Operation of switching elements of an inverter

Abstract

The invention relates to a method for operating switching elements (24) of an inverter (20) which is electrically coupled to a multiphase alternating voltage (16), wherein the inverter has at least one series circuit (22) which consists of the switching elements (24) for each of the phases of the multiphase alternating voltage in order to electrically couple the multiphase alternating voltage to a direct voltage intermediate circuit (28) which is connected to the inverter by applying switching signals (40) to the switching elements of the series circuits, wherein the switching signals are determined on the basis of a clock signal using pulse width modulation, wherein, depending on phase currents and an intermediate circuit direct voltage based on a field-oriented controller (36), a space vector is determined and the switching signals are determined depending on the space vector and using phase target voltages. According to the invention, a damping space vector is superimposed on the space vector to determine the switching signals and is determined depending on the phase currents such that phase positions of the phase target voltages are shifted by a predetermined phase value with respect to at least one predetermined frequency which is greater than the fundamental vibration.

Description

[0001] Description

[0002] Operating switching elements of an inverter

[0003] The invention relates to a method for operating switching elements of an inverter which is electrically coupled to a multi-phase alternating voltage of an alternating voltage network via an AC-side filter, wherein the inverter has at least one series circuit of the switching elements for each of the phases of the multi-phase alternating voltage in order to electrically couple the multi-phase alternating voltage to a DC intermediate circuit connected to the inverter by applying switching signals to the switching elements of the series circuits, wherein the switching signals are determined based on a clock signal using pulse width modulation, for which purpose the phase currents of phases of the multi-phase alternating voltage and an intermediate DC voltage are detected, wherein a clock period of the clock signal is smaller than an oscillation period of a fundamental oscillation of the phase voltages,wherein, depending on the detected phase currents and the detected intermediate DC voltage, a space vector is determined based on a field-oriented control, and the switching signals are determined depending on the space vector by means of a space vector modulation and using phase target voltages. Furthermore, the invention relates to a control device for operating switching elements of an inverter which is electrically coupled to a multi-phase alternating voltage of an alternating voltage network via an AC-side filter, wherein the inverter has at least one series circuit of the switching elements for each of the phases of the multi-phase alternating voltage, wherein the control device is designed to apply switching signals to the switching elements of the series circuits in order to electrically couple the multi-phase alternating voltage to a DC intermediate circuit connected to the inverter,to determine the switching signals based on a clock signal using pulse width modulation, for which purpose the phase currents of phases of the multi-phase alternating voltage and an intermediate circuit direct voltage are detected, wherein a clock period of the clock signal is smaller than an oscillation period of a fundamental oscillation of the phase voltages, and to determine a space vector depending on the detected phase currents and the detected intermediate circuit direct voltage based on field-oriented control and to determine the switching signals depending on the space vector by means of space vector modulation and using phase target voltages. Finally, the invention also relates to an energy converter for electrically coupling a multi-phase electrical alternating voltage of an alternating voltage network to a direct voltage intermediate circuit, with an inverter,the inverter has at least one series circuit of the switching elements for each of the phases of the multi-phase alternating voltage, wherein a respective series circuit is electrically coupled to the DC voltage intermediate circuit and a respective center terminal of a respective series circuit is electrically coupled to a respective phase of the multi-phase alternating voltage, an AC-side filter for electrically coupling the inverter to the multi-phase alternating voltage, and a control device for operating the switching elements of the inverter.

[0004] Energy converters with inverters, control devices therefor, and methods for operating switching elements of inverters are extensively known in the prior art, so that separate written documentation is not required. Nowadays, electrical energy converters, also called electrical energy transformers, are used in the form of so-called static energy converters or energy transformers. This means that, unlike dynamic energy converters, they do not require any mechanically movable, particularly rotatable, parts for the purpose of energy conversion.Static energy converters are generally designed as clocked electrical energy converters and for this purpose have at least one inverter with switching elements which are interconnected in a suitable manner and are connected, for example, to phase connections, called phases for short, of a multi-phase alternating voltage and at least partially to a DC voltage intermediate circuit in order to electrically couple the multi-phase alternating voltage to the DC voltage intermediate circuit, so that the desired conversion function of the energy converter can be achieved by operating the switching elements in a suitable switching mode.

[0005] A switching element within the meaning of this disclosure is preferably a controllable electronic switching element, for example a controllable electronic semiconductor switch such as a transistor, a thyristor, combination circuits thereof, preferably with parallel-connected freewheeling diodes, a gate turn-off thyristor (GTO), an insulated-gate bipolar transistor (IGBT), combinations thereof, or the like. In principle, the switching element can also be formed by a field-effect transistor, in particular a metal-oxide semiconductor field-effect transistor (MOSFET).

[0006] To provide the desired energy conversion functionality, the switching elements are operated in switching mode.

[0007] In relation to a semiconductor switch such as a transistor, this means that in the switched-on state, a very low electrical resistance is provided between the terminals of the transistor forming a switching path, so that a high current flow is possible with a very low residual voltage. In the switched-off state, however, the switching path of the transistor has a high impedance, i.e. it provides a high electrical resistance, so that even with a high electrical voltage applied to the switching path, there is essentially no current flow, or only a very low, in particular negligible, current flow. This is different from linear operation in transistors, which is not generally used in clocked energy converters.It can be provided that the DC voltage intermediate circuit is provided by a DC voltage network and / or the multi-phase AC voltage is provided by a multi-phase AC voltage network, which can be, for example, a public power supply network. The multi-phase AC voltage is preferably a three-phase AC voltage. Depending on the network, however, it can also be a two-phase AC voltage, a four-phase AC voltage, a five-phase AC voltage, or the like. The number of phases of the multi-phase AC voltage is generally application-dependent and not limited.

[0008] The switching elements of the inverter are coupled to the control device. The coupling is preferably designed such that each of the switching elements can be controlled individually. The control device can be designed as an electronic circuit which provides corresponding control signals for the switching elements so that the desired switching operation of the switching elements can be realized. In addition to electronic components for the predeterminable provision of the control signals, the electronic circuit can also comprise at least one program-controlled computing unit in order to be able to provide the desired function of the control device. Of course, the control device can also consist exclusively of the computing unit.

