacitor C, each inductor L_1i and each inductor L_2j are selected to enable required power transfer and to

minimise current in inductors L_1i,

and/or minimise currents in inductors L_2j,

Preferably, for maximum power transfer in one direction, the converter is operated such that the difference between the AC voltage angles α_1i,α_2j is substantially

$\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}$

and, for maximum power transfer in the opposite direction, the converter is operated such that the difference between the AC voltage angles α_1i,α_2j is substantially

$180-\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}.$

Preferably, the converter is operated such that, for power transfer, the AC voltage angle (s) α_1i is/are substantially 0 degrees; and the AC voltage angle (s) α_2j is/are substantially

$\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}$

for maximum power transfer in one direction, and substantially

$180-\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\mathrm{degrees}$

for maximum power transfer in the opposite direction.

With the third aspect of the invention, the principles of the invention are extended to a converter with n terminals on one side and m terminals on the other side. n and m may each be any positive integer and are not necessarily equal. It will be appreciated that if n=m=1, the third aspect of the invention is equivalent to the second aspect of the invention, and that if n=m=1 and p=1, the third aspect of the invention is equivalent to the first aspect of the invention. As with the second aspect of the invention, the number of phases p may be any positive integer, preferably 1, 2 or 3.

According to a fourth aspect of the invention, there is provided a converter for transferring power between a p phase AC system of AC voltage V_1ac of frequency ω, RMS line-neutral magnitude V_1acm and angle α₁ and a DC system of DC voltage V₂, the converter comprising: —

an AC/DC converter for transforming DC voltage V₂ into a p phase AC voltage V_2ac, of frequency ω, RMS line-neutral magnitude V_2acm and angle α₂; and

a LCL circuit for each phase p, wherein each LCL circuit comprises two inductors L₁,L₂ and a capacitor C connected together at their first terminals, the second terminal of inductor L₁ being connected to the respective phase of AC voltage V_1ac, and the second terminal of inductor L₂ being connected to the respective phase of AC voltage V_2ac;

wherein the value of the capacitor C, inductor L₁ and each inductor L_2j are selected to enable required power transfer and to

minimise current in inductor L₁,

and/or minimise currents in inductors L_2j,

Preferably, for maximum power transfer in one direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially 270 degrees and, for maximum power transfer in the opposite direction, the converter is operated such that the difference between the AC voltage angles α₁,α₂ is substantially 90 degrees.

With the fourth aspect of the invention, the principles of the invention are extended to a converter for transferring power between an AC system and a DC system. As compared with the first and second aspects of the invention, the first AC/DC converter is omitted, and the resonant circuit associated with each phase is connected directly to the respective phase of the AC voltage source.

According to a fifth aspect of the invention, there is provided a converter for transferring power between a p phase AC system of AC voltage V_1ac of frequency ω, RMS line-neutral magnitude V_1acm and angle α₁, and m DC systems respectively of DC voltage V_2j (j=1, 2, . . . m), where m is any positive integer, the converter comprising: —

• • m AC/DC converters for respectively transforming DC voltages V_2j into respective p phase AC voltages V_2jac, of frequency ω, RMS line-neutral magnitude V_2jacm and angle α_2j, and

a resonant circuit for each phase p, wherein each resonant circuit comprises an inductor L₁, m inductors L_2j and a capacitor C connected together at their first terminals, the second terminal of each inductor L₁ being connected to the respective phase of the respective AC voltage V_1ac, and the second terminal of each inductor L_2j being connected to the respective phase of the respective AC voltage V_2jac;

wherein the value of the capacitor C, inductor L₁ and inductor L₂ are selected to enable required power transfer and to

minimise currents in each inductor L_1i,

and/or minimise current in inductors L₂

Preferably, for maximum power transfer in one direction, the converter is operated such that the difference between the AC voltage angles α₁,α_2j is substantially 270 degrees and, for maximum power transfer in the opposite direction, the converter is operated such that the difference between the AC voltage angles α₁,α_2j is substantially 90 degrees. [j1]

With the fifth aspect of the invention, the principles of the invention are extended to a converter for transferring power between an AC system and m DC systems, where m may be any positive integer.

In all aspects of the invention, we aim minimizing current through inductors and therefore minimizing losses and stress in the switches and inductors. This is equivalent to minimizing reactive power circulation within each side of the converter.

According to the second aspect of the invention all values of k₁ in the range

$-\frac{V_2}{V_1}<k_1<1$

will give zero reactive current at low voltage (angle of I_1ac is equal to angle of V_1ac) and at high voltage side (angle of I_2ac is equal to angle of V_2ac), if other conditions are satisfied. All values for k1 in the above range imply minimal magnitude of I_1ac and I_2ac.

The actual value for k₁ will depend on the other design goals. Large positive values will give small values for inductors but the switching losses will be high. The value k₁=k₂=0 will give best responses under DC faults but will still give large currents switching instants and switching losses. A very large negative k1 implies minimal switching losses but the dynamic stability will be poor since natural resonance mode will be close to the switching frequency.

In all aspects of the invention, the converter preferably comprises control means for controlling power flow though the converter, wherein said control means comprises: —

means for varying AC voltage magnitude by changing the width of the AC pulses;

varying the angle difference between the AC sources; and/or

varying the operating frequency of the converter.

Preferably, the first DC system is a low voltage system and the second DC system is a high voltage system. That is to say, V₁ is preferably lower than V₂.

The AC voltages V_1ac,V_2ac of the two AC/DC converters are described as p phase AC voltages. That is, the AC/DC converters are configured to transform the respective DC voltage into an AC voltage with p phases, where p may be any positive integer value. Thus, for example, the AC voltages V_1ac,V_2ac may be single phase, 2 phase, 3 phase etc.

It will be appreciated that, in the case of converters with a single terminal on each side, the inner resonant circuit of a single phase converter will comprise one L₁ inductor, one L₂ inductor and one capacitor, whilst the inner resonant circuit of a 2 phase converter will comprise two L₁ inductors, two L₂ inductors and two capacitors C, and so on. More generally, the number of inductors L_1i,L_2j depends on the number of terminals on the respective side of the converter, as well as the number of phases, whereas the number of capacitors C only depends on the number of phases.

The DC/AC converters may each comprise 2p switches arranged as a bridge (or a half bridge), for transforming a DC voltage into a p phase AC voltage in known manner. Each bridge/half bridge may comprise p legs, where each leg comprises two switches, and corresponds to a different phase of the converter.

In cases where the converter has more than one phase, the second terminals of the capacitors are preferably connected to a common point, ie, in a star configuration. However, as with any other AC system, the capacitors may be connected in other configurations. For example, in delta configuration in the case of a 3 phase converter.

Where sides of the converter are referred to, this is with reference to the inner resonant circuits. In general there will be a low voltage side and a high voltage side, wherein the low voltage side is connected to a relatively low DC voltage and the high voltage side is connected to a relatively high DC voltage. The first DC system(s) of the first to third embodiments may be low voltage system(s) and the second DC system(s) may be high voltage system(s), ie V₁ (or V_1i) may be a relatively low and V₂ (or V_2j) may be relatively high. However, it will be appreciated that the DC voltages on each side of the converter may be equal.

It will be appreciated that in all aspects of the invention, the converter may be operated at less than maximum power by varying the (difference between) the AC voltage angles α₁,α₂ (or α_1i,α_2j) or varying the voltage magnitudes V_1ac, or V_2acm. Thus, whilst the converter may be operated for maximum power transfer by setting the difference between the AC voltage angles to be, for each k1,

$\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1},$

and for case

substantially 270 or 90 degrees, operation at lower k₁=0 it is powers may be achieved by varying the difference between these angles. In general, a difference in the range 180-360 degrees will give power transfer in one direction, and a difference in the range 0-180 degrees will give power transfer in the opposite direction.

It will be appreciated that a p phase AC system has p connecting nodes (wires).

In general, features described herein in relation to one aspect or embodiment of the invention may also be applied to any other aspect or embodiment of the invention described herein, either alone or in combination with other features.

