METHOD FOR OPERATING A CONVERTER, CONVERTER AND COMPUTER PROGRAM PRODUCT

Abstract

A method is configured for operating a converter (10) which is implemented as a modular-multilevel converter and comprises a control arrangement (38) and a number M of phase-legs (21 to 29). The method comprises detecting whether the converter (10) has to be set into one mode of a group comprising a static synchronous compensator mode or a grid unbalance mode, generating mode control signals (MCS) depending on the detected mode, generating balance voltage reference signals (u.sub.bal,ref) depending on a first side frequency (?.sub.g), a second side frequency (?m), second side current reference signals (im,ref) and the mode control signals (MCS), generating a phase-leg control signal (u.sub.ref), generating cell control signals (51 to S4) and providing the cell control signals (51 to S4) to semiconductor switches (41 to 44) of cells (31) of the phase-legs (21 to 29).

Claims

1-15. (canceled)

16. A method for operating a converter, wherein the converter is implemented as a modular-multilevel converter and comprises a control arrangement and a first side, a second side and a number M of phase-legs which are arranged between the first side and the second side, wherein each of the number M of phase-legs comprises at least a cell with a capacitor and semiconductor switches, wherein the converter is realized as a direct AC/AC converter, and wherein each of the number M of phase-legs comprises a number N of cells which are realized as full bridge cells, the method comprising: detecting whether the converter has to be set into one mode of a group comprising: a static synchronous compensator mode in case a voltage of at least one terminal on the first side of the converter has constantly zero Volt and/or a machine coupled to the second side of the converter has a standstill, or a grid unbalance mode, generating mode control signals (MCS) depending on the detected mode, generating balance voltage reference signals (u.sub.bal,ref) depending on a first side frequency (?.sub.g), a second side frequency (?.sub.m), second side current reference signals (i.sub.m,ref), measured phase-leg capacitor signals (u.sub.C?,yxn) and the mode control signals (MCS) by the control arrangement, generating a phase-leg control signal (u.sub.ref) for each of the number M of phase-legs depending on first side voltage reference signals (u.sub.g,conv,ref), second side voltage reference signals (u.sub.m,conv,ref) and the balance voltage reference signals (u.sub.bal,ref) by a reference generator of the control arrangement, generating cell control signals by a modulator of the control arrangement as a function of the phase-leg control signals (u.sub.ref), and providing the cell control signals to the semiconductor switches of the cells.

17. The method of claim 16, wherein the method comprises: generating first side current reference signals (i.sub.g,ref) depending on the mode control signals (MCS) by a first side control module of the control arrangement, and generating the balance voltage reference signals (u.sub.bal,ref) by a current control module of an inner control module of the control arrangement depending on the first side current reference signals (i.sub.g,ref), the second side current reference signals (i.sub.m,ref) and current balance reference signals (i.sub.bal,ref).

18. The method of claim 16, wherein the method comprises generating the balance voltage reference signals (u.sub.bal,ref) by a current control module of the inner control module of the control arrangement depending on first side current reference signals (i.sub.g,ref), the second side current reference signals (i.sub.m,ref), the mode control signals (MCS) and the current balance reference signals (i.sub.bal,ref).

19. The method of claim 16, wherein the method comprises: generating the current balance reference signals (i.sub.bal,ref) by a decoupling matrix module of the inner control module depending on current balance signals (i.sub.bal), generating the current balance signals (i.sub.bal) by a voltage control module of the inner control module depending on clock signals (u.sub.pll) and measured phase-leg capacitor signals (u.sub.C?,yxn) measured at the cells of the number M of phase-legs, and generating the clock signals (u.sub.pll) as a function of the first side frequency (?.sub.g) and the second side frequency (?.sub.m) by the inner module.

20. The method of claim 16, wherein in case the converter is set to the static synchronous compensator mode, the mode control signals (MCS) are generated as a function of a first, a second and a third evaluated current balance signal (i.sub.bal,e1, i.sub.bal,e2, i.sub.bal,e3), wherein the first evaluated current balance signal (i.sub.bal,e1) depends on current balance signals (i.sub.bal) of phase-legs which are connected to a first terminal (a) of the first side of the converter, wherein the second evaluated current balance signal (i.sub.bal,e2) depends on current balance signals (i.sub.bal) of phase-legs which are connected to a second terminal (b) of the first side of the converter, and wherein the third evaluated current balance signal (i.sub.bal,e3) depends on current balance signals (i.sub.bal) of phase-legs which are connected to a third terminal (c) of the first side of the converter.

21. The method of claim 20, wherein the method comprises: generating the mode control signals (MCS) as a function of rescaled signals and clock signals (u.sub.pll) by a multiplication module of the rescale module, and generating the clock signals (u.sub.pll) as a function of the first side frequency (?.sub.g) and the second side frequency (?.sub.m) by the inner control module.

22. The method of one of claim 16, wherein in case the converter is set to the static synchronous compensator mode, the mode control signals (MCS) are generated as a function of common mode voltage signals, and wherein the common mode voltage signals are generated by a processing module of the control arrangement as a function of measured signals and comprise a common mode voltage amplitude (?.sub.G.sup.0) and a common mode voltage angle (?.sub.uG.sup.0).

23. The method of claim 22, wherein the method comprises calculating a value of the common mode voltage angle (?.sub.uG.sup.0) and a value of the common mode voltage amplitude (?.sub.G.sup.0) by: $?_{u G}^{0} = a \tan \frac{k_{6} k_{1} - k_{3} k_{4}}{k_{3} k_{5} - k_{6} k_{2}}$ ${\hat{u}}_{G}^{0} = {\begin{matrix} - \frac{k_{3}}{k_{1} \cos ?_{u G}^{0} + k_{2} \sin ?_{u G}^{0}} \\ or - \frac{k_{6}}{k_{4} \cos ?_{u G}^{0} + k_{5} \sin ?_{u G}^{0}} \end{matrix}$ wherein k1 to k6 are values of a first to a sixth parameter which are calculated as a function of the measured signals.

24. The method of claim 20, wherein in case the converter is set to the static synchronous compensator mode, the mode control signals (MCS) are generated as a function of: rescaled signals and common mode voltage signals, with a weighting factor.

25. The method claim 16, wherein in case the converter is set to the grid unbalance mode, the mode control signals (MCS) are generated as a function of a first and a second circulating current signal.