[0009] The control device is designed to operate the switching elements in switching mode such that the energy converter provides the predeterminable energy conversion functionality. Furthermore, the control device is designed to set phase currents of the multi-phase alternating voltage in a predeterminable manner. The control device therefore provides a control functionality by means of which the phase currents can be set in a predeterminable manner. This makes it possible to regulate the phase current on a respective phase in a predeterminable manner. Suitable current sensors or the like can be used, for example, to detect the phase currents. The current sensors can be included in the energy converter. However, it can also be provided that the energy converter has connections for the current sensors, which are included, for example, in the corresponding electrical alternating voltage network to which the regulated phase currents are to be applied.

[0010] To control the phase currents, it is usual to provide suitable control functionality. One frequently used option for providing the phase currents is field-oriented control using space vector modulation, which - depending on the design - is sometimes also referred to as vector control and which, in the three-phase case, is based on the use of a d / q transformation, also called Park transformation. The Park transformation is used to transform three-phase variables of the three-phase electrical machine into a two-axis coordinate system which has reference axes d and q. This enables vector control or vector control to be implemented, in which space vector representations can be used to set the operating states of the inverter.The d / q transformation is related to a Clarke transformation and differs from the latter in that the d / q coordinate system of the d / q transformation rotates with a rotor of the three-phase electrical machine during stationary operation. In this case, a respective pair of values ​​d, q can represent a temporally constant quantity for this operating state. Using field-oriented control, phase target voltages can be determined for each of the phases of the multi-phase alternating voltage. The switching signals can then be determined as a function of the phase target voltages within the framework of pulse width modulation.

[0011] Inverters that are used as frequency converters, particularly grid inverters, and are therefore connected to a multi-phase AC network are usually connected via network-side filters, particularly LCL filters, to smooth the phase currents or to reduce the harmonic content of phase voltages of the multi-phase AC voltage and / or conducted interference to such an extent that the applicable network guidelines and / or network standards are adhered to. Such filters usually have a combination of inductances and capacitances. This leads to oscillating systems that can develop resonances. Damping elements are therefore required to avoid very large resonance peaks in individual frequency ranges around resonant frequencies.Traditionally, passive ohmic resistors are integrated into a filter structure as attenuation elements, or parasitic ohmic properties of components are used for passive attenuation.

[0012] The disadvantages of this proven damping technology, however, are the resulting power loss, which leads to reduced efficiency, as well as the associated additional technical effort for the dissipation of waste heat, additional installation space for the damping elements and the associated additional costs, and finally a reduced filtering effect of the filter due to the passive damping elements in the actually desired frequency range of the filter, which further increase the costs and installation space.

[0013] The invention is based on the object of developing methods of the generic type as well as control devices and energy converters in order to reduce the need for attenuators in filters with inductances and capacitances.

[0014] As a solution, the invention proposes a method, a control device and an energy converter according to the independent claims.

[0015] Advantageous further developments arise from features of the dependent claims. With regard to a generic method, the invention proposes in particular that the space vector is determined exclusively as a function of the detected phase currents and the detected intermediate circuit DC voltage, and that a damping space vector is additionally superimposed on the space vector for determining the switching signals, which is determined as a function of a damping signal which, in turn, is determined as a function of the phase currents, so that phase positions of the desired phase voltages are shifted by a predetermined phase value with respect to at least one predetermined frequency which is greater than the fundamental oscillation.

[0016] With regard to a generic control device, the invention proposes in particular that the control device is further designed to determine the space vector exclusively as a function of the detected phase currents and the detected intermediate circuit DC voltage and to additionally superimpose a damping space vector on the space vector for determining the switching signals, which is determined as a function of a damping signal which in turn is determined as a function of the phase currents, so that phase positions of the desired phase voltages are shifted by a predetermined phase value with respect to at least one predetermined frequency which is greater than the fundamental oscillation.

[0017] With regard to a generic energy converter, the invention proposes in particular that the control device is designed according to the invention.

[0018] The invention is based, among other things, on the idea that a control structure with filtering of the target voltage calculated before the field-oriented control, taking into account a damping signal, can be used to reduce damping elements. The control structure achieved in this way can, for example, have a digital filter, although preferably no band-stop filter characteristic is configured, and not even as a high-pass or low-pass filter. This design avoids a situation in which a stop-stop filter characteristic with a respective stopband could prevent excitation of a resonance point, but at the same time a resulting control characteristic would lose a damping effect in the event of external interference excitations in the affected frequency range.By shifting the phase positions of the phase currents by the specified phase value with respect to the at least one specified frequency, any excitation of AC-side resonances can be reduced or even prevented, without the need for damping elements. This allows, in particular, the disadvantages associated with damping elements to be reduced or avoided.

[0019] As a rule, grid-side resonance frequencies can be excited or amplified in conjunction with the AC-side filter by the clock operation of the inverter. A grid-side resonance frequency is usually greater than a frequency of the fundamental oscillation of the AC voltage. The predetermined frequency can also be a frequency range that is preferably above the frequency of the fundamental oscillation. If a resonance effect occurs, this is indicated by a corresponding amplitude in the phase currents at this frequency. Such resonance frequencies can be greater and / or smaller than a frequency corresponding to the clock period. The control device is preferably designed to carry out the phase shift for one or more predetermined frequencies or frequency ranges.In particular, it can be provided that the control device does not carry out the phase shift essentially for the frequency of the fundamental oscillation or a range comprising the frequency of the fundamental oscillation.

[0020] The damping signal is an electrical signal that can be linked to the space vector, which can also be formed by one or more electrical signals. For example, at least partial superposition, in particular addition, can be provided.

