CONVERTERS AND METHOD FOR THEIR OPERATION
Patent Information
- Application Number
- DE502020009663
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-04-20
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2040-04-20
AI Technical Summary
Multilevel converters face challenges in minimizing electrical losses and peak current amplitudes due to limitations in existing methods that primarily focus on second-order harmonics, which do not adequately reduce losses or control branch currents effectively.
The method involves generating and controlling a fourth current harmonic within the converter, which has four times the fundamental frequency, in addition to the first and second current harmonics, to further reduce electrical losses and peak current amplitudes by adjusting the phase positions and amplitude ratios of these harmonics.
This approach reduces electrical losses and peak current amplitudes more effectively than using only first and second current harmonics, allowing for smaller semiconductor switches and easier control of branch currents, while maintaining efficient energy transfer.
Description
[0001] The invention relates to converters, in particular multilevel converters, and methods for their operation.
[0002] Multilevel converters are known, for example, from the publication "Modular Multilevel Converter: An universal concept for HVDC networks and extended DC bus applications" (R. Marquardt, The 2010 International Power Electronics Conference, 978-1-4244-5393-1 / 10, 2010 IEEE). The multilevel converter described therein comprises three parallel series circuits, whose outer terminals form the multilevel converter's DC voltage terminals. Each of the parallel series circuits comprises at least two series-connected submodules, each of which includes at least two switches and one capacitor. The voltage level at the DC voltage terminals can be adjusted by appropriately controlling the switches.
[0003] From the publication "Loss optimization of MMC by Second-order-Harmonic Circulating Injection" (Yang, Li, Li, Wang, Xu, Gou, IEEE Transactions on Power Electronics, DOI 10.1109 / TPEL.2017.2751068), a method with the features according to the preamble of patent claim 1 is known. In the known method, with the aim of reducing electrical losses in the converter, an additional circulating current is generated that flows only within the converter. This circulating current has twice the frequency - relative to the fundamental frequency of the converter - and is subsequently referred to as the second current harmonic.
[0004] A method for operating a multilevel converter is known from WO 2013 / 071947 A1. In this known method, a control device controls switching modules of the converter in such a way that at least one additional circuit current harmonic flows in each series circuit of the converter. The invention is based on the object of further improving a method of the type described.
[0005] This object is achieved according to the invention by a method having the features according to claim 1. Advantageous embodiments of the method according to the invention are specified in subclaims.
[0006] According to the invention, by controlling the sub-modules, an additional circulating current flowing within the converter - hereinafter referred to as the fourth current harmonic - is generated, which has four times the fundamental frequency.
[0007] A key advantage of the method according to the invention is that – compared to applying only the second current harmonic – the additional application of the fourth current harmonic allows for even further reduction of electrical losses. Alternatively or additionally, it is possible to reduce the peak current in the converter or the maximum amplitude of the branch currents using the fourth current harmonic.
[0008] It is advantageous if the electrical losses of the converter are reduced by the fourth current harmonic to a level that is smaller than when the converter is operated with the first and second current harmonics alone.
[0009] By controlling the sub-modules, the phase position of the second current harmonic is preferably kept constant relative to the phase of the first current harmonic.
[0010] Preferably, the control of the sub-modules is carried out in such a way that the maxima of the first current harmonics each coincide with a maximum of the second current harmonic when the direct current component in the branch currents is positive, and the minima of the first current harmonics each coincide with a minimum of the second current harmonic when the direct current component in the branch currents is negative.
[0011] For a cosinusoidal fundamental oscillation or cosinusoidal first current harmonic according to Iz 1 = Az 1 cos ωt mit Az1 > 0 where Iz1 is the first current harmonic, Az1 is the amplitude of the first current harmonic, ω is the angular frequency of the current and t is the time, the following preferably applies: Iz2 t = sign Izdc ⋅ Az 2 ⋅ cos 2 ωt mit Az2 > 0 where Iz2 denotes the second current harmonic, Az2 the amplitude of the second current harmonic and Izdc the DC component in the respective branch current.
[0012] In other words, it is advantageous if the second current harmonic Iz2 – in the cosine system – is applied exactly in phase with the first current harmonic Iz1 and multiplied by the current direction of the direct current component Izdc of the branch current Iz. The applied second current harmonic Iz2 is thus applied in such a way that it increases the maximum magnitude value or magnitude peak of the natural branch current occurring within the grid period by exactly the amplitude Az2 of the second current harmonic Iz2.