According to an aspect of the invention there is provided a converter for transferring power up to maximum level P₁₂ between a DC system of low voltage V₁ and a DC system of high voltage V₂, comprising of:

an AC/DC converter transforming DC voltage V₁ into a single phase AC voltage V_ac1, of frequency ω, root mean square magnitude V_ac1m and angle α₁,

an AC/DC converter transforming DC voltage V₂ into a single phase AC voltage V_aC2, of same frequency ω, root mean square magnitude V_aC2m and angle α₂,

two inductors and a capacitor connected in a star L₁CL₂ circuit at their first terminals; the second terminal of capacitor C and second terminal of first inductor L₁ connecting to the first AC voltage V_ac1, the second terminal of the capacitor C and the second terminal of the second inductor L₂ connecting to the second AC voltage V_ac2,

where the components L₁, C and L₂ are selected for maximum power transfer at minimal losses as:

k ₁=1−ω²L₂C

k ₂=1−ω²L₁C

k ₃ =L ₁ +L ₂+ω²L₁L₂C

and where k₁ is selected in the region:

$-\frac{V_2}{V_1}<k_1<1$

and where

$k_2=k_1s^2$ $k_3=\frac{-p8V_1V_2\sqrt{1-{\left(V_1/V_2\right)}^2k_1^2}}{P_{1\mathrm{ac}}{\mathrm{\omega \pi }}^2}$

and operated in such way that for maximum power transfer the difference between AC voltage angles is

${\alpha }_2-\alpha -1=\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

for one power direction, or

${\alpha }_2-{\alpha }_1=180-\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

for the opposite power direction.

According to an aspect of the invention there is provided a converter for transferring power up to maximum level P₁₂ between a DC system of low voltage V₁ and a DC system of high voltage V₂, comprising of:

an AC/DC converter transforming DC voltage V₁ into a p phase AC voltage V_ac1, of frequency ω, root mean square magnitude V_ac1m and angle α₁,

an AC/DC converter transforming DC voltage V₂ into a p phase AC voltage V_ac2, of same frequency ω, root means square magnitude V_ac2m and angle α₂,

two inductors and a capacitor per each phase connected in a star L₁CL₂ circuit at their first terminals; the second terminal of first inductor L₁ connecting to one phase of first AC voltage V_ac1, the second terminal of the second inductor L₂ connecting to corresponding phase of the second AC voltage V_ac2, and the second terminals of all capacitors C connected to a common point

where the components L₁, C and L₂ are selected for maximum power transfer at minimal losses as:

k ₁=1−ω²L₂C

k ₂=1−ω²L₁C

k ₃ =L ₁ +L ₂−ω₂L₁L₂C

and where k₁ is selected in the region:

$-\frac{V_2}{V_1}<k_1<1$

and where

$k_2=k_1s^2$ $k_3=\frac{-p8V_1V_2\sqrt{1-{\left(V_1/V_2\right)}^2k_1^2}}{P_{1\mathrm{ac}}\omega {\pi }^2}$

and operated in such way that for maximum power transfer the difference between AC voltage angles is

${\alpha }_2-{\alpha }_1=\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

for one power direction, or

${\alpha }_2-{\alpha }_1=180-\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

for the opposite power direction.

According to an aspect of the invention there is provided a converter having n DC terminals on low voltage side and m DC terminals on high voltage side, transferring up to maximum power P_nm between low voltage and high voltage side, and comprising of:

n, AC/DC converters transforming DC voltages V_1i (i=1, 2, . . . n), into p phase AC voltages V_ac1i, of frequency ω, root mean square magnitude V_ac1im and angle α₁,

m, AC/DC converter transforming DC voltages V_2j (j=1, 2, . . . m), into a p phase AC voltage V_ac2j, of same frequency ω, magnitude V_ac2jm and angle α₂,

n L₁ inductors, m L₂ inductors and a capacitor C per each phase connected in a star L₁CL₂ connection at their first terminals, the second terminals of inductors L_1i connecting to one phase of AC voltages V_ac1i, the second terminals of inductors L_2j connecting to respective phases of AC voltages V_ac2j, and the second terminals of all capacitors C connected in a common point

where the components L₁, C and L₂ are selected for maximum power transfer and minimal currents. One feasible solution is:

the capacitor:

$C={\mathrm{nmP}}_{\mathrm{nm}}/\left(p\omega \sum _{i=1}^nV_{1\mathrm{iac}}\sum _{j=1}^mV_{2\mathrm{iac}}\right)$

the inductors L₁=n/(ω²C), L₂=m/(ω²C)

operated in such way that for maximum power transfer the AC voltage angles at low voltage side are α_1i=0 and the AC voltage angles at high voltage side are α_2j=270 for one power direction, but the AC voltage angles at high voltage side are α_2j=90 for the opposite power direction.

According to an aspect of the present invention there is provided a converter for transferring power up to maximum level P₁₂ between a p phase AC system of voltage V_ac1 with frequency ω, root mean square magnitude V_ac1m, angle α₁, and a DC system of voltage V₂, comprising of:

an AC/DC converter transforming DC voltage V₂ into a p phase AC voltage V_ac2, of frequency ω, root means square magnitude V_aC2m and angle α₂,

two inductors and a capacitor per each phase connected in a star L₁CL₂ circuit at their first terminals; the second terminal of first inductor L₁ connecting to one phase of AC voltage V_ac1, the second terminal of the second inductor L₂ connecting to corresponding phase of the voltage V_ac2, and the second terminals of all capacitors C connected to a common point

where the components L₁, C and L₂ are selected for maximum power transfer at minimal losses as:

k ₁=1−ω²L₂C

k ₂=1−ω²L₁C

k ₃ =L ₁ +L ₂−ω²L₁L₂C

and where k₁ is selected in the region:

$-\frac{V_2}{V_1}<k_1<1$

and where

$k_2=k_1s^2$ $k_3=\frac{-p8V_1V_2\sqrt{1-{\left(V_1/V_2\right)}^2k_1^2}}{P_{1\mathrm{ac}}{\mathrm{\omega \pi }}^2}$

and operated in such way that for maximum power transfer the difference between AC voltage angles is for

${\alpha }_2-{\alpha }_1=\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

one power direction, or

${\alpha }_2-{\alpha }_1=180-\mathrm{arctg}\sqrt{\frac{V_2^2}{V_1^2k_1^2}-1}$

for the opposite power direction.

Preferably, power flow is controlled using one of the following methods:

a) varying AC voltage magnitude by changing width of the AC pulses,

b) varying the angle difference between the AC sources,

c) varying the operating frequency.

Embodiments of the present invention will now be described with reference to the accompanying drawings in which: —

FIG. 1 a shows a 2 terminal, single phase DC/DC converter with half-bridge which embodies the invention;

FIG. 1 b shows a 2 terminal, 2 phase, DC/DC converter which embodies the invention;

FIG. 1 c shows a 2 terminal, 3 phase, DC/DC converter which embodies the invention;

FIG. 2 shows the AC voltage generated by the low voltage DC/AC bridge in the 2 phase converter of FIG. 1b;

FIG. 3 is a phasor diagram which illustrates operation of a 2 terminal converter;

FIGS. 4 a-4b illustrate the results of a PSCAD/EMTDC simulation of normal operation of a 2 terminal 2 phase converter;

FIGS. 5 a-5b illustrate the response of the PSCAD/EMTDC simulation to a change in control angle γ₁, for a 2 terminal 2 phase converter;

FIGS. 6 a-6b illustrate a PSCAD/EMTDC simulation of a reversal in power transfer direction, for a 2 terminal 2 phase converter;

FIGS. 7 a-7b illustrate the response of the PSCAD PSCAD/EMTDC simulation after a zero impedance fault on V₂;

FIG. 8 shows a 3 terminal, 2 phase DC/DC converter which embodies the invention;

FIG. 9 is a phasor diagram which illustrates operation of a 3 terminal converter;

FIGS. 10 a-10e illustrate the results of a PSCAD/EMTDC simulation of the 3 terminal 2 phase converter shown in FIG. 9;

FIG. 11 shows a multi-terminal 2 phase DC/DC converter which embodies the invention;

FIG. 12 shows a 2 terminal 3 phase DC/AC converter which embodies the present invention;

FIGS. 13 a-13c illustrates the response of a PSCAD PSCAD/EMTDC simulation of a 2 terminal 2 phase DC/AC converter for a low impedance fault on 80 kV DC bus; and

FIG. 14 shows a 2 phase DC/AC converter with 1 AC terminal and m DC terminals, which embodies the present invention.