[0021] The attenuation signal can have a separate value for each phase of the alternating voltage. The attenuation space vector is determined as a function of the attenuation signal by means of an attenuation unit of the control device. The attenuation unit can have a digital filter, for example. The attenuation space vector is preferably provided as a Clarke transform. The superposition with the space vector can be achieved by a combination which can comprise vector addition or the like, for example. An attenuation space vector can be assigned to an individual resonant frequency, for example. The attenuation space vector, in turn, can be formed from a superposition of several individual frequency-specific attenuation space vectors. This allows several occurring resonant frequencies to be treated essentially together or simultaneously. However, no settings need to be made for this purpose with the invention.The invention essentially makes it possible to adapt to occurring resonance frequencies and to suppress them.

[0022] An advantageous system property can be to reduce the dependence of filter resonance points, for example with regard to frequency and amplitude enhancement, on properties of the AC network, particularly with regard to a complex network impedance at the connection point. In an electrical AC network such as the public power grid, the complex network impedance can vary, for example due to changing other electrical devices that are also electrically coupled to the AC network. This means that the resonance points of the AC network can change slowly or abruptly during operation and are therefore generally largely unknown.In addition, additional mains filters from other inverters can be connected to the same connection point of the AC network and, depending on their operation in particular, can lead to additional or changed resonance points in the overall system of the AC network.

[0023] The invention therefore provides a method and a control device for active damping, in particular of an LC (L) mains filter, which is able to ensure the necessary damping effect for a variable position of one or more resonance points over a wide band, comparable to a real ohmic resistor as a damping element. The invention is therefore particularly suitable for filters designed in the manner of an LCL filter. The invention avoids having to measure an electrical voltage across the filter or an electrical current across a filter branch and taking these into account when determining the space vector. In particular, there is no need to record electrical variables on the filter side, such as a filter current, an electrical voltage, in particular in the region of the filter, or the like. The invention manages without such measurements.It is based on the use of phase currents, which are already available for the implementation of field-oriented control. Likewise, there is no need to provide a frequency-selective filter such as a band-stop filter, which would be impractical anyway due to the notch frequency that would have to be adjusted depending on the system resonance subject to variation.

[0024] The invention can be implemented using either analog or digital technology. In this respect, the control device can have an analog and / or a digital hardware circuit to implement the function according to the invention. In principle, the function could also be implemented using a program-controlled computer unit. However, the hardware circuit is preferably used due to its generally faster response time compared to the program-controlled computer unit. The invention can be implemented, for example, in the manner of an AS IC, an FPGA or the like. In particular, the invention can also be retrofitted to existing control devices or control device designs with little effort. If a computer unit is provided, the invention can also be at least partially covered by the computer program.

[0025] The invention is particularly suitable for use in grid inverters, such as those disclosed, for example, in EP 3 576 284 A1. However, the use of the invention is not limited to this and can also be used, among other things, when an electrical machine is electrically coupled to the inverter instead of the multiphase alternating voltage. The invention can be particularly advantageously combined with a PWM in the manner of a discontinuous modulation method.

[0026] According to the invention, only broadband filters need to be used, and in particular, no band-stop filters are required, which would have to be tuned to the current filter resonance and thus continuously adjusted to changing network parameters, especially resonance frequencies. This makes it possible, for example, to work with a fixed controller setting over a wide range of network impedance variations.

[0027] In summary, the invention may be characterized in particular by one or more of the following features:

[0028] - Active filter damping without additional measured variables, but exclusively with the help of recorded inverter currents or phase currents in the phases of the multi-phase AC voltage network.

[0029] - Extension of the current control loop with a digital phase-compensating filter, in particular a higher-order lead-lag element or a modified all-pass filter with a suitable phase response, - no model or observer is introduced into the field-oriented control, whereby such a component would require at least partial knowledge of parameters of the multi-phase AC network and would thus reduce the robustness of the control system as a whole against fluctuations of these unknown parameters,

[0030] - Integration of the, preferably entire, control into an integrated circuit, for example AS IC or the like, which preferably includes a measured value acquisition and processing, a current controller, at least a proportional controller, a phase-compensating, in particular digital, filter, for example 2nd order, a calculation of the switching signals in the modulator for the space vector modulation,

[0031] - Use of an integrating current actual value acquisition, for example in the manner of a sigma-delta measurement to suppress high-frequency interfering frequency components in the current measurement, and / or

[0032] - Use of flat modulation to minimize power dissipation and to compensate for the resulting increased dead time by a higher-order digital filter in the modulator.

[0033] It is further proposed that the phase positions be shifted in advance. This allows for good damping of at least one resonance of the AC voltage network. The advance of the phase positions is preferably selected such that response times of the control device, in particular with regard to field-oriented control, space vector modulation and / or the detection of the phase currents, are taken into account. A sufficient phase reserve should be taken into account in the design so that damping of resonance frequencies can be reliably achieved even at high resonance frequencies or in stiff AC voltage networks.

[0034] It is further proposed that the damping signal is provided by means of a damping unit which determines the at least one predetermined frequency and which evaluates the phase currents and determines the phase value depending on the evaluation. The damping unit is, for example, part of the control device. The damping unit can provide a filter functionality which implements the shifting of the phase, in particular the leading of the phase. Preferably, the damping signal is provided for several, in particular all, resonance frequencies or one or more frequency ranges. It can be taken into account that the resonance frequency can depend on an interaction between the AC voltage network and the filter. The damping signal therefore does not need to be provided permanently during normal operation.It can be provided that the phase currents are evaluated and, depending on this, a determination is made as to whether the inverter is operating in the range of a resonant frequency. If such an operating state does not exist, the damping signal can be deactivated. This refinement can reduce or even eliminate dead time in the control system.

[0035] According to a further development, it is proposed that the recorded phase currents are digitized using an integrating measuring method. The integrating measuring method can be, for example, a sigma-delta method, an oversampling method or the like. In this further development, at least partially digital signal processing is provided. The sigma-delta method provides an integrating method for recording currents so that, for example, harmonics can be reduced during recording, which can arise, for example, when sampling a signal value that is provided by the respective current sensor when recording the respective phase current. For example, a 1-bit signal stream can be provided. In principle, however, a comparable integrating method can also be used for digitization instead of the sigma-delta method.In particular, a measuring interval with very small, especially no, dead time can be achieved.