26. The method of claim 25, wherein the method comprises calculating the first circulating current signal and the second circulating current signal by: ${\hat{i}}_{? ?}^{M} = {\hat{i}}_{? ?}^{M} = - \frac{{\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{P S} \cos (?_{i G}^{N S} - ?_{u G}^{P S}) + {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{N S} \cos (?_{i G}^{P S} - ?_{u G}^{N S})}{2 {\overset{?}{u}}_{M}}$ ${\hat{i}}_{? ?}^{M} = {\hat{i}}_{? ?}^{M} = \frac{{\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{P S} \sin (?_{i G}^{N S} - ?_{u G}^{P S}) - {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{N S} \sin (?_{i G}^{P S} - ?_{u G}^{N S})}{2 {\overset{?}{u}}_{M}}$ $?_{? ?}^{M} = ?_{u M} + ?$ $?_{? ?}^{M} = ?_{u M} + 3 ? / 2$ $?_{? ?}^{M} = ?_{u M}$ $?_{? ?}^{M} = ?_{u M} + ? / 2$ wherein ?.sub.G.sup.NS is a value of a negative-sequence grid current magnitude, ?.sub.G.sup.PS is a value of a positive-sequence grid voltage magnitude, ?.sub.iG.sup.NS is a value of a negative-sequence grid current angle, ?.sub.uG.sup.PS is a value of a positive-sequence grid voltage angle, ?.sub.G.sup.PS is a value of a positive-sequence grid current magnitude, ?.sub.G.sup.NS is a value of a negative-sequence grid voltage magnitude, ?.sub.iG.sup.PS is a value of a positive-sequence grid current angle, ?.sub.uG.sup.NS is a value of a negative-sequence grid voltage angle, ?.sub.??.sup.M is a value of a first circulating current amplitude of the first circulating current signal, ?.sub.??.sup.M is a value of a second circulating current amplitude of the second circulating current signal, ?.sub.??.sup.M is a value of a first circulating current angle of the first circulating current signal, ?.sub.??.sup.M is a value of a third circulating current angle, ?.sub.??.sup.M is a value of a second circulating current angle of the second circulating current signal, ?.sub.??.sup.M is a value of a fourth circulating current angle and ?.sub.uM is a value of a second side voltage magnitude.

27. The method of claim 24, wherein in case the converter is set to the grid unbalance mode, the mode control signals (MCS) are generated as a function of: the first and the second circulating current signal and the rescaled signals and/or the common mode voltage signals, with a weighting factor.

28. A converter, comprising: a control arrangement and a number M of phase-legs, wherein each of the number M of phase-legs comprises at least a cell with a capacitor and semiconductor switches, wherein the converter is realized as a modular-multilevel converter and is configured to executedetecting whether the converter has to be set into one mode of a group comprising: a static synchronous compensator mode in case a voltage of at least one terminal on the first side of the converter has constantly zero Volt and/or a machine coupled to the second side of the converter has a standstill, or a grid unbalance mode, generating mode control signals (MCS) depending on the detected mode, generating balance voltage reference signals (u.sub.bal,ref) depending on a first side frequency (?.sub.g), a second side frequency (?.sub.m), second side current reference signals (i.sub.m,ref), measured phase-leg capacitor signals (u.sub.C?,yxn) and the mode control signals (MCS) by the control arrangement, generating a phase-leg control signal (u.sub.ref) for each of the number M of phase-legs depending on first side voltage reference signals (u.sub.g,conv,ref), second side voltage reference signals (u.sub.m,conv,ref) and the balance voltage reference signals (u.sub.bal,ref) by a reference generator of the control arrangement, generating cell control signals by a modulator of the control arrangement as a function of the phase-leg control signals (u.sub.ref), and providing the cell control signals to the semiconductor switches of the cells.

29. A computer program product, comprising instructions to cause a converter to execute: detecting whether the converter has to be set into one mode of a group comprising: a static synchronous compensator mode in case a voltage of at least one terminal on the first side of the converter has constantly zero Volt and/or a machine coupled to the second side of the converter has a standstill, or a grid unbalance mode, generating mode control signals (MCS) depending on the detected mode, generating balance voltage reference signals (u.sub.bal,ref) depending on a first side frequency (?.sub.g), a second side frequency (?.sub.m), second side current reference signals (i.sub.m,ref), measured 130275-0111US01/7227454.2 7 phase-leg capacitor signals (u.sub.C?,yxn) and the mode control signals (MCS) by the control arrangement, generating a phase-leg control signal (u.sub.ref) for each of the number M of phase-legs depending on first side voltage reference signals (u.sub.g,conv,ref), second side voltage reference signals (u.sub.m,conv,ref) and the balance voltage reference signals (u.sub.bal,ref) by a reference generator of the control arrangement, generating cell control signals by a modulator of the control arrangement as a function of the phase-leg control signals (u.sub.ref), and providing the cell control signals to the semiconductor switches of the cells.

30. The method of claim 22, wherein in case the converter is set to the static synchronous compensator mode, the mode control signals (MCS) are generated as a function of: rescaled signals and the common mode voltage signals, with a weighting factor.

31. The method of claim 23, wherein in case the converter is set to the grid unbalance mode, the mode control signals (MCS) are generated as a function of: the first and the second circulating current signal and rescaled signals and/or the common mode voltage signals, with a weighting factor.

32. The method of claim 26, wherein in case the converter is set to the grid unbalance mode, the mode control signals (MCS) are generated as a function of: the first and the second circulating current signal and rescaled signals and/or common mode voltage signals, with a weighting factor.

Description

[0062] The following description of Figures of examples or embodiments may further illustrate and explain aspects of the converter and the method for voltage conversion. Arrangements, devices and modules with the same structure and the same effect, respectively, appear with equivalent reference symbols. In so far as arrangements, devices and modules correspond to one another in terms of their function in different Figures, the description thereof is not repeated for each of the following Figures.

[0063] FIGS. 1 and 2 show an exemplary embodiment of a converter and of details of a converter;

[0064] FIGS. 3 to 5 show an exemplary embodiment of a control arrangement of a converter and details of a control arrangement;

[0065] FIGS. 6 and 7 show exemplary embodiments of a control arrangement of a converter and of signals of a converter; and

[0066] FIGS. 8A to 8D, 9A and 9B show exemplary embodiments of details of a control arrangement of a converter and of signals.

[0067] FIG. 1 shows an exemplary embodiment of a converter 10 that is realized as an AC/AC converter. The converter 10 is implemented as a direct AC/AC converter. The converter 10 has a first side 11 and a second side 12. The first side 11 includes three terminals a, b, c for three phases. The second side 12 also includes three terminals 1, 2, 3 for three phases. The converter 10 is realized as modular-multilevel converter, abbreviated as MMC. The first side 11 of the converter 10 is coupled to a grid 13. A transformer 14 couples the first side 11 of the converter 10 to the grid 13.

[0068] The second side 12 of the converter 10 is coupled to a machine 15. The machine 15 e.g. includes a generator 16 and a turbine 17. The generator 16 and the turbine 17 are mechanically connected to each other via a shaft. A first side of the generator 16 is connected to the second side 12 of the converter 10. Optionally, a machine disconnector 20 couples the first side of the generator 16 to the second side 12 of the converter 10. A second side of the generator 16 is coupled to an earth terminal 18. In impedance 19 may be arranged between the second side of the generator 16 and the earth terminal 18. An arrangement 9 or system comprises the converter 10, the grid 13, the transformer 14 and the machine 15. The converter 11 includes a control arrangement 38 coupled via not-shown connection lines and/or wireless to the phase-legs 21 to 29 and the other parts of the arrangement.