[0013] By controlling the sub-modules, the phase position of the fourth current harmonic is preferably kept constant relative to the phase of the first current harmonic.
[0014] With regard to the fourth current harmonic, it is considered advantageous if the control of the sub-modules is carried out in such a way that the maxima of the first current harmonics each coincide with a minimum of the fourth current harmonic when the direct current component in the branch currents is positive, and the minima of the first current harmonics each coincide with a maximum of the fourth current harmonic when the direct current component in the branch currents is negative.
[0015] For a cosinusoidal fundamental oscillation or cosinusoidal first current harmonic, the phase position is preferably set so that the following relationship is fulfilled: Iz4 t = − sign Izdc ⋅ Az 4 ⋅ cos 4 ωt mit Az4 > 0 where Iz4 denotes the fourth current harmonic and Az4 denotes the amplitude of the fourth current harmonic.
[0016] The fourth current harmonic Iz4 is thus applied inversely to the second current harmonic Iz2 in the cosine system of the fundamental oscillation. It thus reduces the maximum magnitude of the natural branch current by the amplitude value Az4 of the fourth current harmonic Iz4.
[0017] The resulting branch current Iz is thus: Iz t = Izdc + Az 1 ⋅ cos ωt + sign Izdc ⋅ Az 2 ⋅ cos 2 ωt − sign Izdc ⋅ Az4 ⋅ cos 4 ωt
[0018] By inverting the fourth current harmonic with a fixed relative phase position, the branch current peak can be significantly reduced compared to applying only the second current harmonic. Furthermore, the fixed relative phase position of the second and fourth current harmonics makes it easy to always determine the peak value of the branch current: The peak value corresponds to the sum of the DC component and the amplitudes of the first and second current harmonics, minus the amplitude of the fourth current harmonic.
[0019] The amplitude ratio between the amplitude of the fourth current harmonic and the amplitude of the second current harmonic is preferably kept constant by controlling the sub-modules.
[0020] It is advantageous if the control of the sub-modules keeps the amplitude ratio between the amplitude of the fourth current harmonic and the amplitude of the second current harmonic within a range between 0.4 and 0.7. The resulting current harmonic feedforward is very easy to implement with the latter variant, as it can be adjusted and controlled using a single variable, the amplitude of the second current harmonic. The fourth current harmonic follows a fixed factor k.
[0021] The value to which the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic is controlled is preferably chosen such that the electrical losses of the converter are minimized.
[0022] It is advantageous if the control of the submodules regulates the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic to a value that depends on the operating point of the converter. The operating point is preferably defined by the voltages, currents, power, and / or reactive power of the converter.
[0023] It is particularly advantageous if, by controlling the sub-modules, the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic is regulated to a value that depends on the power factor of the converter. In a preferred variant, the amplitude ratio is calculated by multiplying the power factor of the converter by a constant.
[0024] It is particularly advantageous if the value to which the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic is controlled has been determined as optimal in previous simulations of the converter for different operating points, in particular with regard to minimal electrical losses of the converter at the respective operating point.
[0025] A suitable or optimal amplitude ratio can be determined and saved, for example, through simulation calculations in the form of a characteristic map or a characteristic curve for various operating points of the converter. For this purpose, the amplitude ratio Az2 / Az1 can be varied for set or specified operating points, and the resulting converter losses can be calculated. The amplitude ratio Az2 / Az1 that results in the lowest losses for the respective operating point (e.g., defined by the active and reactive power of the converter) and / or simultaneously fulfills other boundary conditions (e.g., regarding falling below a maximum amplitude of the branch current Iz) is entered into the characteristic map and later read out again during converter operation.
[0026] The procedure can be carried out particularly easily if the value to which the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic is regulated is read from a memory in which the value to be set is stored for different operating points.
[0027] The converter is preferably equipped with at least one AC voltage terminal, to which an alternating current can be fed in or drawn, and at least one DC voltage terminal, to which a direct current can be fed in or drawn. At least two parallel-connected series circuits preferably form the converter's external DC voltage terminals.