[0036] Preferably, a digitization clock period is at least a factor of 100, preferably at least a factor of 1000, smaller than the clock period of the clock signal. The digitization clock period corresponds to a sampling rate when digitizing the signal value. Because the digitization clock period is considerably smaller than the clock period of the clock signal, precise and reliable digitization can be achieved despite the integrating sampling method.

[0037] According to a further development, discontinuous pulse width modulation is proposed. Discontinuous pulse width modulation can be used, for example, to reduce switching losses in the switching elements. Discontinuous pulse width modulation is particularly suitable for use in energy converters that convert large amounts of electrical power.

[0038] It is further proposed that an averaging process be performed for the detected phase currents, extending over at least two clock periods of the clock signal. This allows the method implementation of the invention to be further improved. This embodiment can, for example, be particularly advantageously combined with discontinuous pulse width modulation. It is possible for measured values ​​for the phase currents to be detected symmetrically with respect to the pulse width modulation, in particular the clock signal.

[0039] It is further proposed that the phase currents be detected in respective detection periods, with each of the detection periods extending over a respective clock period of the clock signal. This makes it possible to use the detected current values ​​of the phase currents directly for the subsequent determination of the switching signals. The detection of the phase currents advantageously extends over a respective clock period of the clock signal of the pulse width modulation.

[0040] Furthermore, it is proposed that the phase voltages of the multiphase alternating voltage be detected before determining the space vector, with the field-oriented control being initialized depending on the voltage signals corresponding to the phase voltages. This allows synchronization of the field-oriented control with the alternating voltage to be achieved, especially if the inverter is a grid-connected inverter.

[0041] It is further proposed that a transfer function is used to provide the damping signal, which at least in the range of the at least one predetermined frequency essentially has an amplification factor in a range of approximately 1. The resonant frequency results, among other things, from an interaction of the AC voltage network providing the AC voltage with the filter, also called a mains filter, and possibly other connected electrical devices. At least within the range, the control device provides the transfer function, on the basis of which the damping signal is determined as a function of the phase currents. The damping signal can preferably be determined exclusively as a function of the phase currents. An amplitude is therefore preferably changed as little as possible when determining the damping signal.In contrast to this, a phase position of the damping signal is determined in a suitable manner so that the desired preferential advance in the phase position can be achieved within the framework of the superposition with the space vector.

[0042] The control device preferably has a phase shifter circuit for providing the attenuation signal. The desired phase shift can be achieved in a simple manner using hardware technology using the electronic phase shifter circuit. The phase shifter circuit can be provided at least partially by the attenuation unit. A phase shifter circuit is an electronic circuit that shifts the phase of an electrical oscillation. For lagging phase shifts, electronic delay lines or delay circuits can be used. For leading phase shifts, phase locked loops (PLL) or the like can be used, particularly for periodic signals. The phase shifter circuit enables a desired phase position to be set quickly and accurately.In particular, the phase shifter circuit can be designed as a lead-lag element.

[0043] For use cases or application situations that may arise during the method and which are not explicitly described here, it may be provided that, in accordance with the method, an error message and / or a request to enter user feedback is issued and / or a standard setting and / or a predetermined initial state is set.

[0044] The advantages and effects stated for the method according to the invention naturally also apply equally to the control device according to the invention and the energy converter according to the invention, and vice versa. In this respect, method features can also be formulated as device features, and vice versa.

[0045] The features and combinations of features mentioned above in the description as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures can be used not only in the respectively specified combination but also in other combinations without departing from the scope of the invention. The exemplary embodiments explained below are preferred embodiments of the invention. The features and combinations of features mentioned above in the description as well as the features and combinations of features mentioned in the following description of exemplary embodiments and / or shown alone in the figures can be used not only in the respectively specified combination but also in other combinations.Thus, embodiments are also encompassed by the invention or are to be regarded as disclosed which are not explicitly shown and explained in the figures, but which arise and can be produced by separate combinations of features from the explained embodiments. The features, functions and / or effects illustrated with reference to the exemplary embodiments can, taken on their own, each represent individual features, functions and / or effects of the invention which can be viewed independently of one another and which each also develop the invention independently of one another. Therefore, the exemplary embodiments are intended to also include combinations other than those in the explained embodiments. Furthermore, the described embodiments can also be supplemented by further features, functions and / or effects of the invention already described.

[0046] In the figures, the same reference symbols designate the same features or functions.

[0047] It shows :

[0048] FIG 1 is a schematic circuit diagram of an energy converter connected to an AC voltage network, which has an inverter, a filter for electrically coupling the AC voltage network and a control device;

[0049] FIG 2 is a schematic circuit diagram like FIG 1, in which the filter is designed as a T-filter and a current sensor detects an electrical current in a branch of the T-filter;

[0050] FIG 3 is a schematic circuit diagram like FIG 2, in which an electrical filter voltage is detected instead of the electrical current;

[0051] FIG 4 shows a schematic block diagram of a field-oriented control and a space vector modulation of the control device, in which a space vector is additionally processed by means of a phase shift circuit as an active damping circuit;

[0052] FIG 5 is a schematic diagram of a frequency-dependent gain curve of a transfer function of the damping circuit according to FIG 4;

[0053] FIG 6 is a schematic diagram of a frequency-dependent phase curve of the transfer function of the damping circuit according to FIG 4;

[0054] FIG 7 is a schematic diagram of a frequency-dependent gain curve of the open control loop according to FIG 4 without the active damping circuit;

[0055] FIG 8 is a schematic diagram of a frequency-dependent phase response of an open control loop according to FIG 4 without the active damping circuit;

[0056] FIG 9 is a schematic diagram like FIG 7 with the active damping circuit;

[0057] FIG 10 is a schematic diagram like FIG 8 with the active damping circuit;

[0058] FIG. 11 shows a schematic diagram like FIG. 5, in which different transfer functions are represented by means of three graphs; FIG. 12 shows a schematic diagram like FIG. 6, in which different transfer functions are represented by means of three graphs as in FIG. 5;

[0059] FIG 13 is a schematic block diagram of the active damping circuit according to FIG 4; and

[0060] FIG 14 is a schematic diagram of a Clarke transform of a space vector with a superimposed damping space vector.