[0069] The converter 10 comprises a number M of phase-legs 21 to 29. In FIG. 1, the number M is nine. The phase-legs 21 to 29 can be called legs, branches or arms. A first to a third phase-leg 21 to 23 couples a first terminal 1 of the second side 12 to a first, second and third terminal a, b, c of the first side 11. Correspondingly, a fourth to a sixth phase-leg 24 to 26 couples a second terminal 2 of the second side 12 to the first, second and third terminal a, b, c of the first side 11. Similarly, a seventh to a ninth phase-leg 27 to 29 couples a third terminal 3 of the second side 12 to the first, second and third terminal a, b, c of the first side 11. The first to the third terminal a, b, c of the first side 11 can be named grid terminals. The first to the third terminal 1, 2, 3 on the second side 12 can be named motor terminals.

[0070] Each of the phase-legs 21 to 29 comprises a number N of cells 31 to 36. In the example shown in FIG. 1, the number N is 6. However, the number N may also be 2, 3, 4, 5 or higher than 6. The number N of cells 31 to 36 of a phase-leg are e.g. connected in series. Additionally, each of the phase-legs 21 to 29 comprise a phase-leg reactor 37 that is connected in series to the number N of cells 31 to 36. Each of the phase-legs 21 to 29 has e.g. the same structure as a first phase-leg 21. A first to a ninth phase-leg voltage u1 to u9 can be tapped across the first to the ninth phase-leg 21 to 29. Measured phase-leg capacitor signals u.sub.C?,yxn are typically different from the first to the ninth phase-leg voltage u1 to u9. The phase-leg reactor 37 can be realized as inductor, inductance, coil or impedance.

[0071] In an example, the arrangement 9 is configured for a pumped hydro installation with a nominal power of e.g. 80 MW and a variable motor frequency of e.g. around ? of the nominal grid frequency. The machine 15 is realized e.g. as a synchronous machine, which may feature higher flexibility and efficiency compared to a doubly fed induction machine (abbreviated DFIM). CFSM which is the abbreviation for converter fed synchronous machine means a converter is processing the whole power flow from the synchronous machine, as opposed to a DFIG, where the converter is placed to supply the rotor of an asynchronous machine. The configuration of the direct MMC is shown in FIG. 1. The arrangement 9 is implemented as a direct AC/AC MMC with three-phase grid connection on the first side 11 and a synchronous machine 15 on the second side 12, connected with a pump/turbine 17 by a mechanical shaft. The grounding of the converter 10 can possibly be made with the impedance 19 connected to the machine stator midpoint as shown here, but this is not the only option. The impedance 19 may include at least one of a resistor, resistors, inductor and inductors. The impedance 19 may be realized as a resistive and/or an inductive impedance.

[0072] The first to third terminal x={a,b,c} on the first side 11 which may be named grids terminals are connected to the first to third terminal y={1,2,3} on the second side 12 which may be named motor terminals via the number M of phase-legs 21 to 29 which can be named branches. Each phase-leg 21 to 29 comprises a string of cells 31 to 36 plus the phase-leg impedance 37. Each cell 31 to 36 is a bipolar cell (full-bridge), since the phase-leg voltages u1 to u9 have a dual frequency ac waveform.

[0073] A possible goal of machine standstill is to remain in converter ready state (waiting to start pump or turbine operation) while maintaining the cells 31 to 36 charged, while a goal in STATCOM mode is to provide ancillary services to the three-phase grid 13. Since in an example the arrangement 9 does not to have the machine disconnector 20, the arrangement 9 cannot be reduced to three independent Y-STATCOM converters during the operation. For the Y-STATCOM, either negative sequence currents or zero sequence voltage may typically be used to perform the energy balancing control. As the voltage at the machine terminal is zero (for example at one, two or three of the first to the third terminals 1, 2, 3), the degrees of freedom for balancing are strongly limited. If no suitable control action is performed, the average phase-leg capacitor voltages diverge and the robustness against grid faults and the STATCOM operation is compromised due to a reduced margin with respect to the cell over-voltage (abbreviated OV) trip level and/or the cell under-voltage (abbreviated UV) trip level.

[0074] The first to third terminal on the first side 11 of the converter 11 (being the grid terminals, namely the terminals on the transformer secondary side) are defined with the index x={a,b,c}. The first to third terminal on the second side 12 (being the machine terminals) are defined with the index y={1,2,3}. The phase-legs indices are [1a,1b,1c,2a,2b,2c,3a,3b,3c]. When one terminal voltage is zero, paralleled phase-legs (1a-2a-3a/1b-2b-3b/1c-2c-3c) can be balanced. For this purpose, a special control method is employed, which is described below. However, balancing different sets of phase-legs (e.g. transferring energy from a <->b, b<->c, c<->a) is not possible.

[0075] FIG. 2 shows an example of a cell 31 that can be used in the embodiment of the converter 10 shown in FIG. 1. The cell 31 is realized as full bridge cell. In an example, the number N of cells 31 to 36 of the first phase-leg 31 is realized such as the cell 31. The cell 31 comprises four semiconductor switches 41 to 44. In an example, the semiconductor switches 41 to 44 are realized as insulated-gate bipolar transistors (abbreviated IGBTs), integrated gate commutated thyristors (IGCTs), gate turn-off thyristors (abbreviated GTOs) and/or metal-oxide-semiconductor field-effect transistors (abbreviated MOSFETs). The semiconductor switches 41 to 44 are configured to be able to interrupt the current flowing. The cell 31 comprises four diodes 45 to 48 which are coupled to the four semiconductor switches 41 to 44. The four diodes 45 to 48 operate as freewheeling diodes. The four diodes 45 to 48 are connected as anti-parallel diodes to the four semiconductor switches 41 to 44. Thus, for example, the diode 45 is anti-parallel diode to the semiconductor switch 41 etc. The cell 31 includes a storage arrangement 49. The storage arrangement 49 comprises e.g. a capacitor 50. A cell voltage uC (which may be named capacitor voltage) is stored by the storage arrangement 49. Control terminals of the semiconductor switches 41 to 44 are coupled to the control arrangement 38 (shown in FIG. 1A). A phase-leg current i flows through the cell 31. The phase-leg current i might flow through the cell capacitor 50 with the same or opposite sign, or only flows through the semiconductor switches 41 to 44.

[0076] In an example, a measured signal is determined by measurements of the current i, the cell voltage uC and/or the first phase-leg voltage u1 by current sensors and/or voltage sensors (not shown).

[0077] The cell 31 is realized as bipolar cell, i.e. a voltage across the cell terminals can be +uC,0,?uC. In an example, the cell 31 could provide more than three voltage levels (e.g. +uC,+uC/2,0,?uC/2,?uC).

[0078] In order to modify on the converter energy balance by shifting energy from one phase-leg 21 to 29 to another phase-leg, internal currents (commonly named circulating currents) which are neither visible at the terminals of the grid 13 nor the terminals of the machine 15, shall be used.