[0028] It is advantageous if, by controlling the sub-modules, the amplitude ratio between the amplitude of the second current harmonic and the amplitude of the first current harmonic is regulated to a value which is selected or determined depending on the respective energy flow, in particular active power flow and / or reactive power flow, between the DC voltage side or the DC voltage connections and the AC voltage side or the AC voltage connections of the converter.
[0029] A six-pulse multilevel converter is preferred as the converter.
[0030] The invention further relates to a converter, in particular a multilevel converter, which has at least two module branches equipped with sub-modules and a control device which, by controlling the sub-modules in the module branches, generates a current - hereinafter referred to as the first current harmonic - which has a predetermined fundamental frequency, and, by controlling the sub-modules of the converter, generates an additional circulating current flowing within the converter - hereinafter referred to as the second current harmonic - which has twice the fundamental frequency.
[0031] According to the invention, the control device is designed in such a way that, by controlling the sub-modules, it generates an additional circulating current flowing within the converter - hereinafter referred to as the fourth current harmonic - which has four times the fundamental frequency.
[0032] With regard to the advantages of the converter according to the invention, reference is made to the advantages of the method according to the invention explained above, since the advantages of the method according to the invention essentially correspond to those of the converter according to the invention.
[0033] It is considered advantageous if the control device reduces the electrical losses of the converter and / or the maximum value of the magnitude of the branch currents by the fourth current harmonic so that these are smaller than when the converter is operated solely with the first and second current harmonics.
[0034] The invention is explained in more detail below using exemplary embodiments; by way of example, Figure 1 shows an embodiment of a converter according to the invention, Figure 2 shows an embodiment of a sub-module which can be used to form module branches in the converter according to Figure 1, Figure 3 shows a further embodiment of a sub-module which can be used to form module branches in the converter according to Figure 1can be used, Figure 4 shows a profile of the second and fourth current harmonics relative to the first current harmonic and relative to the natural branch current for a positive direct current component, Figure 5 shows a profile of the second and fourth current harmonics relative to the first current harmonic and relative to the natural branch current for a negative direct current component, Figure 6 shows a two-dimensional characteristic diagram for determining a suitable or optimal amplitude ratio between the amplitude of the first current harmonic and the amplitude of the second current harmonic as a function of the active and reactive current, and Figure 7 shows a characteristic curve for determining a suitable or optimal amplitude ratio between the amplitude of the first current harmonic and the amplitude of the second current harmonic as a function of the active current.
[0035] For the sake of clarity, the same reference symbols are always used in the figures for identical or comparable components.
[0036] The Figure 1 shows a converter 10, which has on its first connection side (AC voltage side) AS1 three AC voltage terminals L1, L2 and L3, at each of which an AC current can be fed into or taken from the converter 10. On a second connection side (DC voltage side) AS2 there are two DC voltage terminals, at which a DC current Idc can be fed into or taken from the converter 10; these are shown in the Figure 1 marked with the reference symbols L+ and L-.
[0037] The DC voltage between the DC terminals L+ and L- is designated Udc. The AC voltage between the AC terminals L1, L2, and L3 is designated Uac.
[0038] The converter 10 has three series circuits R1, R2, and R3, whose outer terminals form the DC voltage terminals L+ and L- of the converter 10. The series circuits R1, R2, and R3 each comprise two module branches TS connected in series.
[0039] Each of the module branches TS has at least two series-connected sub-modules SM, each comprising at least two switching elements and a capacitor. Examples of suitable sub-modules SM are described below in connection with the Figures 2 and 3 explained.
[0040] The converter 10 has a control device 20 suitable for controlling the submodules SM and thus for controlling the module branches TS. For this purpose, the control device 20 has a computing device 21 and a memory 22. A control program module SPM, which determines the operation of the computing device 21, is stored in the memory 22.
[0041] A possible operating procedure for the converter 10 is described below in connection with the Figures 4 to 7 explained in more detail.
[0042] The Figure 2 shows an embodiment of a sub-module SM, which comprises two switching elements S in the form of transistors, each of which has a diode D connected in parallel (or antiparallel), and a capacitor C as an energy storage device. The above-mentioned components form a half-bridge circuit, which is controlled by the switching elements S - by the control device 20 according to Figure 1 - enables unipolar operation of the capacitor C. The transistors and the diodes D connected in parallel can be prefabricated components, as indicated by boxes in the figures; the transistors and the diodes D connected in parallel can be, for example, IGBT components.