[0061] FIG. 1 shows a schematic circuit diagram of an energy converter 10 connected to an AC voltage network 14, which comprises an inverter 20, a filter 12 for electrically coupling the inverter 20 to the AC voltage network 14, and a control device 30. In the present case, the AC voltage network 14 provides a three-phase AC voltage 16 as a multi-phase electrical AC voltage. In alternative embodiments, the AC voltage can also have a different number of phases.

[0062] The energy converter 10 serves to electrically couple the three-phase electrical alternating voltage 16 of the alternating voltage network 14 to a direct voltage intermediate circuit 18, to which an intermediate direct voltage 28 is applied. The inverter 20 has at least one series circuit 22 of switching elements 24 for each of the phases of the three-phase alternating voltage 16. Each series circuit 22 is electrically connected to the direct voltage intermediate circuit 18. A respective center connection 26 of each series circuit 22 is electrically coupled to a respective phase of the three-phase alternating voltage 16. On the alternating voltage side, the filter 12 is connected to the inverter 20, via which filter the inverter 20 is electrically coupled to the three-phase alternating voltage 16. A control device 30 serves, among other things, to operate the switching elements 24 of the inverter 20 by providing corresponding switching signals 40.In the present embodiment, a clock rate for providing the switching signals 40 is approximately 4 kHz. However, in alternative embodiments, the clock rate may also be selected differently.

[0063] In the illustration according to FIG 1, the AC voltage network 14 is symbolically represented by a three-phase AC voltage source, which is schematically connected in series with a network inductance L gri d is switched on. The network inductance L grid symbolically represents an inductive effect of the alternating voltage network 14 .

[0064] The filter 12 is designed here as an LCL filter with a T-filter structure and has a series connection of two inductances, namely an inductance L k and an inductance L e . A current generated by the inductance L e provided terminal is electrically connected to the AC voltage network 14. A through the inductance Lk The terminal provided is electrically connected to the respective phase terminals of the inverter 20. In this case, the inductances L grid , L kf L e each intended for each phase.

[0065] Respective center terminals of the series circuits of the inductors L k and L e are connected to a respective series circuit of an electrical resistor R d and an electrical capacitance Cf. These series circuits are connected by their opposite terminals to an electrical reference potential 44, which in this case is a ground potential.

[0066] FIG 2 shows a schematic circuit diagram like FIG 1 , in which an electrical current sensor 46 is arranged in a branch of the LCL filter 12, which is formed by the respective series circuit of the electrical resistance R dand the electrical capacitance Cf, detects an electrical current, and provides a corresponding current signal 50 to the control device 30. This embodiment is provided separately for each phase of the three-phase alternating voltage 16. The current signals 50 of the current sensors 46 are fed to the control device 30.

[0067] FIG. 2 also schematically shows one of the series circuits 22 with its switching elements 24 of the inverter 20 for one of the phases of the three-phase alternating voltage 16. The series circuits 22 are provided for all phases of the three-phase alternating voltage 16. It can be seen that the series circuits 22 are designed as half-bridge circuits in the present case. In the present embodiment, it is further provided that the switching elements 24 are formed by IGBTs. In alternative embodiments, other suitable switching elements can of course also be provided here.

[0068] FIG. 2 also shows that the control device 30 has a damping controller 38. The damping controller 38 is connected to the current sensors 46 and receives the corresponding current signals 50 from the respective current sensors 46. Furthermore, current sensors 54 are provided, with which the respective phase currents iu of the three-phase alternating voltage 16 can be detected. The current sensors 54 are connected to a field-oriented controller 36.

[0069] The field-oriented controller 36 determines a space vector 80, which is linked to a damping signal 32 of the damping controller 38 by means of a link 56. This signal is then fed to a space vector modulation 42, which provides the switching signals 40 for the switching elements 24, so that the desired switching operation can be achieved by means of the inverter 20.

[0070] FIG 3 shows a schematic circuit diagram like FIG 2 , in which instead of detecting the electrical current by means of the current sensors 46 for each phase, an electrical filter voltage is measured by means of a respective voltage sensor 48 at the respective center connection of the inductors L e and L k with respect to the reference potential 44. Corresponding voltage signals 52 are again fed to the damping control 38. The further structure corresponds to that already explained in connection with FIG. 2, which is why additional reference is made to the relevant explanations.

[0071] FIG. 4 shows, in a schematic block diagram, among other things, a section of the control device 30 as used in FIGS. 1 to 3. Not shown in FIG. 4 is that the detected intermediate circuit DC voltage 28 is fed via an analog-to-digital converter to an intermediate circuit controller (not shown). Likewise, a comparison value for the intermediate circuit DC voltage 28 is fed to the intermediate circuit controller.

[0072] Also not shown is that phase voltages of the multi-phase alternating voltage 16 are detected and fed to a PLL circuit by means of a further analog-to-digital converter. The PLL circuit provides a phase signal. Detecting the phase voltages is not absolutely necessary during normal operation and only needs to be performed, for example, once, to commence normal operation of the energy converter 10 or the inverter 20.

[0073] Finally, phase currents 70 are detected and fed to a current controller (not shown) by means of a further analog-to-digital converter 58 (FIG. 13). Current sensors 54 can be used for this purpose. The current controller is also supplied with the values ​​Id, ref / I q , ref as well as the phase signal are supplied. The current controller determines a space vector 80 from the supplied variables. During normal operation, this field-oriented control requires only the intermediate circuit DC voltage in addition to the phase currents. Unlike in the prior art, no further electrical variables, in particular those of the filter 12, need to be detected and made available. It can be seen from FIG 4 that the space vector 80 of the space vector modulation 42 is not supplied directly but via a link circuit 84. The link circuit 84 links the space vector 80 with a damping space vector 114 in the manner of a vector addition. The damping space vector 114 serves to dampen grid-side resonance frequencies during normal operation of the energy converter 10 or the inverter 20.