[0079] Mathematically speaking, that is not the only option: [0080] circulating currents interact with terminal voltages (determined for a given energy transfer) and create dc (plus ac harmonics of less relevance since zero mean) power components if they have the same frequency used to make an energy transfer between phase-legs (reduce energy stored in cell capacitors in one or more phase-legs and increase energy stored in cell capacitors in one or more phase-legs). [0081] common mode voltage interacts with terminal currents (determined for a given energy transfer) and create dc (plus ac harmonics of less relevance since zero mean) power components if they have the same frequency used to make an energy transfer between phase-legs (reduce energy stored in cell capacitors in one or more phase-legs and increase energy stored in cell capacitors in one or more phase-legs).

[0082] If the topology does not allow for circulating currents, e.g. a Y-STATCOM, other means are required, such as common mode voltage (if terminal currents are sufficiently large) or manipulating the terminal currents otherwise.

[0083] The balancing control for a wide class of MMCs is presented in FIG. 3. The control can be divided into sub-categories or modules as illustrated in FIGS. 3 and 4. The control concept for the energy balancing at machine standstill and other modes shown in FIGS. 5 to 9 extends the concepts shown in FIGS. 3 and 4.

[0084] FIG. 3 shows an exemplary embodiment of a control arrangement 38 of a converter 10 which is a further development of the embodiments shown above. In FIG. 3, control of a direct 3-phase/3-phase MMC for a pumped-storage hydro power plant is elucidated, wherein an inner control module 61 is realized e.g. according to FIG. 4.

[0085] The control arrangement 38 at least realizes the following modules: The inner control module 61 is coupled to an input side of a reference generator 62. The reference generator 62 is coupled to an input side of a modulator 63. The inner control module 61 includes a voltage control module 64 and a current control module 65. An output of the voltage control module 64 is coupled to an input of the current control module 65. The output of the current control module 65 is coupled to the input side of the reference generator 62.

[0086] A first side control module 66 is coupled on its output side to the inner control module 61 and the reference generator 62. More specifically, the first side control module 66 is coupled to the voltage control module 64 and to the current control module 65. The first side control module 66 can also be named grid control module or grid side current controller.

[0087] Correspondingly, a second side control module 67 is coupled on its output side to the inner control module 61 and the reference generator 62. More specifically, the second side control module 67 is coupled to the voltage control module 64 and to the current control module 65. The second side control module 66 can also be named machine control module or machine current controller.

[0088] The modulator 63 generates cell control signals S1 to S4 for the number M of phase-legs 21 to 29 as a function of phase-leg control signals uref More specifically, different cell control signals S1 to S4 are provided to the semiconductor switches 41 to 44 of each cell of the number N of cells 31 to 36 of each phase-leg of the number M of phase-legs 21 to 29. The reference generator 62 generates the phase-leg control signals u.sub.ref for each phase-leg of the number M of phase-legs 21 to 29 as a function of first side voltage reference signals u.sub.g,conv,ref, second side voltage reference signals u.sub.m,conv,ref and balance voltage reference signals u.sub.bal,ref.

[0089] Mode control signals MCS are provided to the first side control module 66 and/or to the inner control module 61, especially to the current control module 65 of the inner control module 61. Thus, in FIG. 3, the mode control signals MCS are shown twice, indicating the possible modules to which the mode control signals MCS can be applied in the different examples.

[0090] Thus, in an example, the mode control signals MCS are provided to the first side control module 66.

[0091] The current control module 65 generates the balance voltage reference signals u.sub.bal,ref as a function of first side current reference signals i.sub.g,ref, second side current reference signals i.sub.m,ref and current balance reference signals i.sub.bal,ref or current balance signals i.sub.bal. Phase-leg measured current signals i.sub.yx are provided to the current control module 65. Phase-leg measured current signals i.sub.yx represent currents in each of the nine phase-legs 21 to 29. The current control module 65 generates the balance voltage reference signals u.sub.bal,ref as a function of a comparison ofon one handthe phase-leg measured current signals i.sub.yx andon the other handof a signal gained from the first side current reference signals i.sub.g,ref, the second side current reference signals i.sub.m,ref and the current balance reference signals i.sub.bal,ref or current balance signals i.sub.bal.

[0092] The voltage control module 64 generates the current balance reference signals i.sub.bal,ref or the current balance signals i.sub.bal as a function of a first side frequency ?g, a second side frequency ?.sub.m and measured phase-leg capacitor signals u.sub.C?,yxn.

[0093] The first side control module 66 provides the first side current reference signals i.sub.g,ref depending on the mode control signals MCS and first side measured current signals i.sub.g,abc. The first side measured current signals i.sub.g,abc are measured e.g. at the first to the third terminal a, b, c of the first side 11. Furthermore, the first side control module 66 provides the first side frequency ?.sub.g and the first side voltage reference signals u.sub.g,conv,ref for example depending on the mode control signals MCS and/or the first side measured current signals i.sub.g,abc.

[0094] Thus, the mode control signals MCS is an input for the grid side control (negative-sequence currents). The first side control module 66 includes e.g. controllers with an integrator (e.g. PI controllers for proportional-integral or PR controllers for proportional-resonant). The mode control signals MCS are part of the first side current reference signals i.sub.g,ref; so there is no need for an additional input of the mode control signals MCS to the current control module 65. In case the mode control signals MCS would alternatively be applied to the current control module 65 and the current control module 65 would include proportional controllers, the mode control signals MCS would be erased.

[0095] The second side control module 67 provides the second side frequency ?.sub.m, the second side current reference signals i.sub.m,ref and the second side voltage reference signals u.sub.m,cov,ref for example depending on a second side reference frequency ?.sub.m,ref. The power control could be assigned to either of the first and/or the second side. The second side control module 67 receives second side measured current signals i.sub.m,123 which are measured at the first to the third terminal 1, 2, 3 of the second side 12 of the converter 11. Typical speed or power controls are applied. If one side performs the power control, then the opposite side performs the energy stored control. For example, the first side control module 66 or the second side control module 67 adjusts the energy stored inside the capacitor 50 of the cells 31 to 36. A value W.sub.C of the energy can be calculated using the equation:

[00003] $W_{C} = \frac{1}{2} C_{c e l l} \underset{y, x, n}{.Math.} u_{C, yxn}^{2}$

with C.sub.cell is a capacitance value of the capacitor 50 in one of the full-bridge cells 31 to 36, u.sub.Cxyn is a value of a capacitor voltage (which is equal to the cell voltage uC), x is a value of a terminal on the first side 11 (x is 1, 2 or 3), y is a value of a terminal at the second side (y is 1, 2 or 3) and n is a number of the phase-leg (n={1, . . . M}).

[0096] The control arrangement 38 may comprise one or more than one microprocessor or microcontroller. The control arrangement 38 may include one or more than one field-programmable gate array, abbreviated FPGA. For example, the inner control module 61 and the reference generator 62 may be implemented by a microcontroller or microprocessor. The modulator 63 may be realized by a microcontroller, microprocessor, FPGA or an analog circuit. The modulator 63 is configured to generate a pulse pattern. Thus, the modules of the control arrangement 38 (shown in FIGS. 3 to 5, 6, 8A, 8C and 8D etc.) are realized by at least one microcontroller or microprocessor using a computer program or several software parts forming the computer program and/or are realized by hardware performing logical operations and/or analog signal processing steps.