[0043] During active operation of the sub-module SM, one of the two switching elements S is switched on and the other switching element S is switched off. If the Figure 2 upper switching element S is switched on and the lower switching element S is switched off, the capacitor C is connected between the two terminals A1 and A2 and is electrically active; if the Figure 2 If the upper switching element is switched off and the lower switching element is switched on, the capacitor C is bridged and separated from the two terminals A1 and A2 by the diodes.
[0044] In passive operation of the sub-module SM, both switching elements S are switched off.
[0045] The Figure 3shows an embodiment of a sub-module SM, which comprises four switching elements S in the form of transistors, each of which has a diode connected in parallel, and a capacitor C as an energy storage device. The above-mentioned components form a full-bridge circuit, which is controlled by the switching elements S - by the control device 20 according to Figure 1 - enables bipolar operation of the capacitor C.
[0046] In active operation of the sub-module SM according to Figure 3 Two of the four switching elements S are switched on and the other two are switched off. In passive operation of the sub-module SM according to Figure 3 all four switching elements S are switched off.
[0047] Preferred operating modes of the SPM control program module are described below as examples: In a known manner, the submodules SM in each of the module branches are controlled with respect to the fundamental frequency of the converter, for example, with respect to the fundamental frequency of 50 Hz or 60 Hz, such that a desired energy flow occurs between the connection sides AS1 and AS2, or desired alternating currents flow on the first connection side AS1 and a desired direct current Idc flow on the second connection side AS2. The currents flowing in the module branches at the fundamental frequency are hereinafter referred to as first current harmonics.
[0048] In addition, the sub-modules SM in the module branches are controlled in such a way that converter-internal circulating currents flow, i.e. currents that only circulate in the branches of the converter 10 and do not leave the converter 10 to the outside, i.e. are not visible on either the first or the second connection side AS1 or AS2. The converter-internal circulating currents are in the Figure 1 marked with an arrow with the reference symbol Ik.
[0049] The so-called injection of internal converter circulating currents and the associated measures for controlling the SM submodules are well known, so reference is made to the literature in this regard. For example, the aforementioned publication "Loss optimization of MMC by Second-order-Harmonic Circulating Injection" (Yang, Li, Li, Wang, Xu, Gou, IEEE Transactions on Power Electronics, DOI 10.1109 / TPEL.2017.2751068) describes the injection of circulating currents at twice the fundamental frequency, i.e., the second current harmonic.
[0050] As described in more detail below, the control program module SPM will control the switches of the sub-modules SM in such a way that not only a second current harmonic, but also a fourth current harmonic is present in the circuit currents. The following explanations refer to the branch current Iz in the Figure 1upper module branch TS of the right series connection R3; however, they apply accordingly to the other module branches TS in the converter 10.
[0051] In the following, it is assumed that the first current harmonic Iz1 in the branch current Iz is cosinusoidal, i.e. the following applies: Iz 1 = Az 1 cos ωt , where Az1 denotes the amplitude of the first current harmonic Iz1 and has a positive sign or is greater than zero. ω denotes the angular frequency of the current, and t denotes time.
[0052] The above assumption of a cosine-shaped curve is chosen here only for the purpose of explanation and refers to a correspondingly arbitrarily chosen time base. With a different time base, the first current harmonic would not be cosine-shaped, but rather phase-shifted, for example, sinusoidal. However, this is irrelevant for the following explanations; with a different time base, only the respective phase shift relative to this other time base needs to be mathematically accounted for in the formula.
[0053] Since in normal operation, even without the addition of further current harmonics, a direct current Izdc usually occurs in the module branches, this leads to a so-called natural branch current Iznat according to: Iznat = Izdc + Az 1 cos ωt .
[0054] By applying the second current harmonic Iz2 and the fourth current harmonic Iz4, the resulting branch current Iz is: Iz = Iznat + Iz 2 t + Iz 4 t bzw . Iz = Izdc + Az 1 cos ωt + Iz 2 t + Iz4 t .