FIGS. 1 a to 1b illustrate three DC/DC converters which embody the invention. The converters of FIGS. 1a to 1b are configured to transfer power between a first bipolar, centrally grounded DC system of low voltage ±V₁ and a second bipolar, centrally grounded DC system of high voltage ±V₂.

It will be appreciated that power may be transferred from the low voltage system to the high voltage system or from the high voltage system to the low voltage system. Thus, each voltage system may be either a source or sink, depending on the direction of power transfer.

FIG. 1 a shows the single phase topology of the converter. The converter comprises a low voltage DC/AC converter 10, an inner resonant circuit 12 and a high voltage DC/AC converter 14.

The two DC/AC converters 10, 14 each comprise two switches, (respectively S₁,S₂ and S₅,S₆) arranged as a half bridge. The low voltage DC/AC converter is connected to convert the low DC voltage V₁ into a first single phase AC voltage V_1ac. Similarly, the high voltage DC/AC converter is connected to convert the high DC voltage V₂ into a second single phase AC voltage V_2ac.

The inner resonant circuit 12 comprises two inductors L₁,L₂ and a capacitor C connected as a star L₁CL₂ circuit at their first terminals. The second terminal of inductor L₁ and the second terminal of the capacitor C are connected to the first AC voltage V_1ac. The second terminal of inductor L₂ and the second terminal of the capacitor C are connected to the second AC voltage V_2ac.

In FIG. 1a, each pole of the low voltage system is shown to be connected to a first terminal of a resistor R_dc, the second terminal of which is connected to a first terminal of a respective capacitor C₁. The second terminal of each capacitor C₁ is connected to ground. A similar arrangement exists on the high voltage side of the converter where the capacitor is labelled C₂. The capacitors C₁,C₂ filter the harmonics and improve power quality. The resistors R_dc are included to represent a general DC load. The current I_dc1 (I_dc2) consists of half cycles of single polarity, whereas current I₁ (I₂) is true DC current as would be required in many applications. A half-bridge (single phase) topology suffers very high harmonic content. Thus, multi phase topologies may be more desirable for higher-power applications.

FIG. 1 b shows the 2 phase topology of the converter. The circuit of FIG. 1b is similar to that of FIG. 1a. However, the two DC/AC converters 10′, 12′ each comprise four switches (respectively S₁-S₄ and S₅-S₈) arranged as a full bridge (two legs).

The low voltage DC/AC converter is connected to convert the low DC voltage V₁ into a first 2 phase AC voltage V_1ac. Similarly, the high voltage DC/AC converter is connected to convert the high DC voltage V₂ into a second 2 phase AC voltage V_2ac.

The inner resonant circuit 12′ comprises two inductors L₁, L₂ and a capacitor C, for each phase of the converter. The components associated with each phase are connected as a star L₁CL₂ circuit at their first terminals, and the second terminals of the inductors L₁,L₂ are respectively connected to the corresponding phases of the AC voltages V_1ac and V_2ac. The second terminals of the capacitors for both phases are connected to a common point.

FIG. 1 c shows the 3 phase topology of the converter. The circuit is similar to that of FIG. 1b. However, the DC/AC bridges 10″,14″ comprise 6 switches (respectively S₁-S₆ and S₇-S₁₂) arranged in three legs, to produce a 3 phase AC voltage. Correspondingly, the inner resonant circuit 12″ comprises three sets of inductors L₁, L₂ and capacitor C, one set associated with each phase. A converter with a greater number of phases may be implemented by applying the same principles. In the circuit of FIG. 1c, the capacitors C are connected to a common point (star point). However, alternative delta connection of the capacitors is also shown.

In what follows, reference is made to the 2 phase converter illustrated in FIG. 1b. However, it will be appreciated that the same or a similar analysis applies to topologies with different numbers of phases.

A design goal is controllable power transfer between a low voltage DC system V₁ and a high voltage system V₂. It is assumed that V₁ and V₂ do not change polarity, but that they can deliver or sink power. Thus, bi-directional power flow is achieved by means of current direction change.

Each of the DC/AC bridges 10′,14′ may be operated in a square-wave fashion or using other methods for generating AC from DC. FIG. 2 shows the AC voltage output V_1ac of the low voltage DC/AC bridge. V₁ is the magnitude of the low voltage DC source, α₁ is the phase angle of V_1ac, γ₁ is the conduction angle of V_1ac, V_1acf is the fundamental component of V_1ace and V_1acmm is the peak magnitude of V_1ac.

Square-wave operation of the DC/AC bridges is achieved by sequentially firing gates g₁ and g₂ on the low voltage side, and gates g₅ and g₆ on the high voltage side. The converter AC voltages V_1ac and V_2ac can be expressed as:

V _1ac =V _1acm∠α₁=V_1acx+jV_1acy  (1)

V _2ac =V _2acm∠α₂=V_2acx+jV_2acy  (2)

where V_1acm is the magnitude of V_1ac, V_2acm is the magnitude of V_2ac, α₁ is the phase angle of V_1ac and α₂ is the phase angle of V_2ac.

The firing angles α₁,α₂ and the conduction angles γ₁,γ₂ can be externally manipulated from the converter controller

(not shown). Using Fourier series expansion, the root mean square (RMS) of the fundamental component of the AC voltage magnitude is (line neutral):

V _1acm=4V₁/(√2π)·sin(γ₁/2),V_2acm=4V₂/(√2π)·sin(γ₂/2)  (3)

In equation (3), the conduction angles γ₁,γ₂ can be used to control the magnitude of the AC voltage, considering that the DC voltages V₁ and V₂ are constant. The maximum power is obtained for full conduction, ie γ₁=γ₂=180 degrees.

Further the xy components of control signal M_1acx and M_1acy can be represented as

$\begin{array}{c}{V}_{1\mathrm{acx}}=\frac{4V_1}{\pi \sqrt{2}}\sin \frac{{\gamma }_1}{2}\cos {\alpha }_1\\ =\frac{4V_1}{\pi \sqrt{2}}M_{1m}\cos {\alpha }_1\\ =\frac{4V_1}{\pi \sqrt{2}}M_{1x}\end{array}$ $\begin{array}{c}{V}_{1\mathrm{acy}}=\frac{4V_1}{\pi \sqrt{2}}\sin \frac{{\gamma }_1}{2}\sin {\alpha }_1\\ =\frac{4V_1}{\pi \sqrt{2}}M_{1m}\sin {\alpha }_1\\ =\frac{4V_1}{\pi \sqrt{2}}M_{1y}\end{array}$ $\begin{array}{c}{M}_{1m}=\sqrt{M_{1x}^2+M_{1y}^2}\\ =\sin \frac{{\gamma }_1}{2}\end{array}$

The basic equations for the inner resonant circuit are:

I _1ac =I _1acm∠β₁=(V_1ac−V_c)/(jωL₁),  (4)

I _2ac =I _2acm∠β₂=(V_2ac−V_c)/(jωL₂)  (5)

jωCV _c =I _1ac +I _2ac  (5)

where  =2πf_s and f_s is the converter switching frequency. The coordinate frame can be positioned arbitrarily. Thus, without loss of generality, V_1ac can be positioned on the x-axis (α₁=0). Thus:

V _1ac =V _1acx =V _1acm ,V _1acy=0  (6)

We will conveniently rewrite equations (4)-(5) as:

$\begin{array}{cc}\stackrel{\_}{I_{1\mathrm{ac}}}=\frac{V_{1\mathrm{acm}}k_1-\stackrel{\_}{V_{2\mathrm{ac}}}}{j\omega ·k_3}& \left(17\right)\\ \stackrel{\_}{I_{2\mathrm{ac}}}=\frac{\stackrel{\_}{V_{2\mathrm{ac}}}k_2-V_{1\mathrm{acm}}}{j\omega ·k_3}& \left(18\right)\end{array}$

where the coefficients k₁, k₂ and k₃ are:

k ₁=1−ω²L₂C  (19)

k ₂=1−ω²L₁C  (20)

k ₃ =L ₁ +L ₂−ω²L₁L₂C  (21)

The above variables k₁, k₂ and k₃ are conveniently introduced to study converter behavior. Ultimately, we need to determine the three parameters: L₁, L₂ and C, which can be obtained from the three equations (19)-(21). The coefficient k₃ is a positive non-zero constant that is fully determined by the power transfer level. Coefficients k₁ and k₂ are manipulated in the design stage, as it is discussed in section C below. Since ω>0, L₁>0, L₂>0, C>0, from (19)-(21) we have upper limits:

k ₁<1,k₂<1,k₃<L₁+L₂  (22)

A. Converter Controllability for Minimal Reactive Power Circulation

It is essential to minimize reactive power flow in LCL circuit in order to operate with minimum current magnitudes and to reduce switching losses.