[0097] The measurements of the cell voltage uC, other voltages, currents and other signals are performed by current sensors, voltage sensors and other sensors (not shown), e.g. used in the field of power conversion systems. The sensors couple the terminals at the first and the second side 11, 12 and the phase-legs 21 to 29 to the control arrangement 38.

[0098] In an alternative embodiment, not shown, the current control module 65 generates the balance voltage reference signals u.sub.bal,ref as a function of first side current reference signals i.sub.g,ref, second side current reference signals i.sub.m,ref, mode control signals MCS and current balance reference signals i.sub.bal,ref or current balance signals i.sub.bal.

[0099] FIG. 4 shows an exemplary embodiment of details of a control arrangement 38 of the converter 10 which is a further development of the embodiments shown above. In FIG. 4, an example of the inner control module 61 of a direct ac/ac MMC is explained. A signal generator 74 provides clock signals u.sub.pll as a function of the first side frequency ?.sub.g and the second side frequency ?.sub.m. The signal generator 74 takes as input the frequency information ?.sub.g and ?.sub.m, e.g. already retrieved by a phase-locked loop, a power synchronization loop or another algorithm retrieving the frequency information in the grid and machine side modules. The inner control module 61 comprises a decoupling matrix module 75 that couples the voltage control module 64 to an input side of the current control module 65. The voltage control module 64 receives measured signals. The measured signals are the measured phase-leg capacitor signals u.sub.C?,yxn. The measured phase-leg capacitor signals u.sub.C?,yxn represent a number M of average capacitor voltages or a number M of signals representing average capacitor voltages. The measured phase-leg capacitor signals u.sub.C?,yxn depend on the average of the cell voltages uC of a phase-leg of the number M of phase-legs 21 to 29. Moreover, the voltage control module 64 also receives the measured phase-leg capacitor signals u.sub.C?,yxn representing an average of the first number M of average capacitor voltages. An averaging module 76 of the inner control module 61 generates the average of the first number M of the measured phase-leg capacitor signals u.sub.C?,yxn.

[0100] The inner control module 61 includes a cascaded average capacitor voltage/phase-leg current control. The decoupling matrix module 75 with a symbol Kdec or custom-character .sub.dec is configured to avoid modifying the grid and motor currents while balancing the cells, i.e. the current balance reference signal i.sub.bal,ref are internal (circulating) currents.

[0101] FIG. 5 shows an exemplary embodiment of details of a control arrangement 38 which is a further development of the embodiments shown in FIGS. 1 to 4. In FIG. 5, a first method is described which uses negative sequence grid currents. The method can be used in one of the following cases: The machine 15 has a standstill, the converter 10 is in a standby mode with machine standstill and/or the converter 10 is in STATCOM mode with machine standstill. In these cases, the machine control module 67 is inactive (u.sub.ref.sup.INU=0 and i.sub.ref.sup.INU=0). The balancing control loses some degrees of freedom (coming from the machine terminals 1, 2, 3) and an alternate method should be employed in order to continue operation without diverging cell voltages uC, if no additional measure is taken. Principle wise, it is not anymore possible to balance the converter 10 with circulating currents. Some (barely noticeable) currents should be allowed to flow on the first side 11 (named grid side), as shown in FIG. 5. Since these currents are not internal, the output is routed to the first side control module 66 (named grid-side current controller). In fact, mainly negative sequence currents are used for balancing. Note that the clock signals u.sub.PLL only contain the first side frequency ?.sub.g (named grid frequency). In FIG. 5, a modification to the balancing control to enable a stable operation in machine standstill and STATCOM modes is shown.

[0102] These balancing currents are routed to the first side control module 66, since PI controllers (which are controllers using an integrator part) are used and would cancel out any additional contribution that is not present in their references. Using the method, the measured phase-leg capacitor signals u.sub.C?,yxn are not diverging, demonstrating the effectiveness of the proposed balancing control method.

[0103] The control arrangement 38 comprises a rescale module 80 which is coupled to the input side of the current control module 65. The rescale module 80 includes an amplification limitation module 84 and a multiplication module 85 that is arranged between the amplification limitation module 84 and the input side of the current control module 65. The rescale module 80 comprises a first, second and third evaluating module 81 to 83 connected to an input side of the amplification limitation module 84. The input side of the first, second and third evaluating module 81 to 83 are coupled to the number M of phase-legs 21 to 29, e.g. to a number M of current sensors of the number M of phase-legs 21 to 29.

[0104] In case the converter 10 is set to the static synchronous compensator mode (abbreviated STATCOM mode), a number M of current balance signals i.sub.bal are applied to the first, second and third evaluating module 81 to 83. The converter 10 is set by automatically switching over to the STATCOM mode. A first evaluated current balance signal i.sub.bal,e1 is generated by the first evaluating module 81 depending on a first, fourth and seventh current balance signal i.sub.bal, e.g. by averaging these three signals. A second evaluated current balance signal i.sub.bal,e2 is generated by the second evaluating module 82 depending on a second, fifth and eighth current balance signal i.sub.bal, e.g. by averaging these three signals. A third evaluated current balance signal i.sub.bal,e3 is generated by the third evaluating module 83 depending on a third, sixth and ninth current balance signal i.sub.bal, e.g. by averaging these three signals. The first, second and third evaluated current balance signal i.sub.bal,e1, i.sub.bal,e2, i.sub.bal,e3 are modified by the amplification limitation module 84. The multiplication module 85 multiplies signals provided by the amplification limitation module 84 with the clock signals u.sub.pll. Output signals of the multiplication module 85 are rescaled signals which are fed to the current control module 65. The mode control signals MCS are realized as the rescaled signals.

[0105] FIG. 6 shows an exemplary embodiment of details of a control arrangement 38 of a converter 10 which is a further development of the embodiments shown above. In FIG. 6, a second method is described which uses a zero sequence common mode voltage injection. The method can be used in one of the following cases: The machine 15 has a standstill, the converter 10 is in a standby mode with machine standstill and/or the converter 10 is in STATCOM mode with machine standstill.

[0106] An alternative to the use of negative sequence grid currents (as shown in FIG. 5) is the use of a zero sequence common mode voltage. When the machine 15 is at standstill, there is a significant voltage reserve (about 50%) in the cells 31 to 36 (a cell can be named valve).

[0107] The control arrangement 38 comprises a processing module 90 that is coupled on its output side to the first side control module 66 and, thus, indirectly to the current control module 65. The processing module 90 provides the mode control signals MCS to the first side control module 66. The mode control signals MCS are generated as a function of common mode voltage signals. A signal generator of the control arrangement 38 e.g. assembles the magnitude and angle with the first side frequency co g (named grid frequency) for generating the mode control signals MCS. The signal generator may use equations similar to equations (6.1)-(6.4) below. The common mode voltage signals comprises a common mode voltage amplitude ?.sub.G.sup.0 and a common mode voltage angle ?.sub.uG.sup.0. The common mode voltage amplitude ?.sub.G.sup.0 and the common mode voltage angle ?.sub.uG.sup.0 are generated by the processing module 90 as a function of measured signals, as described below. The measured current and voltage signals are transformed into positive sequence signals and the negative sequence signals. The common mode voltage amplitude ?.sub.G.sup.0 and the common mode voltage angle ?.sub.uG.sup.0 are a function of the amplitudes and of the angles (phase angles) of the positive sequence signals and the negative sequence signals.