[0055] Regarding the amplitude and phase position of the second current harmonic Iz2 and the fourth current harmonic Iz4, the control program module SPM will preferably proceed as follows: Second current harmonic:
[0056] The phase position of the second current harmonic Iz2 is preferably adjusted such that the maxima of the first current harmonic Iz1 each coincide with a maximum of the second current harmonic Iz2 when the direct current component Idc in the branch current Iz is positive, and the minima of the first current harmonic Iz1 each coincide with a minimum of the second current harmonic Iz2 when the direct current component Idc in the branch current Iz is negative. The phase position is therefore adjusted such that the following relationship is satisfied: Iz 2 t = sign Izdc ⋅ A 2 z ⋅ cos 2 ωt mit A2z > 0
[0057] The second current harmonic Iz2 is applied – in the cosine system – exactly "in phase" (relative to time t=0) to the first current harmonic Iz1 and multiplied by the current direction of the direct current component Idc of the branch current Iz. The applied second current harmonic Iz2 is thus applied in such a way that it increases the maximum magnitude value or magnitude peak of the natural branch current Iznat occurring within the grid period by exactly the amplitude Az2 of the second current harmonic Iz2.
[0058] The Figure 4 shows an exemplary curve of the second current harmonic Iz2 relative to the first current harmonic Iz1 and relative to the natural branch current Iznat for a positive direct current component Izdc.
[0059] The Figure 5shows an exemplary curve of the second current harmonic Iz2 relative to the first current harmonic Iz1 and relative to the natural branch current Iznat for a negative direct current component Izdc.
[0060] The amplitude Az2 of the second current harmonic Iz2 is preferably adjusted depending on the amplitude Az1 of the first current harmonic Iz1. The following variants are considered advantageous in this regard: Variant 1:
[0061] In a first variant, it is provided that by controlling the sub-modules, the amplitude ratio between the amplitude Az2 of the second current harmonic Iz2 and the amplitude Az1 of the first current harmonic Iz1 is regulated to a value that depends on the operating point of the converter.
[0062] It is advantageous, for example, if the amplitude ratio Az2 / Az1 to be set is read from a characteristic map in which the respective target amplitude ratio is plotted as a function of an active power P to be set or already set by the converter (or active power flow between the DC voltage side and the AC voltage side) and a reactive power Q to be set or already set (or reactive power flow between the DC voltage side and the AC voltage side).
[0063] Alternatively, it can advantageously be provided that the amplitude ratio Az2 / Az1 to be set is read from a characteristic map in which the respective target amplitude ratio is plotted as a function of an active current Ip to be set by the converter and a reactive current Iq to be set or already set.
[0064] The Figure 6shows an example of such a two-dimensional characteristic diagram depending on the active current Ip and reactive current Iq.
[0065] During operation of the converter 10, the converter control or the control program module SPM according to Figure 1 reads out the corresponding amplitude ratio Az2 / Az1 based on the current operating point of the converter, i.e. based on the respective active and reactive power or based on the respective active and reactive current, and then sets it.
[0066] The map according to Figure 6is preferably calculated in project-specific simulations before commissioning the converter 10. For this purpose, for example, the amplitude ratio Az2 / Az1 is varied for set or specified support points (operating points of the converter), and the resulting converter losses are calculated. The amplitude ratio Az2 / Az1 that results in the best or lowest losses and / or simultaneously fulfills other boundary conditions (e.g., regarding falling below a maximum amplitude of the branch current Iz) is entered into the characteristic map and is then available for control during converter operation.
[0067] An advantage of this first variant is its very precise control, allowing for the amplitude of the branch current Iz to be considered for specific operating points. For operating points located between support points in the characteristic map, the amplitude ratio Az2 / Az1 can be determined, for example, by interpolation.
[0068] Instead of the Figures 6 A multi-dimensional characteristic map can also be created from the two-dimensional characteristic map shown and later used to determine the amplitude ratio Az2 / Az1 if more than two operating parameters of the converter 10 are to be taken into account. Variant 2:
[0069] In a second variant, it is provided that the amplitude ratio Az2 / Az1 between the amplitude Az2 of the second current harmonic Iz2 and the amplitude Az1 of the first current harmonic Iz1 is determined by evaluating a one-dimensional characteristic curve and a subsequent multiplication of an intermediate value ZW determined on the basis of the one-dimensional characteristic curve by a power factor cos(ϕ).