The converter has 3 control signals that can be manipulated M_1d, M_2d and M_2q. to minimise reactive power. Using (17) the low voltage side AC current I_1ac is:

$\begin{array}{cc}\begin{array}{c}\stackrel{\_}{I_{1\mathrm{ac}}}=\frac{V_{1\mathrm{acm}}k_1-V_{2\mathrm{acd}}-jV_{2\mathrm{acq}}}{j\omega ·k_3}\\ =\frac{-V_{2\mathrm{acq}}}{\omega ·k_3}-j\frac{V_{1\mathrm{acm}}k_1-V_{2\mathrm{acd}}}{\omega ·k_3}\end{array}& \left(23\right)\end{array}$

The condition for zero reactive power at terminal 1, I_1acq=0, is obtained from (23):

0=V₁M_1dk₁−V₂M_2d  (24)

Therefore the control law for I_1acq=0 is from (24):

$\begin{array}{cc}{M}_{1d}=\frac{V_2M_{2d}}{V_1k_1}& \left(25\right)\end{array}$

Using (25) and (23) we derive power expression:

$\begin{array}{cc}\begin{array}{c}{P}_{1\mathrm{ac}}=\mathrm{Re}\left(\stackrel{\_}{V_{1\mathrm{ac}}}·{\stackrel{\_}{I_{1\mathrm{ac}}}}^{*}\right)\\ =-{\mathrm{pV}}_{1\mathrm{acm}}\frac{V_{2\mathrm{acq}}}{\omega k_3}\\ =-{\mathrm{pV}}_{1\mathrm{acm}}\frac{2\sqrt{2}V_2M_{2q}}{\pi \omega k_3}\end{array}& \left(26\right)\end{array}$

where p is the number of phases (p=2 for the converter in FIG. 1). Therefore M_2q is used for controlling power.

On the high voltage side, AC current is from (18):

$\begin{array}{cc}\begin{array}{c}\stackrel{\_}{I_{2\mathrm{ac}}}=\frac{V_{2\mathrm{acd}}k_2-jV_{2\mathrm{acq}}k_2-V_{1\mathrm{acm}}}{j\omega ·k_3}\\ =\frac{-V_{2\mathrm{acq}}k_2}{\omega ·k_3}+j\frac{V_{1\mathrm{acm}}-V_{2\mathrm{acd}}k_2}{\omega ·k_3}\end{array}& \left(27\right)\end{array}$

Using (27), zero reactive power condition at terminal 2 is:

$\begin{array}{cc}\begin{array}{c}\frac{V_{2\mathrm{acq}}}{V_{2\mathrm{acd}}}=\frac{I_{2\mathrm{acq}}}{I_{2\mathrm{acd}}}\\ =\frac{V_{1\mathrm{acm}}-V_{2\mathrm{acd}}k_2}{V_{2\mathrm{acq}}k_2}\end{array}& \left(28\right)\end{array}$

After rearranging (28) and using (25):

$\begin{array}{cc}{M}_{2q}^2V_2^2k_2=\frac{M_{2d}^2V_2^2}{k_1}-M_{2d}^2V_2^2k_2& \left(29\right)\end{array}$

Therefore replacing (24) in (29), to provide zero reactive power on high voltage side we operate the converter as:

$\begin{array}{cc}{M}_{2d}=\frac{M_{2q}}{\sqrt{\frac{1}{k_2k_1}-1}}& \left(30\right)\end{array}$

The above equation implies that the phase angle of AC voltage V_2ac is constant and for a given k₁ and k₂:

$\begin{array}{cc}\begin{array}{c}{\alpha }_2=\mathrm{arctg}\frac{M_{2q}}{M_{2d}}\\ =\mathrm{arctg}\sqrt{\frac{1}{k_2k_1}-1}\end{array}& \left(31\right)\end{array}$

B. Calculation of k₁, k₂ and k₃

The influence of k₁, k₂ and k₃ will be discussed in this section, considering full power transfer. We assume that maximum control signals are provided at maximum power transfer (in order to get minimal currents).

M _1mmax =M _2mmax=1  (32)

Replacing (32) in (25):

$\begin{array}{cc}{M}_{2d\max }=\frac{V_1k_1}{V_2}& \left(33\right)\end{array}$

Since |M_2d|≦1, |M_1d|≦1, we get the boundaries for k₁:

$\begin{array}{cc}-\frac{1}{s}<k_1<1& \left(34\right)\end{array}$

where s is the stepping ratio V₁/V₂. Using (33), we can also calculate the maximum q component control signal M_2qmax:

M _2qmax=√{square root over (1−M_2dmax²)}=√{square root over (1−s²k₁²)}  (35)

Replacing (33) in (29) at maximum power (32):

M _2q ² V ₂ ² k ₂ =V ₁ ² k ₁ −M _2d ² V ₂ ² k ₂  (36)

Re-writing (36), we obtain formula for calculating k₂:

k ₂ =k ₁ s ²  (37)

The coefficient k₃ is calculated from active power condition (26), and replacing (33):

$\begin{array}{cc}\begin{array}{c}{k}_3=\frac{-{\mathrm{pV}}_{1\mathrm{acm}}M_{2q\max }2\sqrt{2}V_2}{P_{1\mathrm{ac}}\omega \pi }\\ =\frac{-{\mathrm{pV}}_18V_2\sqrt{1-s^2k_1^2}}{P_{1\mathrm{ac}}\omega {\pi }^2}\end{array}& \left(38\right)\end{array}$

The converter design procedure is as follows: —

• 1) Given data are V₁, V₂ and P₁ (assuming lossless converter P₁=P_1ac)
• 2) Using equation (3), calculate V_1acm and V_2acm.
• 3) Considering the physical nature of the switches, select a suitable switching frequency f_s, where ω=2πf_s).
• 4) Select the number of phases p.
• 5) Select value of k1 according to the performance requirements for particular converter. Larger negative k1 will give lower switching loses but larger fault currents. The lowest fault currents are achieved for k₁=0. Note that all k₁ from the above range will give minimal currents.
• 6) Determine k2 and k3 from equations (xx) (yy).
• 7) Determine C, L1 and L2 from equations (zz)

A good initial guess for k1 is middle of the range.

A test system module has been developed using a PSCAD/EMTDC simulator. The test system data for a 3 MW, 2 kV/40 kV converter are given in tables 1a and 1b.

	TABLE 1a
	Given Parameters	Value
	Power P1	3	MW
	Pole voltage V1	2	kV
	Pole voltage V2	40	kV
	Switching frequency f	1	kHz
	DC line resistance Rdc	0.05Ω
	Filter capacitance C1	300	μF
	Filter capacitance C2	20	μF
	Number of phases p	2

	TABLE 1b
	Calculated Parameters	Value
	Line-neutral AC voltage V1ac	1.8	kV
	Line-neutral AC voltage V2ac	36	kV
	Capacitance C	3.68	μF
	Inductance L1	6.88	mH
	Inductance L2	6.88	mH

FIGS. 4 a to 4d show the converter PSCAD model simulation in steady-state in step-up mode. It can be seen that the internal AC circuit correctly achieves the voltage and current stepping, ie, low V_1ac and I_2ac, high V_2ac and I_1ac. It can also be seen that voltages and currents at the respective terminals are in phase, which implies low conduction losses due to the minimal current magnitude, and low switching losses due to zero current switching. FIG. 4a shows that I_2ac has some harmonic content. The voltages in FIGS. 4c and 4b are measured as pole-pole (phase phase) and therefore the values are twice those given in tables 1a and 1b.