[0108] Alternatively, the mode control signals MCS are equal to the common mode voltage signals which comprise the common mode voltage amplitude ?.sub.G.sup.0 and the common mode voltage angle ?.sub.uG.sup.0.

[0109] Phases and amplitudes of positive sequence signals and of negative sequence signals are determined using measured current and voltage signals by the processing module 90. The common mode voltage amplitude ?.sub.G.sup.0 and the common mode voltage angle ?.sub.uG.sup.0 are determined using the phases and the amplitudes of the positive sequence signals and of the negative sequence signals by the processing module 90.

[0110] A thorough mathematical derivation based on the power equations lead to the desired common mode voltage expressions. For sake of brevity, only the results are presented in this document.

[00004] $\begin{matrix} \frac{4}{\sqrt{2}} p_{? 0} = - {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{0} \cos (?_{i G}^{N S} - ?_{u G}^{0}) - {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{P S} \cos (?_{i G}^{N S} - ?_{u G}^{P S}) - {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{0} \cos (?_{i G}^{P S} - ?_{u G}^{0}) - {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{N S} \cos (?_{i G}^{P S} - ?_{u G}^{N S}) & (1.) \end{matrix}$ $\begin{matrix} \frac{4}{\sqrt{2}} p_{?0} = {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{0} \sin (?_{i G}^{N S} - ?_{u G}^{0}) - {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{P S} \sin (?_{i G}^{N S} - ?_{u G}^{P S}) - {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{0} \sin (?_{i G}^{P S} - ?_{u G}^{0}) - {\hat{i}}_{G}^{P S} {\hat{u}}_{G}^{N S} \sin (?_{i G}^{P S} - ?_{u G}^{N S}) & (2.) \end{matrix}$

[0111] Note that the same solution can be found using complex analysis in abc frame starting with p= custom-character {vi*}.

[0112] FIG. 7 show exemplary embodiments of voltage and current signals of a converter 10 using the processing module 90 and the method described above, e.g. by FIG. 6. In FIG. 7, voltages u.sub.g and currents i.sub.g at the first, second and third terminal a, b, c of the first side 11 of the converter 10 are shown. FIG. 7 illustrates the second method that can be used in case of machine standstill operation in standby condition when an asymmetrical grid fault occur (|u.sub.+|=0.1pu and |u.sub.?|=0.4pu), leading to a reactive current injection of |i.sub.+|=0.3pu and |i.sub.?|=0.1pu. The algorithm defines |u.sub.0|=0.325pu and ?u.sub.0=?.

[0113] Remark: The case where |i.sub.+|=|i.sub.?| cannot be balanced.

[0114] FIG. 8A shows an exemplary embodiment of details of a control arrangement 38 of a converter 10 which is a further development of the embodiments shown above. In FIG. 8A, a third method is described which uses a circulating current injection. The method can be used in the following case: The machine 15 is spinning and the converter 10 should react on grid faults and/or unbalances.

[0115] The control arrangement 38 comprises a processing module 91 that is coupled on its output side to the inner control module 61. The processing module 91 provides the mode control signals MCS to the inner control module 61. More specifically, the processing module 91 applies the mode control signals MCS to the current control module 65. The mode control signals MCS are generated as a function of a first circulating current signal and a second circulating current signal as will be explained below.

[0116] This method improves the energy balancing during grid faults and unbalances. Due to the existence of negative sequence currents and/or voltages, the terminal powers are not anymore identical, meaning that the energy balance in the phase-legs 21 to 29 are differently affected.

[0117] The method is performed for ensuring the phase-leg power matching during grid unbalances. The method is based on a current redistribution between phase-legs connected to the same power terminal on the second side 12, namely the terminals 1, 2, 3. This can be formulated as:

[00005] $\begin{matrix} i_{y x} = - \frac{p_{M y}}{\underset{\underset{k_{G}}{?}}{.Math. p_{M i}}} i_{G} + \frac{p_{G x}}{\underset{\underset{k_{M}}{?}}{.Math. p_{G k}}} i_{M} & (3.) \end{matrix}$

[0118] In normal conditions, i.e., when both three-phase systems are balanced, k.sub.G=k.sub.M=1/3. However, the method cannot be applied when ?p.sub.Mi=0 or ?p.sub.Gk=0. This latter case is very likely to happen with voltage dips below 0.5 pu, owing that the grid codes indicate a coefficient k.sub.q=2 for the reactive current injection during faults (i.sub.qRef=k.sub.q?u), meaning that all the available current is used for reactive power. The unit pu is an expression of system quantities as fractions of a defined base unit quantity which is often used in the power systems analysis field of electrical engineering (the so-called per-unit system).

[0119] A rework of the equation (3.0) enabling its application, when the denominator of either k.sub.G or k.sub.M becomes zero, is the following:

[00006] $\begin{matrix} ? p_{G k} = p_{G k} - \frac{.Math. p_{G}}{3} .fwdarw. k_{M} = \frac{1}{3} + \frac{? p_{G k}}{p_{G, nom}}, .Math. ? p_{G k} = 0 ? p_{M i} = p_{M i} - \frac{.Math. p_{M}}{3} .fwdarw. k_{G} = \frac{1}{3} + \frac{? p_{M i}}{p_{M, nom}}, .Math. ? p_{M i} = 0 & (4.) \end{matrix}$

[0120] The equations use the following parameters: [0121] ?p.sub.Gk: active power difference for phase k on the first side 11, [0122] p.sub.Gk: active power for phase k on the first side 11, [0123] k: phase on the first side (a,b,c), [0124] ?p.sub.G/3: active power average on the first side 11 (=?.sub.kp.sub.Gk/3), [0125] p.sub.G,nom: nominal power for the first side 11 (used for normalization in per-unit system), [0126] ?p.sub.Mi: active power difference for phase i on the second side 12, [0127] p.sub.Mi: active power for phase i on the second side 12, [0128] i: phase on the first side (1,2,3), [0129] ?p.sub.M/3: active power average on the second side 12 (=?.sub.ip.sub.Mi/3), [0130] p.sub.M,nom: nominal power for the second side 12 (used for normalization in per-unit system).

[0131] The first side 11 of the converter 10 is the grid side of the converter. The second side 12 of the converter 10 is the machine side of the converter or a further grid side. Instead of having k.sub.G and k.sub.M directly expressed in function of the phase to sum of powers ratio, it is expressed in terms of deviation of the phase power difference with respect to the mean value 1/3.

[0132] FIG. 8B shows exemplary embodiments of voltage and current signals of a converter 10 as described above, e.g. in FIG. 8A. In FIG. 8B, voltages u.sub.g and currents i.sub.g at the first, second and third terminal a, b, c of the first side 11 of the converter 10 are shown. The FIG. 8B illustrate the third method with grid unbalance. The coefficient k.sub.G=[0.415 0.481 0.104] are used in order to achieve input/output phase-leg power matching.