[0070] The Figure 7 shows an example of such a one-dimensional characteristic curve for the intermediate value ZW as a function of the active current Ip. Alternatively, the one-dimensional characteristic curve for the intermediate value ZW can also be plotted as a function of the active power P.
[0071] The one-dimensional characteristic curve is preferably calculated project-specifically in simulations for operating points without reactive power, i.e., only dependent on the active power, before commissioning the converter 10. For this purpose, for example, the amplitude ratio Az2 / Az1 is varied for set or specified support points of the one-dimensional characteristic curve, and the resulting converter losses are calculated. The amplitude ratio Az2 / Az1 that results in the best losses and / or simultaneously fulfills other boundary conditions (e.g., maintaining a maximum amplitude of the branch current Iz) is entered into the one-dimensional characteristic curve as an intermediate value ZW.
[0072] In the one-dimensional characteristic curve, as mentioned, the optimal intermediate values ZW are preferably plotted only for operating points without reactive power, i.e. one-dimensionally over the active power P or the active current Ip, whereby the reactive power Q or the reactive current Iq is assumed or set to zero.
[0073] During operation of the converter 10, the converter control or the control program module SPM according to Figure 1 reads the corresponding intermediate value ZW(P, Ip) based on the converter's current operating point, in this case based on the respective active power P or the respective active current Ip. For operating points located between support points on the characteristic curve, the intermediate value ZW can be determined, for example, by interpolation.
[0074] Subsequently, in order to take the respective reactive power Q into account, the read-out intermediate value ZW(P, Ip) is multiplied by the respective power factor cos(ϕ) of the converter 10, which results in the amplitude ratio Az2 / Az1 according to: Az 2 / Az 1 = ZW P Ip ⋅ cos φ Az 2 / Az 1 = ZW P Ip ⋅ P P 2 + Q 2 where P is the active power and Q is the reactive power.
[0075] An advantage of this second variant is the reduced effort, while still achieving a good result in terms of reducing electrical losses. Variant 3:
[0076] In a third variant, to determine the amplitude ratio Az2 / Az1, a fixed constant K is multiplied by the respective power factor cos(ϕ) of the converter 10 according to: Az 2 / Az 1 = K ⋅ cos φ with cos φ = P P 2 + Q 2
[0077] An advantage of this third variant is that it requires very little effort, while still achieving an acceptable result in terms of reducing electrical losses. Fourth current harmonic:
[0078] The phase position of the fourth current harmonic is preferably adjusted such that the maxima of the first current harmonic Iz1 each coincide with a minimum of the fourth current harmonic Iz4 when the direct current component Idc in the branch current Iz is positive, and the minima of the first current harmonic Iz1 each coincide with a maximum of the fourth current harmonic Iz4 when the direct current component Idc in the branch current Iz is negative. The phase position is therefore adjusted such that the following relationship is satisfied: Iz4 t = − sign Izdc ⋅ A4z ⋅ cos 4 ωt mit A4z > 0
[0079] The fourth current harmonic Iz4 is thus applied inversely to the second current harmonic Iz2 in the cosine system. It thus reduces the maximum magnitude of the natural branch current Iznat by the amplitude value Az4 of the fourth current harmonic Iz4.
[0080] In the described procedure - apart from the reversal of direction depending on the current direction of the direct current component Izdc - there is no variable or variable phase shift dependent on the operating point of the converter 10 between the applied current harmonics Iz2 and Iz4 relative to the given fundamental or first current harmonic Iz1.
[0081] The resulting branch current Iz is thus: Iz t = Izdc + Az1 ⋅ cos ωt + sign Izdc ⋅ Az2cos 2 ωt − Az4 ⋅ cos 4 ωt
[0082] The above-mentioned Figure 4also shows an exemplary curve of the fourth current harmonic Iz4 relative to the first and second current harmonics Iz1, Iz2 and relative to the natural branch current Iznat for a positive direct current component Izdc.
[0083] The Figure 5 shows an exemplary curve of the fourth current harmonic Iz4 relative to the first and second current harmonics Iz1, Iz2 and relative to the natural branch current Iznat for a negative direct current component Izdc.
[0084] It can be found in the Figures 4 and 5 It can be seen that the branch current peak of the branch current Iz is greatly reduced by the fourth current harmonic Iz4, which is inversely connected with a fixed phase position - in comparison to the sole connection of only the second current harmonic Iz2.