[0074] The attenuation space vector 114 is determined by means of a signal-to-space vector converter 116 as a function of an attenuation signal 32. The attenuation signal 32 is in turn determined as a function of the detected digitized phase currents 70 by a phase shifter circuit 34 as an attenuation unit. The logic circuit 84 modifies the space vector 80 with respect to a phase advance and determines therefrom a leading space vector 82. The leading space vector 82 is fed to the space vector modulation 42, by means of which the switching signals 40 for the switching elements 24 of the inverter 20 are determined.

[0075] It can be seen from FIG 4 that the field-oriented control 36 is based on a fundamentally known field-oriented control, which is why further detailed explanations in this regard are omitted here. The field-oriented control 36 is coupled for communication purposes to a setpoint register 124, which provides setpoints for the field-oriented control 36 in a known manner. In contrast to the prior art, the space vector 80 according to the invention is not fed directly to the space vector modulation 42, but is first further processed by means of the logic circuit 84 before it is fed to the space vector modulation 42.

[0076] It can also be seen from FIG 4 that the recorded phase currents 70 are first digitized by means of the analog-to-digital converter 58. For this purpose, it is provided in the present case that the recorded phase currents 70 are digitized using a sigma-delta method as an integrating measuring method. In this case, a digitization clock period is at least a factor of 1000 smaller than the clock period of the clock signal. Then, as part of the digitization, it is provided that an averaging is carried out for the recorded phase currents 70, which extends over two clock periods of the clock signal. The digitized phase currents 70 are then fed to a Clarke transformer 122. The transformed phase currents are then linked to target values ​​120 for the phase currents by means of a linking circuit 126. In the present case, a difference is formed and a difference signal 128 is provided.The difference signal 128 is fed to the phase shifter circuit 34. The setpoint values ​​120 are provided using a block 118 that is communicatively coupled to the setpoint register 124 and provides current setpoint values ​​in dq coordinates. The function of block 118 is known to those skilled in the art, which is why further detailed explanations are omitted here.

[0077] FIG. 13 shows an enlarged schematic representation of a section of FIG. 4 in the area of ​​the phase-shift circuit 34 and the signal processing with respect to the space vector 80. It can be seen that the detected phase currents 70 are converted into digital signals by means of the analog-to-digital converter 58. The sigma-delta method is used for the conversion.

[0078] A digitization clock period used for the sigma-delta method is selected in the present case in a range from approximately 10 MHz to approximately 20 MHz. This makes it possible to provide an integrating current measurement method which is advantageously suited to reducing harmonics caused by switching operations. The phase currents 70 digitized in this way are fed to the phase shifter circuit 34, forming a difference with the desired values ​​120. The phase shifter circuit 34 supplies the attenuation signal 32, which is converted as an attenuation space vector 114 and linked to the space vector 80 by means of the linking circuit 84 so that the leading space vector 82 can be provided. This is then fed to the space vector modulation 42, which determines the switching signals 40 therefrom in a known manner.

[0079] The phase shifter circuit 34 is designed here as a lead-lag element and enables a phase gain to be achieved. This makes it possible to compensate for system-related propagation times, such as those that can occur, for example, in PWM, when processing signals, when detecting sensor values, and / or the like. Furthermore, a sufficient phase reserve can be created so that, in conjunction with the filter 12, resonant frequencies on the AC mains side can be actively damped. The filter function of the phase shifter circuit 34 specifies a frequency range in which resonant frequencies that occur can be damped.

[0080] The function of the invention will be further explained below with reference to diagrams. It is based on an energy converter 10 which has the filter 12 and the inverter 20, wherein a clock rate with respect to the provision of the switching signals 40 is approximately 4 kHz. Depending on the network inductance L grid and thus dependent on a relative system short-circuit voltage or an inverted quantity thereto, namely a relative system short-circuit power, which in the present case lies between approximately 5 and approximately 250, a resonance peak of the entire system including the AC voltage network 14 in the present example lies approximately between 930 Hz and 1400 Hz.

[0081] This is illustrated by FIGS 7 and 8, which represent a Bode diagram. An abscissa in the two schematic diagrams according to FIGS 7 and 8 is assigned to the frequency, whereas in FIG 7 the ordinate is assigned to an amplitude and the ordinate in FIG 8 is assigned to a phase. A graph 86 represents an amplitude curve against the frequency for a relative system short-circuit power RSC 250, whereas a graph 88 shows an amplitude curve for a relative system short-circuit power RSC 5. The corresponding phase curves can be seen from the associated schematic phase diagram according to FIG 8, whereas a graph 90 is assigned to graph 86 and a graph 92 to graph 88.

[0082] The resonance points can be easily identified by the phase shifts. From FIGS. 7 and 8 it can be seen that with increasing size of the network inductance L grida resonance frequency of the filter 12 becomes smaller. The phase gain is considerably smaller here, or it can even become negative. Due to the system propagation times, the phase loss can increase, especially at low resonance frequencies. The system propagation time, also called dead time, takes into account the runtime of the control loop. Within a system propagation time, a resonant oscillation can develop or change by an angular change Acp, which depends, among other things, on the resonance frequency: h(p —tdead * f resonance, where tdead is the system propagation time and f resonance is the resonance frequency. To compensate for the system propagation time in the control loop and to calculate a damping circuit that matches the instantaneous phase position of the resonance oscillation, an ideal digital filter would have to perform a corresponding frequency-dependent correction of the phase position of the setpoint voltage. All-pass filters would generally be suitable for this in terms of their amplitude characteristics, but they do not exhibit suitable phase response characteristics across the entire relevant control frequency range.

[0083] In contrast, with the phase shifter circuit 34, in particular when it is designed as a lead-lag element, the desired adaptation of the phase response or the frequency-dependent phase curve can be achieved, wherein at the same time undesirable side effects with regard to an amplitude response or a frequency-dependent amplitude curve can be kept small or avoided.