[0133] Alternatively, a circulating current with the first side frequency ?.sub.g (grid frequency) or the second side frequency ?.sub.m (machine frequency) can be derived in order to match the phase-leg currents obtained e.g. with (4.0). In this case, the set of equations is:

[00007] $\begin{matrix} p_{??} = - \frac{{\hat{i}}_{? ?}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM})}{4} - \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})}{4} & (5.) \end{matrix}$ $p_{??} = - \frac{{\hat{i}}_{? ?}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})}{4} + \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM})}{4}$ $p_{??} = - \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM})}{4} - \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})}{4}$ $p_{??} = - \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})}{4} + \frac{{\hat{i}}_{??}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM})}{4}$ $\frac{4}{\sqrt{2}} p_{? 0} = - {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{PS} \cos (?_{i G}^{N S} - ?_{u G}^{PS}) - {\hat{i}}_{G}^{PS} {\hat{u}}_{G}^{NS} \cos (?_{i G}^{PS} - ?_{u G}^{NS}) - {\hat{i}}_{??}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM}) - {\hat{i}}_{??}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})$ $\frac{4}{\sqrt{2}} p_{?0} = - {\hat{i}}_{G}^{N S} {\hat{u}}_{G}^{PS} \sin (?_{i G}^{N S} - ?_{u G}^{PS}) - {\hat{i}}_{G}^{PS} {\hat{u}}_{G}^{NS} \sin (?_{i G}^{PS} - ?_{u G}^{NS}) - {\hat{i}}_{??}^{M} {\hat{u}}_{M} \cos (?_{??}^{M} - ?_{uM}) - {\hat{i}}_{??}^{M} {\hat{u}}_{M} \sin (?_{??}^{M} - ?_{uM})$

[0134] The circulating currents shall not affect the power transfer, i.e. p.sub.?0 and p.sub.?0 shall remain unmodified.

[0135] The circulating currents with the amplitudes i.sub.??, i.sub.??, i.sub.??, i.sub.?? and angles are defined such that they can interact with the machine voltages.

[0136] The mode control signals MSC are equal to the first circulating current signal (more specifically equal to the first circulating current amplitude ?.sub.??.sup.M and the first circulating current angle ?.sub.??.sup.M) and the second circulating current signal (more specifically equal to the second circulating current amplitude ?.sub.??.sup.M and the second circulating current angle ?.sub.??.sup.M) or are generated as a function of the first circulating current signal (more specifically as a function of the first circulating current amplitude ?.sub.??.sup.M and the first circulating current angle ?.sub.??.sup.M) and the second circulating current signal (more specifically as a function of the second circulating current amplitude ?.sub.??.sup.M and the second circulating current angle ?.sub.??.sup.M). The mode control signals are generated as a function of the first and second circulating current signal by a signal generator of the control arrangement 38.

[0137] The generation of the time-domain signals for the circulating currents follows the equations (6.1)-(6.4):

i.sub.??(t)=?.sub.??.sup.M cos(?.sub.Mt+?.sub.??.sup.M) (6.1)

i.sub.??(t)=?.sub.??.sup.M cos(?.sub.Mt+?.sub.??.sup.M) (6.2)

i.sub.??(t)=?.sub.??.sup.M cos(?.sub.Mt+?.sub.??.sup.M) (6.3)

i.sub.??(t)=?.sub.??.sup.M cos(?.sub.Mt+?.sub.??.sup.M) (6.4)

[0138] The time-domain signals are reconstructed with the magnitudes and angles ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M and the frequency ?.sub.M by a signal generator (not shown) of the control arrangement 38.

[0139] FIG. 8C shows an exemplary embodiment of details of a control arrangement 38 of a converter 10 which is a further development of the embodiments shown above. The inside of the current control module 65 is the shown in FIG. 8C, assuming that the order of the phase-legs 21 to 29 is [1a, 2a, 3a, 1b, 2b, 3b, 1c, 2c, 3c]. The current control module 65 is configured such that the mode control signal MCS provides a circulating current. In FIG. 8C, the symbol for a sum can mean either additions or subtractions of the input signals. The mode control signals MCS may have the form of current mode control signals i.sub.MCS. The current control module 65 generates the balance voltage reference signals u.sub.bal,ref as a function of a sum or a difference of the phase-leg measured current signals i.sub.yx and of other signals. The current control module 65 generates the other signals as a sum or difference of the first side current reference signals i.sub.g,ref (e.g. divided by a factor three), the second side current reference signals i.sub.m,ref (e.g. divided by a factor three), the current balance reference signals i.sub.bal,ref and the mode control signals MCS. The two steps of forming a sum or difference can be combined or can be divided in several steps. The balance voltage reference signals u.sub.bal,ref are achieved by multiplying the results of the two steps of forming a sum or difference with a multiplication factor. Thus, the current control module 65 implements a P controller.

[0140] FIG. 8D shows an exemplary embodiment of details of a control arrangement 38 of a converter 10 which is a further development of the embodiments shown above. By looking at the equation (4.0) above, one sees that k.sub.M is a vector 1?3 with indices a,b,c and k.sub.G a vector 1?3 with indices 1,2,3. Comparing with FIG. 8C, it could be seen as replacement of the mappings for the grid and machine currents that incorporates the mode control signals MCS having the form of current mode control signals i.sub.MCS. In FIG. 8D, a modified example of the current control module 65 is shown, in case of modified third method from equation (4.0). The symbol for a sum can mean either additions or subtractions of the input signals. The corresponding circulating currents with the first to the fourth circulating current amplitudes i.sub.??, i.sub.??, i.sub.??, i.sub.?? can be defined after transformation to the same ??/??/??/?? frame as the difference between the original grid and machine side reference currents from FIG. 8C and the grid and machine side currents from FIG. 8D. Hence, the two methods are identical.

[0141] The current control module 65 generates the balance voltage reference signals u.sub.bal,ref as a function of a sum or a difference of the phase-leg measured current signals i.sub.yx and of other signals. The current control module 65 generates the other signals as a sum or difference of the first side current reference signals i.sub.g,ref (each signal multiplied by one of three factors k.sub.G1, k.sub.G2, k.sub.G3), the second side current reference signals i.sub.m,ref (each signal multiplied by one of other three factors k.sub.Ma, k.sub.Mb, k.sub.Mc) and the current balance reference signals i.sub.bal,ref. The factors k.sub.G1, k.sub.G2, k.sub.G3 and the other factors k.sub.Ma, k.sub.Mb, k.sub.Mc are a function of the mode control signals MCS. The two steps of forming a sum or difference can be combined or can be divided in several steps. The balance voltage reference signals u.sub.bal,ref are achieved by multiplying the results of the two steps of forming a sum or difference with a multiplication factor. Thus, the current control module 65 implements a P controller.