[0085] Due to the fixed phase position of the current harmonics Iz2 and Iz4, the peak value of the branch current Iz is always easy to determine; it can be calculated by summing the direct current component Izdc and the amplitudes Az1 and Az2 of the first and second current harmonics, minus the amplitude Az4 of the fourth current harmonic.
[0086] The amplitude Az4 of the fourth current harmonic Iz4 is preferably adjusted as a function of the amplitude Az2 of the second current harmonic Iz2. In this regard, it is considered advantageous if the amplitude ratio Az4 / Az2 between the amplitude Az4 of the fourth current harmonic and the amplitude Az2 of the second current harmonic is kept constant.
[0087] It is particularly advantageous if the amplitude ratio Az4 / Az2 is set between 0.4 and 0.7, so: 0,4 ≤ Az 4 / Az 2 ≤ 0,7
[0088] As a rule, an optimal value for Az4 / Az2 is 0.5 ± 10%.
[0089] The methods described above as examples for applying a second and fourth current harmonic can have one, several or all of the features or advantages listed below: By applying the second and fourth current harmonics, the primary critical quantities such as the peak value of the branch current or the effective value of the branch current are only slightly increased or can even be reduced. Unfavorable voltage feedback via the branch reactors can be reduced. A clear compensation of the voltage feedback is enabled. The process is easy to implement. According to simulation calculations, the described application method can achieve a loss reduction of 98% – relative to the "optimal," i.e., the theoretically maximum possible loss reduction. The increase in the peak value (peak and "rms" value) of the branch current is easy to determine; for a given maximum permissible peak value, the permissible strength of the application of the process is easy to determine (e.g., by selecting the size of the amplitudes Az4 and Az2). The voltage feedback is easy to determine.The control implementation is simple. The process can be carried out with just a single reference variable. The branch current peak, as a design-critical variable, can be calculated very easily. The feedback on the capacitor module voltage is easy to calculate. In multilevel converters (MMC) in six-pulse circuits with half-bridge submodules, a reduction in losses and energy swing occurs at all operating points with active power, even if only small, additional amplitudes are or can be applied. This reduces the risk of the converter getting out of control or entering uncontrolled B6 operation due to voltage feedback. Compared to the application of a second current harmonic alone, the additional application of the fourth current harmonic allows semiconductor switches with a lower current carrying capacity to be used than would be possible with the application of the second current harmonic alone.The converter can be operated with a larger external modulation index (ratio of AC voltage to half the DC voltage) than would be possible with the addition of the second current harmonic alone. The fixed phase position of the current harmonics and the optionally fixed amplitude ratio allow for a simple calculation of the voltage feedback across the branch reactors. The required time derivative of the currents is relatively simple due to the simple relationships (elimination of individual phase positions of the current harmonics) and can potentially even be implemented in real-time systems.
[0090] Although the invention has been illustrated and described in detail using preferred embodiments, the invention is not limited by the disclosed examples, and other variations may be derived therefrom by those skilled in the art. The scope of the invention is limited only by the claims.
Claims
1. Method for operating a converter (10), particularly a multilevel converter, which has at least two module arms (TS) equipped with sub-modules (SM), - a current - termed first current harmonic (Iz1) in the following - being generated by activating the sub-modules (SM) in the module arms in each case, which current has a predetermined fundamental frequency, and - an additional circulating current - termed second current harmonic (Iz2) in the following - flowing within the converter (10), being generated by activating the sub-modules (SM), which additional circulating current has double the fundamental frequency, an additional circulating current - termed fourth current harmonic (Iz4) in the following - flowing within the converter (10), being generated by activating the sub-modules (SM), which additional circulating current has four-times the fundamental frequency, characterized in that the second current harmonic (Iz2) is multiplied by a polarity sign of the DC component (Izdc) of a respective arm current (Iz) and injected in the cosine system in phase to the first current harmonic (Iz1) and the fourth current harmonic (Iz4) is multiplied by the inverse polarity sign of the DC component (Izdc) of the respective arm current (Iz) and injected in the cosine system in phase to the first current harmonic, wherein a maximum magnitude value of a natural arm current is reduced by an amplitude value (Az4) of the fourth current harmonic (Iz4) .