Controllability of the converters is illustrated in FIGS. 1a to 1c and will now be discussed.

Replacing (2) and (3) in equation (24) gives the power equation:

P _1ac =pωC4V₁/(√2π)·sin(γ₁/2)·4V₂/(√2π)·sin(γ₂/2)·sin(α₂)  (28)

Considering (28) and (9), the power can be controlled in four ways:

• 1) using the conduction angle γ₁;
• 2) using the conduction angle γ₂;
• 3) using the phase angle α₂ (ie the difference between α₂ and α₁); and/or
• 4) using the switching frequency ω.

The use of γ₁ for converter control has been tested using the non-linear PSCAD simulator. The test results are illustrated in FIGS. 5a-5e. In the test, γ₁ was reduced in three steps, from 180, to 100, to 60 degrees. The main converter variables are shown. It is evident that the power reduces as the conduction angle is reduced, as predicted by (28). It can also be seen that the peak value of the internal AC variables does not increase as the conduction angle is varied.

Since the conduction angle can not be negative, it is not possible to reverse the direction of power transfer by varying the conduction angle. Instead power reversal is achieved by changing α₂ from 270 to 90 degrees, noting that α₂ should stay on the y-axis to enable zero reactive power flow. This power reversal method can be deduced from the phasor diagram in FIG. 3 and from equation (11). FIGS. 6a-6c show a PSCAD simulation of power reversal using the above method (ie, 3MW->−3MW). It can be seen that the converter achieves stable operation at −3MW power transfer.

Operation of the converters illustrated in FIGS. 1a to 1b under faults is discussed below.

An aim of the present invention is to develop a converter which is fully tolerant to external faults on either of the DC voltages V₁, V₂.

Assuming a fault on V1, we put V1=0 in the above equations and the fault current I1acf1m relative to rated current I1acm is:

$\frac{I_{1\mathrm{acf}1m}}{I_{1\mathrm{acrm}}}=\frac{\sqrt{V_{2\mathrm{acrd}}^2+V_{2\mathrm{acrq}}^2}}{V_{2\mathrm{acrq}}}=\frac{1}{\sqrt{1-k_1^2s^2}}$

In the above formula the fault current magnitude is very close to rated value for all reasonable k1.

Similarly for faults on V2, we put V2=0 to obtain:

$\frac{I_{2\mathrm{acf}1m}}{I_{2\mathrm{acrm}}}=\frac{1}{\sqrt{1+s^2\left(1/k_2^2-2k_1/k_2\right)}}$

IN the above formula the fault current is lower than rated current form most k1.

Assuming a special case k1=0, the fault Equations (16) and (17) can be rearranged as follows,

ωL₁=−V_2acm/I_1ac  (29)

ωL₂=−V_1acm/I_2ac  (30)

It is evident from equations (29) and (30) that the converter has constant V/I ratio, which implies favourable fault responses. A voltage depression on one side will cause current reduction on the opposite side, given that ωL is constant. An extreme fault on one side of the converter, eg V₁=0, will cause open circuit on the opposite side, eg I₂=0. Accordingly, such faults are not transferred through the converter. Thus, the converter can be used as a DC circuit breaker, which prevents DC fault propagation.

In contrast, a common 2-coil AC transformer has the following characteristic: V₂/V₁=N₂/N₁, where N₁ and N₂ are the number of turns of the transformer coils. Such a transformer will directly transfer voltage depressions from one side to another. Accordingly, external protection circuits are required to interrupt fault currents (AC circuit breaker). FIGS. 7a-7d show the response of the PSCAD simulation to a worst case zero-impedance pole-pole fault on the high voltage terminals, in step-up mode. It can be seen that the current on the input side (I₁) automatically reduces without any additional control intervention. The current on the faulted side (I₂) transiently increases during the fault, but the switches can be slightly overrated to withstand this temporary overcurrent condition.

FIG. 8 shows a 3 terminal converter which embodies the invention. For simplicity, only the 2 phase topology is discussed herein. However, it will be appreciated that the same or a similar analysis applies to topologies with different numbers of phases. The three terminal configuration may be desirable, for example, for connecting two low voltage sources to a high-power network.

The converter of FIG. 8 is similar to that shown in FIG. 1b, except that the circuit comprises two terminals on the low voltage side, ie, three terminals in total. That is to say, the converter transfers power between two low voltage systems (V₁₁ and V₁₂) and one high voltage source V₂. To simplify study we only consider the case of k₁=0.

Both low voltage terminals are connected to a respective DC/AC bridge which respectively converts the DC voltages V₁₁,V₁₂ into 2 phase AC voltages V_11ac, V_12ac. The high voltage terminal is connected to a DC/AC bridge which converts the DC voltages V₂ into 2 phase AC voltage V_2ac.

For each phase of the converter, the inner resonant circuit comprises two inductors L₁₁ and L₁₂ respectively associated with the two low voltage terminals, a third inductor L₂ associated with the high voltage terminal, and a capacitor C. The components L₁₁,L₁₂,L₂,C associated with each phase are connected together at their first terminals. The second terminals of inductors L₁₁,L₁₂,L₂ associated with each phase are connected to the corresponding phase of the respective AC voltages V_11ac, V_12ac, V_2ac. The capacitors

associated with both phases are connected together at a common point.

The basic converter equations are:

I _11ac=(V_11ac−V_c)/(jωL₁₁),I_12ac=(V_12ac−V_c)/(jωL₁₂),I_2ac=(V_2ac−V_c)/(jωL₁₂)  (31)

jωCV _c =I _11ac +I _12ac +I _2ac  (32)

Replacing (31) in equation (32) gives:

-ω²L₁₁L₁₂L₂CV_c=(V_11ac−V_c)L₁₂L₂+(V_12ac−V_c)L₁₁L₂+(V_2ac−V_c)L₁₁L₁₂  (33)

Equation (33) can be rearranged as:

V _cx +jV _cy=(V_11acL₁₂L₂+V_12acL₁₁L₂+jV_2acL₁₁L₁₂)/(L₁₁L₂+L₁₂L₂+L₁₁L₁₂−ω²L₁₁L₁₂L₂C)  (34)

Assuming the low voltages V₁₁,V₁₂ are aligned with the x-axis and the high voltage V₂ is aligned with the y-axis:

V _11ac =V _11acx ,V _12ac =V _12acx ,V _2ac =jV _2acy  (35)

In order to reduce reactive power circulation on the high voltage side, the y-component of the capacitor voltage should be equal to the voltage on the y-axis:

V _cy =V _2ac  (36)

Replacing (31) and (36) in equation (34) gives:

V _2ac(L₁₁L₂+L₁₂L₂+L₁₁L₁₂−ω²L₁₁L₁₂L₂C)=V_2acL₁₁L₁₂  (37)

Assuming L₁₁=L₁₂, (37) gives:

C=2/(ω²L₁₁)  (38)

Thus, the capacitor size is twice the value calculated for the two-terminal converters of FIGS. 1a-1c (cf equation (19)). The elimination of reactive power circulation on the low voltage side is more complex, because two external voltages are involved. The goal is to get the x-component of the capacitor voltage equal to each of the voltages on the x-axis. As this is not possible to achieve in general, the optimum strategy is to have the x-component of the capacitor voltage equal to the average value of the voltages on the x-axis:

V _cx=(V_11ac+V_12ac)/2  (39)

In the case where V_11ac=V_12ac, the x-component of the capacitor voltage will be equal to both V_11ac and V_12ac, and there will be no reactive power circulation. FIG. 9 shows the phasor diagram for this topology. Using equations (34) and (39):

$\begin{array}{cc}\left(V_{11ac}+V_{12ac}\right)/2·\left(L_{11}L_2+L_{12}L_2+L_{11}L_{12}-{\omega }^2L_{11}L_{12}L_2C\right)=V_{11ac}L_{12}L_2+V_{12ac}L_{11}L_2& \left(40\right)\end{array}$

Combining equations (40) and (37):

L2=L₁₁/2  (41)

The power for the high voltage terminal is:

P _2ac=2V_2acI_2ac=2V_2ac(V_11ac+V_12ac)/(2ωL₂)=2V_2acωC(V_11ac+V_12ac)/2  (42)

If the voltages on the low voltage side are equal, equation (42) becomes:

P _2ac=2V_2acV_11acωC  (43)

If the voltages on the x-axis are equal, the power is:

P _11ac =P _12ac=2V_11acI_11ac=2V_11acV_2ac/ωL₁₁=V_11acV_2acωC  (44)

From equations (43) and (44) it can be deduced that the high-voltage converter should be rated for the sum of powers on the low voltage converters.