[0142] FIGS. 9A and 9B show exemplary embodiments of current signals of a converter 10 as described above and using FIGS. 8A to 8D. In FIGS. 9A and 9B, the first to the fourth circulating current signal with the first to the fourth circulating current amplitudes ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M, ?.sub.??.sup.M are shown which are obtained with the alternate implementation of the third method for the same case as in FIG. 8A to 8E.

[0143] In a development, a transition between the methods is performed. A seaming less transition between the methods is possible, for example depending on the machine speed in case zero sequence common mode injection is one of them.

[0144] This disclosure proposes a novel balancing control method for a direct 3ph/3ph MMC in case one terminal voltage is zero using either external currents and/or zero sequence voltage. 3ph is the abbreviation for three phase.

[0145] This disclosure proposes a balancing control method for a direct 3ph/3ph MMC in case of grid unbalances using either circulating currents and/or zero sequence voltage.

[0146] The method described in this disclosure realizes e.g. at least one of the following: [0147] Control of a direct 3ph/3ph MMC with one terminal voltage equal to zero using the first method. 3ph/3ph MMC is the abbreviation for three phase/three phase modular-multilevel converter. [0148] Control of a direct 3ph/3ph MMC with one terminal voltage equal to zero using the second method. [0149] Control of a direct 3ph/3ph MMC with one terminal voltage equal to zero using a combination of the first and the second method with a weighting factor. [0150] Control of a direct 3ph/3ph MMC under grid unbalance using the third method. [0151] Control of a direct 3ph/3ph MMC under grid unbalance using a combination of the first and the third method with a weighting factor. [0152] Control of a direct 3ph/3ph MMC under grid unbalance using a combination of the second and the third method with a weighting factor (for example at low machine speed). [0153] Control of a direct 3ph/3ph MMC in case the second side 12 is a further ac system, e.g. with a different frequency with respect to the first side 11; for example, the converter 10 performs a 50 Hz-60 Hz ac/ac conversion.

[0154] The methods introduced above can be straight-forwardly implemented at a direct 3ph/n-ph MMCs, with n>3. Thus, the number of terminals and phases at the second side 12 of the converter 10 may be larger than three.

[0155] The converter 10 is implemented e.g. as a modular-multilevel converter, a matrix converter and/or a direct converter. The converter 10 can be used e.g. for a pumped hydro installation. The converter 10 can be operated in a STATCOM mode. The converter 10 e.g. implements a cascaded control, reduces a voltage ripple, provides energy balancing, provides a capacitor balancing and uses e.g. a circulating current.

[0156] Signals mentioned above include e.g. three or nine partial signals.

[0157] For example, the following signals include a number M of partial signals (namely a partial signal corresponding to each phase-leg 21 to 29): current balance signals i.sub.bal, current balance reference signals i.sub.bal,ref, balance voltage reference signals u.sub.bal,ref, phase-leg control signals u.sub.ref

[0158] For example, the following signals include a first number L1 partial signals (namely a partial signal corresponding to each terminal a, b, c at the first side 11): first side current reference signals i.sub.g,ref, first side voltage reference signals u.sub.g,conv,ref

[0159] For example, the following signals include a second number L2 partial signals (namely a partial signal corresponding to each terminal 1, 2, 3 at the second side 12): second side current reference signals i.sub.m,ref, second side voltage reference signals u.sub.m,conv,ref

[0160] Single signals are e.g. first evaluated current balance signal i.sub.bal,e1, second evaluated current balance signal i.sub.bal,e2, third evaluated current balance signal i.sub.bal,e3, first circulating current amplitude ?.sub.??.sup.M, second circulating current amplitude ?.sub.??.sup.M.

[0161] The mode control signals MCS include more than one partial signal. The clock signals u.sub.pll also include more than one partial signal.

[0162] The embodiments shown in the FIGS. 1 to 9 as stated represent exemplary embodiments of the improved converter and method for operating a converter; therefore, they do not constitute a complete list of all embodiments according to the improved converter and method for operating a converter. Actual converters and methods may vary from the embodiments shown in terms of arrangements, devices, modules, cells, phase-legs, steps and signals for example.

LIST OF REFERENCE SIGNS

[0163] 1, 2, 3 terminal [0164] 9 arrangement [0165] 10 converter [0166] 11 first side [0167] 12 second side [0168] 13 grid [0169] 14 transformer [0170] 15 machine [0171] 16 generator [0172] 17 turbine [0173] 18 earth terminal [0174] 19 impedance [0175] 20 machine disconnector [0176] 21 to 29 phase-leg [0177] 31 to 36 cell [0178] 37 phase-leg reactor [0179] 38 control arrangement [0180] 41 to 44 semiconductor switch [0181] 45 to 48 diode [0182] 49 storage arrangement [0183] 50 capacitor [0184] 61 inner control module [0185] 62 reference generator [0186] 63 modulator [0187] 64 voltage control module [0188] 65 current control module [0189] 66 first side control module [0190] 67 second side control module [0191] 74 signal generator [0192] 75 decoupling matrix module [0193] 76 averaging module [0194] 80 rescale module [0195] 81 to 83 evaluating module [0196] 84 amplification limitation module [0197] 85 multiplication module [0198] 90, 91 processing module [0199] a, b, c terminal [0200] i current [0201] i.sub.bal current balance signals [0202] i.sub.bal,e1 first evaluated current balance signal [0203] i.sub.bal,e2 second evaluated current balance signal [0204] i.sub.bal,e3 third evaluated current balance signal [0205] i.sub.bal,ref current balance reference signals [0206] i.sub.g,abc first side measured current signals [0207] i.sub.g,ref first side current reference signals [0208] i.sub.m,ref second side current reference signals [0209] i.sub.m,123 second side measured current signals [0210] i.sub.yx phase-leg measured current signals [0211] ?.sub.??.sup.M first circulating current amplitude [0212] ?.sub.??.sup.M second circulating current amplitude [0213] MCS mode control signal [0214] p.sub.ref power-reference signal [0215] S1 to S4 cell control signal [0216] t time [0217] u.sub.bal,ref balance voltage reference signals [0218] uC cell voltage [0219] u.sub.Cxyn capacitor voltage [0220] u.sub.C?,yxn measured phase-leg capacitor signals [0221] u.sub.g,conv,ref first side voltage reference signals [0222] u.sub.m,conv,ref second side voltage reference signals [0223] u.sub.pll clock signals [0224] u.sub.ref phase-leg control signals [0225] u1 to u9 phase-leg voltage [0226] ?.sub.G.sup.0 common mode voltage amplitude [0227] ?.sub.uG.sup.0 common mode voltage angle [0228] ?.sub.g first side frequency [0229] ?.sub.m second side frequency [0230] ?.sub.m,ref second side reference frequency

METHOD FOR OPERATING A CONVERTER, CONVERTER AND COMPUTER PROGRAM PRODUCT

Inventors

Cpc classification

Classification Explorer

H02M7/4835

ELECTRICITY

Classification Explorer

H02M7/797

ELECTRICITY

Classification Explorer

H02M5/271

ELECTRICITY

International classification

Classification Explorer

H02M5/27

ELECTRICITY

Classification Explorer

H02M7/797

ELECTRICITY

Abstract

Claims

Description