[0084] Figures 7 and 8 illustrate the function of the open control loop without the phase-shifting circuit 34. It can be seen that at a frequency of approximately 1400 Hz, the frequency response exhibits a phase shift of approximately -180°. This control loop is therefore unstable in the range of this resonant frequency.

[0085] FIGS. 9 and 10 schematically depict a Bode diagram similar to FIGS. 7 and 8. This diagram shows the control loop with the phase-shifting circuit 34 activated. As can be seen from graphs 86 to 92, in both cases shown, stable operation can be achieved even in the range of the resonant frequencies because sufficient phase reserve can be provided.

[0086] The following provides design criteria for the phase-shifting circuit 34. One design criterion is that a sufficient phase advance should be provided for the highest resonant frequency. Furthermore, it should preferably be possible to achieve a phase response that is as linear as possible from the highest resonant frequency to the lowest resonant frequency, among other things, so that the system propagation times can be at least partially compensated.

[0087] Finally, an amplitude increase, especially at high frequencies, should preferably be kept as small as possible, preferably approximating an all-pass behavior. Furthermore, it is naturally particularly advantageous to reduce system runtimes by designing the entire control loop with a correspondingly minimal runtime. For this reason, the field-oriented control 36, and preferably also the space vector modulation 42, are implemented as a hardware circuit, in particular a digital hardware circuit, for example in the form of an AS IC, FPGA, or the like.

[0088] FIGS. 11 and 12 show a further Bode diagram in a schematic representation, which represents a transfer function of the phase-shifting circuit 34. The abscissas of the schematic diagrams are assigned to the frequency, whereas an ordinate in FIG. 11 is assigned to the amplitude and an ordinate in FIG. 12 is assigned to a phase. Graphs 94, 96, and 98 show three different amplitude curves of the transfer function, with graphs 100, 102, and 104 representing correspondingly assigned phase curves of the transfer function.

[0089] In the present embodiment it is further provided that in this case a discontinuous modulation, also called flat modulation, is implemented as space vector modulation 42. This is a variant of space vector modulation in which only two of three phases are switched by means of the switching elements within a clock cycle. This type of modulation is therefore particularly suitable for use in grid inverters which are operated at high electrical voltage, in particular at the limit of overload, whereby a reduction in switching losses and an increase in efficiency can be achieved. However, a disadvantage is that a current curve is only symmetrical to a complete pulse period. With the integrating current detection of the phase currents 70 at a clock rate of 4 kHz it is advantageous to average accordingly over a period of 250 ps.For other modulation types, however, averaging over a period of 125 ps, i.e., half a period, may be sufficient. The resulting longer period for averaging results in a longer propagation time in the damping control loop and thus a smaller permissible fluctuation range or phase margin in the range of the resonance frequencies, which can result in a smaller stability range during control, particularly with varying network parameters.

[0090] With the phase shifter circuit 34, particularly when it is designed as a lead-lag element, it is possible to at least partially compensate for this system-related additional propagation time, as can be seen from FIGS. 11 and 12. Graphs 94 and 100 show this clearly for a phase shifter circuit 34, whereas graphs 98 and 104 represent the function with the aforementioned averaging, and graphs 96 and 102 represent a combination of the two aforementioned options. The phase shifter circuit 34 has been parameterized such that the additional averaging, as explained above, is almost completely compensated and the resulting phase loss can thus be considerably reduced.

[0091] 5 and 6 show a further Bode diagram, with FIG. 5 showing a frequency-dependent amplitude curve and FIG. 6 a frequency-dependent phase curve of the transfer function of the phase-shifting circuit 34. The abscissas of FIGS. 5 and 6 are again assigned to the frequency. The ordinate of FIG. 5 is assigned to the amplitude and the ordinate of FIG. 6 is assigned to the phase. The filter transfer function of the phase-shifting circuit 34 is shown in graphs 106 and 110. It can be seen from graphs 108 and 112 that the additional filtering or averaging mentioned above is able to provide a further improvement in the transfer function for the desired application. In particular, an almost linear phase increase can be provided in the desired range, as can be seen from the schematic diagram in FIG. 6.This proves to be particularly advantageous for the function. FIG. 14 shows a schematic diagram of the relationship between the space vectors 80, 82, 114 as a Clarke transform. It can be seen that the space vector 82 is formed from a vector addition of the space vector 80 and the damping space vector 114. It is evident that the space vector 80 has a considerably lower frequency than the damping space vector 114. This means that during a comparatively slow revolution of the space vector 80, a fast revolution of the damping space vector 114 is superimposed. This is represented by a circle 130 in FIG. 14. This also clearly shows the effect of the invention. If no resonance frequency occurs, then due to the function of the damping unit there is essentially no significant damping space vector 114.If, however, a resonance frequency occurs, the phase shift circuit 34 or the damping unit provides a suitable damping space vector 114 which makes it possible to counteract the resonance.

[0092] Overall, the invention can achieve the following with a sensible design:

[0093] Active damping can be achieved without additional measured variables because the phase-shifting circuit 34 only requires the phase currents 70, which are already required for the intended operation of the inverter 20, to function. In particular, separate phase voltages do not need to be permanently measured.

[0094] Furthermore, only the phase shifter circuit 34 needs to be provided. In particular, no model or observer needs to be introduced into the control loop. In any case, this component would require at least partial knowledge of the parameters of the AC voltage network 14 and thus reduce the stability of the control system with regard to variations in these unknown parameters. Furthermore, the entire control loop can be integrated into a circuit, such as the AS IC, the FPGA or the like. This integrated circuit should preferably also comprise measured value acquisition and / or measured value processing as well as the current controller, a phase-compensated digital filter, preferably second order, and the calculation of switching actions in the space vector modulation 42.Furthermore, an integrating detection of the phase currents, in particular according to a sigma-delta method, can preferably be used to suppress, for example, high-frequency interfering frequency components in the current measurement. Finally, discontinuous pulse width modulation (PWM) can preferably be used to minimize power loss and to compensate for resulting delays, for example by using higher-order digital filters in the modulator.

[0095] The invention allows only the phase currents 70 in the inverter 20 to be required as measured values. In particular, no external currents and voltages, for example from a mains filter or the mains, need to be taken into account.