The converter design procedure is as follows: —

• 1) Given data are V₁₁,V₁₂,V₂ and P₂ (P₂=P_2ac).
• 2) Using equation (3), calculate V_11ac, V_12ac, V_2ac assuming maximum power transfer (γ=180).
• 3) Considering the physical nature of the switches, select a suitable switching frequency f_s, where ω=2πf_s).
• 4) Select the number of phases p.
• 5) Determine a value for capacitance C using (43), which rearranges as:

C=P ₂ /{pωV _2ac(V_11ac+V_12ac)}  (45)

• 6) Determine values for inductors L₁₁,L₁₂,L₂ from equations (38) and (41), ie:

L2=1/(ω²C),L₁₁=L₁₂=2/(ω²C)  (46)

The value for L₂ may be tuned in the final tests.

A test system has been developed using a PSCAD/EMTDC simulator. The test system data are given in tables 2a and 2b.

	TABLE 2a
	Given Parameters	Value
	Power P11ac	3	MW
	Power P12ac	3	MW
	Pole voltage V11	2	kV
	Pole voltage V12	2	kV
	Pole voltage V2	40	kV
	Switching frequency fs	1	kHz
	DC line resistance Rdc	0.05Ω
	Filter capacitance C11	300	μF
	Filter capacitance C12	300	μF
	Filter capacitance C2	20	μF
	Number of phases p	2

	TABLE 2b
	Calculated Parameters	Value
	Line-neutral AC voltage V11ac	1.8	kV
	Line-neutral AC voltage V12ac	1.8	kV
	Line-neutral AC voltage V2ac	36	kV
	Capacitance C	7.4	μF
	Inductance L11	6.88	mH
	Inductance L12	6.88	mH
	Inductance L2	3.44	mH

FIGS. 10 a-10e show the 3-terminal converter PSCAD model simulation in steady state. A negative step on the V₁₂ terminal conduction angle γ₁₂ is applied to illustrate controllability of the converter.

It can be seen that the converter steady-state variables are as predicted by the above equations. All terminals operate at zero reactive power flow. The reduction in γ₁₂ reduces power on the V₁₂ terminal, and also reduces the exit power P₂, which is the sum of P₁₁ and P₁₂. It can also be seen that P₁₁ is unchanged, which demonstrates that power on each terminal can be individually controlled.

FIG. 11 shows a multi-terminal converter which embodies the invention. For simplicity, only the 2 phase topology is discussed. However, it will be appreciated that the same or a similar analysis applies to topologies with different numbers of phases.

The converter of FIG. 11 is similar to that shown in FIG. 8, except that the circuit comprises n terminals on the low voltage side and m terminals on the high voltage side, where n and m can each be any positive integer and are not necessarily equal.

The n low voltage terminals are connected to respective DC/AC bridges which respectively convert the DC voltages V_1i into 2 phase AC voltages V_1iac. The m high voltage terminals are connected to respective DC/AC bridges which respectively convert the DC voltages V_2i into 2 phase AC voltage V_2iac.

For each phase of the converter, the inner resonant circuit comprises n inductors L₁₁ respectively associated with the n low voltage terminals, m inductors L_2j respectively associated with the m high voltage terminals, and a capacitor C. The components L_1i,L_2j,C associated with each phase are connected together at their first terminals. The second terminals of inductors L_1i,L_2j associated with each phase are connected to the corresponding phase of the respective AC voltages V_1iac, V_2jac. The capacitors associated with both phases are connected together at a common point.

In order to minimize reactive power circulation, the capacitor voltages on one axis should be close to the average value. Thus, all low voltages should be aligned with the x-axis, and all high voltages with the y-axis.

The basic circuit equations are:

$\begin{array}{cc}{I}_{11ac}=\left(V_{11ac}-V_c\right)/\left(j\omega L_{11}\right),I_{12ac}=\left(V_{12ac}-V_c\right)/\left(j\omega L_{12}\right),\dots I_{1nac}=\left(V_{1\mathrm{nac}}-V_c\right)/\left(j\omega L_{1n}\right)& \left(47\right)\\ I_{21ac}=\left(V_{21ac}-V_c\right)/\left(j\omega L_{21}\right),I_{22ac}=\left(V_{22ac}-V_c\right)/\left(j\omega L_{22}\right),\dots I_{2\mathrm{mac}}=\left(V_{2\mathrm{mac}}-V_{c}\right)/\left(j\omega L_{2m}\right)& \left(48\right)\\ j\omega {\mathrm{CV}}_c=\sum _{i=1}^nI_{1\mathrm{iac}}+\sum _{j=1}^mI_{2\mathrm{jac}}& \left(49\right)\end{array}$

Assuming that L₁₁=L_1i=L_1n=L₁ and L_2i=L_2i=L_2n=L₂, and replacing (47) and (48) in equation (49) gives:

$\begin{array}{cc}-{\omega }^2L_1^n{\mathrm{CV}}_c=\sum _{i=1}^n\left(V_{1\mathrm{iac}}-V_c\right)L_1^{n-1}L_2^m+\sum _{j=1}^m\left(V_{2\mathrm{jac}}-V_c\right)L_1^nL_2^{m-1}& \left(50\right)\end{array}$

Given the assumption that all n-voltages are located on the x-axis and all m-voltages on the y-axis, (50) becomes:

$\begin{array}{cc}{V}_{\mathrm{cx}}+jV_{\mathrm{cy}}=\frac{\left(\sum _{i=1}^nV_{1\mathrm{iac}}L_1^{n-1}L_2^m+j\sum _{j=1}^mV_{2\mathrm{jac}}L_{1}^nL_2^{m-1}\right)}{\left({\mathrm{nL}}_1^{n-1}L_1^m+mL_1^nL_1^{m-1}-{\omega }^2L_1^nL_2^mC\right)}& \left(51\right)\end{array}$

The goal is for the y-component of the capacitor voltage to be equal to the average voltage on the y-axis:

$\begin{array}{cc}{V}_{\mathrm{cy}}=\left(1/m\right)\sum _{j=1}^mV_{2\mathrm{jac}}& \left(52\right)\end{array}$

Replacing (52) in equation (50) gives:

$\begin{array}{cc}\left(1/m\right)\sum _{j=1}^mV_{2\mathrm{jac}}\left({\mathrm{nL}}_1^{n-1}L_1^m+mL_1^{m-1}-{\omega }^2L_1^nL_2^mC\right)=\sum _{i=1}^mV_{2\mathrm{iac}}L_1^nL_2^{m-1}& \left(53\right)\end{array}$

Solving equation (52) gives the capacitance:

C=n/(ω²L₁)  (54)

Minimising current magnitudes on the low voltage side, the goal is for the x-component of the capacitor voltage to equal the average voltage on the y-axis:

$\begin{array}{cc}{V}_{\mathrm{cx}}=\left(1/n\right)\sum _{i=1}^nV_{1\mathrm{iac}}& \left(55\right)\end{array}$

Replacing (55) in equation (51) gives:

$\begin{array}{cc}\left(1/n\right)\sum _{i=1}^nV_{1\mathrm{iac}}\left({\mathrm{nL}}_1^{n-1}L_1^m+mL_1^nL_1^{m-1}-{\omega }^2L_1^nL_2^mC\right)=\sum _{i=1}^nV_{1iac}L_1^{n-1}L_2^m& \left(56\right)\end{array}$

Solving (56) gives the capacitance:

C=m/( ²L₂)  (57)

From equations (57) and (54):

L2=mL₁/n  (58)