2. Method according to one of the preceding claims, characterized in that the phase position of the second current harmonic (Iz2) is kept constant relative to the phase of the first current harmonic (Iz1).
3. Method according to one of the preceding claims, characterized in that the activation of the sub-modules (SM) causes - the maxima of the first current harmonic (Iz1) in each case to coincide with a maximum of the second current harmonic (Iz2) when the DC component (Izdc) in the respective arm current (Iz) is positive, and - the minima of the first current harmonic (Iz1) in each case to coincide with a minimum of the second current harmonic (Iz2) when the DC component (Izdc) in the respective arm current (Iz) is negative.
4. Method according to one of the preceding claims, characterized in that due to the activation of the sub-modules (SM) - the maxima of the first current harmonic (Iz1) in each case coincide with a minimum of the fourth current harmonic (Iz4) when the DC component (Izdc) in the respective arm current (Iz) is positive, and - the minima of the first current harmonic (Iz1) in each case coincide with a maximum of the fourth current harmonic (Iz4) when the DC component (Izdc) in the respective arm current (Iz) is negative.
5. Method according to one of the preceding claims, characterized in that the amplitude ratio between the amplitude of the fourth current harmonic (Iz4) and the amplitude of the second current harmonic (Iz2) is kept constant.
6. Method according to one of the preceding claims, characterized in that the amplitude ratio between the amplitude of the fourth current harmonic (Iz4) and the amplitude of the second current harmonic (Iz2) is kept within a range between 0.4 and 0.7.
7. Method according to one of the preceding claims, characterized in that the value to which the amplitude ratio between the amplitude of the second current harmonic (Iz2) and the amplitude of the first current harmonic (Iz1) is adjusted is chosen in such a manner that the electrical losses of the converter (10) become minimal.
8. Method according to one of the preceding claims, characterized in that the amplitude ratio between the amplitude of the second current harmonic (Iz2) and the amplitude of the first current harmonic (Iz1) is adjusted to a value which depends on the operating point of the converter (10).
9. Method according to one of the preceding claims, characterized in that the amplitude ratio between the amplitude of the second current harmonic (Iz2) and the amplitude of the first current harmonic (Iz1) is adjusted to a value which depends on the power factor of the converter (10), and in particular is formed by multiplying the power factor of the converter (10) by a constant.
10. Method according to one of the preceding claims, characterized in that the value to which the amplitude ratio between the amplitude of the second current harmonic (Iz2) and the amplitude of the first current harmonic (Iz1) is adjusted has been determined as optimal in the context of previous simulations of the converter (10) for various operating points, specifically in particular with a view to minimal electrical losses of the converter (10) at the respective operating point.
11. Method according to one of the preceding claims, characterized in that the value to which the amplitude ratio between the amplitude of the second current harmonic (Iz2) and the amplitude of the first current harmonic (Iz1) is adjusted is read from storage in which the value to be set for various operating points in each case is stored.
12. Method according to one of the preceding claims, characterized in that the converter operated is a six-pulse multilevel converter.
13. Converter, particularly a multilevel converter, which has at least two module arms (TS) equipped with sub-modules (SM) and also a control device (20), which - in each case generates a current (Iz1) - termed first current harmonic (Iz1) in the following - by activating the sub-modules (SM) in the module arms, which current has a predetermined fundamental frequency, and - generates an additional circulating current - termed second current harmonic (Iz2) in the following - flowing within the converter (10), by activating the sub-modules (SM) of the converter (10), which additional circulating current has double the fundamental frequency, the control device (20) being configured in such a manner that, by activating the sub-modules (SM), it generates an additional circulating current - termed fourth current harmonic (Iz4) in the following - flowing within the converter (10), which additional circulating current has four-times the fundamental frequency, characterized in that the control device (20) is furthermore equipped to multiply the second current harmonic (Iz2) by a polarity sign of the DC component (Izdc) of a respective arm current (Iz) and inject it in the cosine system in phase to the first current harmonic (Iz1) and to multiply the fourth current harmonic (Iz4) by the inverse polarity sign of the DC component (Izdc) of the respective arm current (Iz) and inject it in the cosine system in phase to the first current harmonic, wherein a maximum magnitude value of a natural arm current is reduced by an amplitude value (Az4) of the fourth current harmonic (Iz4) .