[0096] In particular, precise knowledge of current grid parameter values ​​is not necessary over a wide range of grid properties. The controller of the invention is robust to changing grid properties; these do not need to be known.

[0097] Furthermore, the control loop does not require any bandpass or blocking filters, whose blocking frequency would have to be precisely adapted to the current network properties.

[0098] Even if the invention is explained here based on the application to a three-phase alternating voltage as a multi-phase alternating voltage, the invention is not limited to this and can be used with any multi-phase alternating voltage, for example with four-, five-, or six-phase alternating voltages.

[0099] The embodiments serve solely to explain the invention and are not intended to limit it.

Claims

Patent claims 1. Method for operating switching elements (24) of an inverter (20) which is electrically coupled to a multi-phase alternating voltage (16) of an alternating voltage network (14) via an AC-side filter (12), wherein the inverter (20) has at least one series circuit (22) of the switching elements (24) for each of the phases of the multi-phase alternating voltage (16) in order to electrically couple the multi-phase alternating voltage (16) to a DC voltage intermediate circuit (18) connected to the inverter (20), in that the switching elements (24) of the series circuits (22) are supplied with switching signals (40), wherein the switching signals (40) are determined based on a clock signal using pulse width modulation, for which purpose phase currents (70) of phases of the multi-phase alternating voltage (16) and an intermediate circuit direct voltage (28) are detected, wherein a clock period of the clock signal is smaller than an oscillation period of a fundamental oscillation of phase voltages of the multi-phase alternating voltage (28), wherein a space vector (80) is determined depending on the detected phase currents (70) and the detected intermediate circuit direct voltage (28) based on a field-oriented control (36), and the switching signals (40) are determined depending on the space vector (80) by means of space vector modulation (42) and using phase target voltages, characterized in thatthat the space vector (80) is determined exclusively as a function of the detected phase currents (70) and the detected intermediate circuit DC voltage (28), and a damping space vector (114) is superimposed on the space vector (80) for determining the switching signals (40), which is determined as a function of a damping signal (32), which in turn is determined as a function of the phase currents (70), so that phase positions of the desired phase voltages are shifted by a predetermined phase value with respect to at least one predetermined frequency that is greater than the fundamental oscillation.

2. Method according to claim 1, characterized in that the phase positions are shifted in advance.

3. Method according to one of the preceding claims, characterized in that the attenuation signal (32) is provided by means of an attenuation unit (34) by which the at least one predetermined frequency is determined and which evaluates the phase currents (70) and determines the phase value depending on the evaluation.

4. Method according to one of the preceding claims, characterized in that the detected phase currents (70) are digitized by means of an integrating measuring method.

5. Method according to one of the preceding claims, characterized in that a discontinuous pulse width modulation is used as the pulse width modulation.

6. Method according to one of the preceding claims, characterized in that an averaging is carried out for the detected phase currents (70), which extends over at least two clock periods of the clock signal.

7. Method according to one of the preceding claims, characterized in that the detection of the phase currents (70) takes place in respective detection periods, wherein a respective one of the detection periods extends over a respective clock period of the clock signal.

8. Method according to one of the preceding claims, characterized in that the phase voltages of the multi-phase alternating voltage (16) are detected before determining the space vector (80), wherein initialization of the field-oriented control (36) takes place depending on voltage signals corresponding to the phase voltages.

9. Method according to one of the preceding claims, characterized in that for providing the damping signal (32) a transfer function is used which has an amplification factor in a range of 1 at least in the range of the at least one predetermined frequency.

10. Control device (30) for operating switching elements (24) of an inverter (20) which is connected to a multi-phase AC voltage (16) of an AC voltage network (14) is electrically coupled via an AC-side filter (12), wherein the inverter (20) has at least one series circuit (22) of the switching elements (24) for each of the phases of the multi-phase AC voltage (16), wherein the control device (30) is designed, - to apply switching signals (40) to the switching elements (24) of the series circuits (22) in order to electrically couple the multi-phase alternating voltage (16) to a direct voltage intermediate circuit (18) connected to the inverter (20), - to determine the switching signals (40) based on a clock signal using pulse width modulation, for which purpose phase currents (70) of phases of the multi-phase alternating voltage (16) and an intermediate circuit direct voltage (28) are detected, wherein a clock period of the clock signal is smaller than an oscillation period of a fundamental oscillation of phase voltages of the multi-phase alternating voltage (16), and - to determine a space vector (80) depending on the detected phase currents (70) and the detected intermediate circuit voltage (28) based on a field-oriented control (36) and to generate the switching signals (40) depending on the space vector (80) by means of a space vector modulation (42) and using phase target voltages, characterized in that the control device (30) is further designed, - to determine the space vector (80) exclusively as a function of the detected phase currents (70) and the detected intermediate circuit DC voltage (28), and - to superimpose a damping space vector (114) on the space vector (80) in addition to determining the switching signals (40), which is determined as a function of a damping signal (32), which in turn is determined as a function of the phase currents (70), so that phase positions of the phase target voltages are shifted by a predetermined phase value with respect to at least one predetermined frequency which is greater than the fundamental oscillation.

11. Control device according to claim 10, characterized by a phase shifter circuit (34) for providing the attenuation signal (32).

12. Energy converter (10) for electrically coupling a multi-phase alternating voltage (16) of an alternating voltage network (14) with a direct voltage intermediate circuit (18), with - an inverter (20) which has at least one series circuit (22) of the switching elements (24) for each of the phases of the multi-phase alternating voltage (16), wherein a respective series circuit (22) is electrically coupled to the DC voltage intermediate circuit (18) and a respective center terminal (26) of a respective series circuit (22) is connected to a respective phase of the multi-phase alternating voltage (16) can be electrically coupled, - an AC-side filter (12) for electrically coupling the inverter (20) to the multi-phase AC voltage (16), and - a control device (30) for operating the switching elements (24) of the inverter (20), characterized in that the control device (30) is designed according to one of claims 10 or 11.

Images