The power expressions can be derived as:

$\begin{array}{cc}{P}_{1\mathrm{iac}}=2V_{1\mathrm{iac}}I_{1\mathrm{iac}}=2V_{1\mathrm{iac}}\left(1/m\right)\sum _{j=1}^mV_{2\mathrm{jac}}/\left(\omega L_1\right)=2V_{1\mathrm{iac}}\sum _{j=1}^mV_{2\mathrm{jac}}\omega C/\left(\mathrm{mn}\right)& \left(59\right)\\ P_{2\mathrm{iac}}=2V_{2\mathrm{jac}}I_{2\mathrm{iac}}=2V_{2\mathrm{jac}}\left(1/n\right)\sum _{i=1}^nV_{1\mathrm{iac}}/\left(\omega L_2\right)=2V_{2\mathrm{iac}}\sum _{i=1}^nV_{1\mathrm{iac}}\omega C/\left(nm\right)& \left(60\right)\end{array}$

If all voltages on corresponding axes are equal, V₁₁=V_1i=V_1n, V_2i=V_2j=V_2n, then:

P _1iac=2V_1iacI_1iac=2V_1iacV_2jac/(ωL₁)=2V_1iacV_2jacωC/n  (61)

P _2iac=2V_2jacI_2jac=2V_2jacV_1iac/(ωL₂)=2V_2jacV_1iacωC/m  (62)

The converter design procedure is as follows: —

• 1) Given data are V_1i,V_2i and the power transfer function between the x and y axes P_xy, where:

$P_{\mathrm{xy}}=\sum _{i=1}^nP_{1i}=\sum _{j=1}^mP_{2j}$

• 2) Using equation (3), and assuming maximum power (γ=180) calculate V_1iac,V_2jac.
• 3) Considering the physical nature of the switches, select a suitable switching frequency f_s, where ω=2πf_s).
• 4) Select the number of phases p.
• 5) Determine a value for capacitance C using:

$\begin{array}{cc}C={\mathrm{nmP}}_{\mathrm{xy}}/\left(p\omega \sum _{i=1}^nV_{1\mathrm{iac}}\sum _{j=1}^mV_{2\mathrm{jac}}\right)& \left(63\right)\end{array}$

• 6) Determine values for inductors L₁ and L₂ using (54) and (57), ie,

L ₁ =n/(ω²C),L₂=m/((ω²C)  (64)

Although FIG. 11 shows at least three terminals on each side, in the general case, n and m may each be any positive integer, and are not necessarily equal.

The converter concept described above can also be extended to exchange power on the inner AC circuit. One application is to interconnect a DC source/sink with an AC source/sink. This can be achieved if a converter bridge is replaced with an AC source. FIG. 12 shows a 2 terminal, 3 phase converter which embodies the present invention. The converter shown in FIG. 12 is similar to that shown in FIG. 1c, except that the low voltage side converter bridge is omitted. Instead, the resonant circuit is connected directly to a source of 3 phase AC voltage source. The converter retains all the good properties related to voltage stepping, power control and fault isolation discussed in connection with the preceding embodiments of the invention.

FIGS. 13 a-13c show simulation responses for a converter which connects an AC system with a DC system. The test system is similar to that used to obtain the results of FIGS. 4a-4d, and uses the test parameters set out in tables 1a and 1b, except that the 4 kV converter is replaced by an AC source. This test system would correspond, for example, to an HVDC converter connecting to an AC grid. It can be seen that the converter transfers 3MW and that voltage and current transformation is achieved in normal operation. The simulation imposes a severe DC fault on VAC2 at around 0.307 s. The converter can be seen to have a favourable reaction to the fault in that the power naturally reduces without any overcurrent or overvoltage, and the converter retains normal operation during the fault.

FIG. 12 shows a 3 phase, 2 terminal converter, which connects a low voltage AC system to a high voltage DC system.

However, in the general case, the converter may comprise m DC terminals, where m may be any positive integer. Moreover, the circuit may have p phases, where p is any positive integer.

FIG. 14 shows a 2 phase DC/AC converter which embodies the present invention. The converter shown in FIG. 12 is similar to that shown in FIG. 12, except that the converter transfers power between an AC system on one side and m DC systems on the other side. The converter retains all the good properties related to voltage stepping, power control and fault isolation discussed in connection with the preceding embodiments of the invention.

For clarity of explanation, the converters have been described as having a low voltage side and a high voltage side. However, it will be appreciated that, in certain circumstances, the voltage on both sides of the converter may be equal, such that the converter does not have a low voltage side and a high voltage side. However, it will be appreciated that the above descriptions above still apply in such circumstances, if the terms “low voltage” and “high voltage” are simply regarded as labels for the two sides of the converter.

The converter of the invention comprises control means for controlling AC voltage magnitude, frequency and phase. For example, one way of controlling these parameters is to generate a single square wave as illustrated in FIG. 2, and to vary the width of the square pulses to vary the AC voltage magnitude; the phase shift with respect to an arbitrary coordinate frame to achieve phase angle variation of the AC voltage; and the frequency of the square pulses to vary the frequency of the AC voltage. The square pulse can be achieved in many other ways, using direct square control on switches or phase difference between opposite switches. A multilevel converter or PWM converter is yet another possibility.

A suitable control circuit for the converter of the invention will be straightforward for a person skilled in the art to implement in view of what is disclosed herein.

REFERENCES

• [1] Choi D. K. et al A novel power conversion circuit for cost effective battery fuel cell hybrid system. Elsevier Journal of Power Sources 152, 245-55 (2005)
• [2] Heinemann L. Analysis and design of a modular, high power converter with high efficiency for electrical power distribution systems.” IEEE PESO 2, 713-718 (2002)
• [3] Jovcic D. Off Shore Wind Farm with a Series Multiterminal CSI HVDC. Electric Power Systems Research, Elsevier 78, issue 4, 747-55 (2008)
• [4] Flourentzou N. et al VSC-Based HVDC Power Transmission Systems: An Overview. IEEE Transactions on Power Electronics 24, issue 3, 592-602
• [5] Mohan N. et al Power Electronics Converters, Applications and Design. John Wiley & Sons, (1995)
• [6] Zhao Q. et al High Efficiency, high step up DC-DC converters. IEEE Transactions on Power Electronics, 18, no. 1, 65-73 (2003)
• [7] Hirachi K. et al Circuit configuration of bi-directional DC/DC converter specific for small scale load leveling system. Proc IEE Power conversion conf, 603-609 (2002)
• [8] Abutbul, O. et al Step-up Switching Mode Converter With High Voltage Gain Using a Switched-Capacitor Circuit. IEEE Transactions On Circuit and Systems-I 50 no. 8, 1098-2002 (2003)
• [9] U.S. Pat. No. 5,426,350
• [10] Borage M. et al Analysis and design of LCL-T resonant converter as a constant-current power supply. IEEE Trans. Industrial Electronics 52, no. 6, 1547-54, (2005)
• [11] Dieckerhoff S. et al “Design of an IGBT-based LCL-Resonant Inverter for High-Frequency Induction Heating.” IEEE Industry Applications Conf, 34^th IAS Annual Meeting 3, (1999)
• [12] Uovcic D. Bidirectional high power DC transformer. IEEE Transactions on Power Delivery 24, issue 4, 2276-83 (2009)
• [13] Uovcic D. Step-up DC-DC converter for MW size Applications. IET Power Electronics 2, issue 6, 675-685 (2009)
• [14] WO 2010/029345 A1
• [15] WO 2009/074820 A1
• [16] Jovcic D. Phase Locked Loop System for FACTS. IEEE Transactions on Power Syst 18, no. 3, 1116-24 (2003)
• [17] Manitoba HVDC Research Center PSCAD/EMTDC users' manual. Winnipeg (2003)

Claims

1. A converter for transferring power between a first DC system of DC voltage V1 and a second DC system of DC voltage V2, the converter comprising: — a first AC/DC converter for transforming DC voltage V1 into a first single phase AC voltage V1ac, of frequency ω, root mean square line-neutral magnitude V1acm and angle α1;a second AC/DC converter for transforming DC voltage V2 into a second single phase AC voltage V2ac, of frequency ω, root mean square line-neutral magnitude V2acm and angle α2; andtwo inductors L1,L2 and a capacitor C, wherein the first terminals of the inductors and capacitor are connected together, the second terminal of inductor L1 and the second terminal of the capacitor C are connected to the first AC voltage V1ac, and the second terminal of inductor L2 and the second terminal of the capacitor C are connected to the second AC voltage V2ac;wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to enable required power transfer and tominimise current in inductor L1,and/or minimise current in inductor L2.

2-11. (canceled)

12. A converter according to claim 1 wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to satisfy: k1=1−ω2L2C k2=1−ω2L1C k3=L1+L2−ω2L1L2C and where k1 is selected in the region:

13. A converter according to claim 1 wherein a difference between the AC voltage angles α1,α2 (or α1i,α2j or α1,α2j) of substantially

14. A converter according to claim 1 comprising control means for controlling power flow though the converter, wherein said control means comprises: — means for varying AC voltage magnitude Vacm by changing the width of the AC pulses;varying the angle difference between the AC sources (α1−α2); and/orvarying the operating frequency ω of the converter.

15. A converter for transferring power between a first DC system of DC voltage V1 and a second DC system of DC voltage V2, the converter comprising: — a first AC/DC converter for transforming DC voltage V1 into a first p phase AC voltage V1ac of frequency ω, root mean square line-neutral magnitude V1acm and angle α1;a second AC/DC converter for transforming DC voltage V2 into a second p phase AC voltage V2ac, of frequency ω, root mean square line-neutral magnitude V2acm and angle α2; and a LCL circuit for each phase p, wherein each LCL circuit comprises two inductors L1,L2 and a capacitor C connected together at their first terminals, the second terminal of inductor L1 being connected to the respective phase of the first AC voltage V1ac, and the second terminal of inductor L2 being connected to the respective phase of the second AC voltage V2ac, wherein the second terminals of all capacitors are connected to a common point (star point) or they can be connected in delta between other phases;wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to enable required power transfer and tominimise current in inductor L1,and/or minimise current in inductor L2.

16. A converter according to claim 15 wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to satisfy: k1=1−ω2L2C k2=1−ω2L1C k3=L1+L2−ω2L1L2C and where k1 is selected in the region:

17. A converter according to claim 15 wherein a difference between the AC voltage angles α1,α2 (or α1i,α2j or α1,α2j) of substantially

18. A converter according to claim 15 wherein p=1, such that the AC voltages V1ac and V2ac (or V1iac and V2jac) are single phase AC-voltages.

19. A converter according to claim 15 comprising control means for controlling power flow though the converter, wherein said control means comprises: — means for varying AC voltage magnitude Vacm by changing the width of the AC pulses;varying the angle difference between the AC sources (α1−α2); and/orvarying the operating frequency ω of the converter.

20. A converter according to claim 15 wherein p>1 and the second terminals of the capacitors for all phases are connected to a common point.

21. A converter for transferring power between n DC systems respectively of DC voltage V1i(i=1, 2, . . . n) and m DC systems respectively of DC voltage V2j(j=1, 2, . . . m), where n and m are both any positive integer, the converter comprising: — n AC/DC converters for respectively transforming DC voltages V1i into respective p phase AC voltages V1iac of frequency ω, root mean square line-neutral magnitude V1iacm and angle α1i;m AC/DC converters for respectively transforming DC voltages V2j into respective p phase AC voltages V2jac, of frequency ω, root mean square line-neutral magnitude V2jacm and angle α2j; anda LCL circuit for each phase p, wherein each LCL circuit comprises n inductors L1i, m inductors L2j and a capacitor C connected together at their first terminals, the second terminal of each inductor L1i being connected to the respective phase of the respective AC voltage V1iac, and the second terminal of each inductor L2j being connected to the respective phase of the respective AC voltage V2jac, wherein the second terminals of all capacitors are connected to a common point (star point) or they can be connected in delta between other phases;wherein the value of the capacitor C, each inductor L1i and each inductor L2j are selected to enable required power transfer and tominimise current in inductors L1i,and/or minimise currents in inductors L2j.

22. A converter according to claim 21 wherein a difference between the AC voltage angles α1,α2 (or α1i,α2j or α1,α2j) of substantially

23. A converter according to claim 21 wherein p=1, such that the AC voltages V1ac and V2ac (or V1iac and V2jac) are single phase AC-voltages.

24. A converter according to claim 21 comprising control means for controlling power flow though the converter, wherein said control means comprises: — means for varying AC voltage magnitude Vacm by changing the width of the AC pulses;varying the angle difference between the AC sources (α1−α2); and/orvarying the operating frequency ω of the converter.

25. A converter according to claim 21 wherein p>1 and the second terminals of the capacitors for all phases are connected to a common point.

26. A converter for transferring power between a p phase AC system of AC voltage V1ac of frequency ω, root mean square line-neutral magnitude V1acm and angle α1, and a DC system of DC voltage V2, the converter comprising: — an AC/DC converter for transforming DC voltage V2 into a p phase AC voltage V2ac, of frequency ω, root mean square line-neutral magnitude V2acm and angle α2; anda LCL circuit for each phase p, wherein each LCL circuit comprises two inductors L1,L2 and a capacitor C connected together at their first terminals, the second terminal of inductor L1 being connected to the respective phase of AC voltage V1ac, and the second terminal of inductor L2 being connected to the respective phase of AC voltage V2ac, wherein the second terminals of all capacitors are connected to a common point (star point) or they can be connected in delta between other phases;wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to enable required power transfer and tominimise current in inductor L1,and/or minimise current in inductor L2.

27. A converter according to claim 26 wherein wherein the value of the capacitor C, inductor L1 and inductor L2 are selected to satisfy: k1=1−ω2L2C k2=1−ω2L1C k3=L1+L2−ω2L1L2C and where k1 is selected in the region:

28. A converter according to claim 26 wherein a difference between the AC voltage angles α1,α2 (or α1i,α2j or α1,α2j) of substantially

29. A converter according to claim 26 wherein p=1, such that the AC voltages V1ac and V2ac (or V1iac and V2jac) are single phase AC-voltages.

30. A converter according to claim 26 comprising control means for controlling power flow though the converter, wherein said control means comprises: — means for varying AC voltage magnitude Vacm by changing the width of the AC pulses;varying the angle difference between the AC sources (α1−α2); and/orvarying the operating frequency ω of the converter.

31. A converter according to claim 26 wherein p>1 and the second terminals of the capacitors for all phases are connected to a common point.

32. A converter for transferring power between a p phase AC system of AC voltage V1ac of frequency ω, root mean square line-neutral magnitude V1acm, and m DC systems respectively of DC voltage V2i(i=1,2, . . . m), where m is any positive integer, the converter comprising: — m AC/DC converters for respectively transforming DC voltages V2i into respective p phase AC voltages V2iac, of frequency ω, root mean square line-neutral magnitude V2iacm and angle α2j; anda LCL circuit for each phase p, wherein each LCL circuit comprises an inductor L1, m inductors L2j and a capacitor C connected together at their first terminals, the second terminal of each inductor L1 being connected to the respective phase of the respective AC voltage V1ac, and the second terminal of each inductor L2j being connected to the respective phase of the respective AC voltage V2jac, wherein the second terminals of all capacitors are connected to a common point (star point) or they can be connected in delta between other phases;wherein the value of the capacitor C, inductor L1 and each inductor L2j are selected to enable required power transfer and tominimise current in inductor L1,and/or minimise currents in inductors L2j.

33. A converter according to claim 32 wherein a difference between the AC voltage angles α1,α2 (or α1i,α2j or α1,α2j) of substantially

34. A converter according to claim 32 wherein p=1, such that the AC voltages V1ac and V2ac (or V1iac and V2jac) are single phase AC-voltages.

35. A converter according to claim 32 comprising control means for controlling power flow though the converter, wherein said control means comprises: — means for varying AC voltage magnitude Vacm by changing the width of the AC pulses;varying the angle difference between the AC sources (α1−α2); and/orvarying the operating frequency ω of the converter.

36. A converter according to claim 32 wherein p>1 and the second terminals of the capacitors for all phases are connected to a common point.