DE10127776A1 - Simulation analysis and conditioning of power electronic circuits - Google Patents

Abstract

The power electronics process uses a data process system that communicates with a data base and has input and display facility. This is used to analyse the voltage intermediate circuit (2) between the AC (40) and DC (41) circuits. Input parameters are entered and an optimising process initiated. The Output is generated that specifies components to meet the requirement. A comparison and simulation is made with specification to ensure that performance is within a tolerance band.

Description

The invention relates to a method for simulation, analysis and conditioning power electronic circuits, especially on voltage intermediate circuit Systems based on converters, such as active filter systems, in the field of Power supply, control and drive technology.

The variety of usable power electronic devices and components for Realization of a given by the respective task or problem Circuit design, especially in the field of power supply, control and Drive technology, given the numerous and increasing areas of application and the ever shorter development cycles are getting bigger and bigger unmanageable.

So must with regard to each new task or problem comprehensive and often complex and difficult analysis and selection of any suitable and both with a given circuit structure and set Default values or requirements of compatible components and components to be met be performed. A process about its successful implementation as a rule only field trials and experimental reviews provide definitive certainty.

Conducting field trials requires a fair amount of time as well financial effort, for example by creating the corresponding Test setups, which not least also have to meet safety-related questions. All of this is mostly associated with high costs, especially installation costs, but only leads in the first approximation to the goal. So are up to a final, satisfactory Task or problem solution often multiple test runs with possibly varying Test setups are required, which is driving the spiral of costs upwards. short-term Changes or innovations are only comparatively difficult to do here take into account and / or implement.

For the reasons just mentioned, one is therefore endeavored and attempted with regard to economy, time management, flexibility and variability, some few, well-known or precisely specified power electronic circuits, in different applications or areas of application, as broadband as possible use.

That is, the aim is to have as many different ones as possible Requirements, applications and areas of application with as uniform as possible Range of power electronic circuits and the associated components cover.

For this, the individual components and / or a correspondingly configured one Circuit structure, according to the respective area of application and thus associated default values or boundary conditions or specifications with regard to their Functionality and operational safety are checked in detail.

In order to avoid occurring in practice and / or experiment and often very costly and time-consuming device defects or failures as well as the associated repair and / or to be able to largely exclude downtimes from the outset, it is due to time and cost reasons the different system and To be able to determine or calculate operating parameters in advance as precisely as possible and, if necessary, individual power electronic components in their Circuit environment simulated based on their specification data, for example using a corresponding data processing system to test and operate.

Such a project requires due to the required broadband Possible uses of power electronic components and circuits, as well as the large number different wiring options a generalized procedure which Short-term changes in the simplest possible way, for example the ones used Pulse width modulation strategy, the converter or converter configurations as well of the AC circuit and / or DC voltage intermediate circuit Voltage intermediate circuit converter based active filter system enables or takes into account and a simple and quick determination of the right or optimal one Design and conditioning of the entire circuit structure as well as the used ones Components required operating parameters allowed.

The highest possible accuracy in determining necessary or relevant Operating parameters allow a comparatively optimal design or Conditioning of the entire, underlying circuit and accordingly, For example, an expansion of the permissible safe area or a reduction the safety margin while increasing efficiency and utilization the respective circuit and / or the respective component and thus also one Reduction of installation costs.

With regard to an on-line, i.e. direct use of the method in the context of a Regarding its practicability or practical applicability, in particular, the time required for a process run or its speed of vitally important.

Can be done using known time-based methods, for example under Use of the data processing programs PSPICE and MATLAB / SIMULINK, individual Circuit complexes with a comparatively large time and considerable Just simulating and analyzing computing effort, this requires a simulation and analysis More complex power electronic circuit structures, such as one active filter system, on a data processing device with a Pentium processor already a time expenditure or a computing time, which can be far more than what still allows efficient work.

This is an accurate and exhaustive system or circuit analysis time consuming and no longer practicable from an economic point of view, an on-line use is unthinkable.

For example, a sufficiently precise determination of circuit-relevant operating data is required or parameters, for example a determination of the frequency spectrum that occurs Harmonic components, in simulation and analysis a procedure in very small Time intervals or very small time steps. An approach that in turn leads to an increase in the computing effort and inevitably to one Increases the processing time required.

Common time-based procedures are therefore often inflexible and sluggish. So require, for example, minor modifications to the circuit layout or an exchange individual components usually a complete re-analysis of the present Systems, which makes efficient and speedy work almost impossible.

In comparison, procedures appear based on the technique of frequency analysis are based on the simulation of power electronic circuit structures for the same Increment to work much more accurately than time-based procedures. Besides that is the accuracy that can be achieved there with the time-based method comparable.

So far, none of the known methods offers possibilities in combination Implementation of an efficient and sufficiently precise simulation, analysis and Conditioning of power electronic circuits.

Only partial aspects are covered. So there is often the possibility of Analysis, but the means of simulation are missing.

Rather are for a comprehensive determination of the relevant operating parameters numerous separate conditioning, analysis and simulation processes as well as the associated system tools required. The requirement of combinatorial Use of different system tools to accomplish a given task leads, however, to a reduction in the amount of training required User friendliness, a deterioration in clarity and consequently, for example by multiple data or parameter transfer for an increase the frequency of errors and / or susceptibility to errors.

The object of the invention is based on analysis and processing corresponding input variables of a predetermined power electronic circuit, especially with voltage intermediate circuit converters with pulse width modulation, in As part of a uniform procedure, a determination of the optimized for each Application and the predetermined input parameters coordinated operating parameters and thus optimal conditioning of the entire circuit as well as individual To allow components and / or components.

This task is accomplished through a modular process and associated Device for simulation, analysis and conditioning of power electronics Circuits, in particular based on voltage intermediate circuit converters Power electronic systems, such as active filter systems, solved by by means of a data processing device with an input unit, data memory and Output unit and at least one interface for recording relevant Input variables, such as supply voltages and / or load currents, and if necessary, for the output of corresponding control variables that have been recorded Input variables, based on a predetermined circuit structure, by means of several, linked processing modules are automatically processed and analyzed and depending on the resulting, optimized for the particular circuit and the predetermined input variables matched operating parameters and predetermined Suitable components and components for setting up the Realization or completion of the respective power electronic circuit automatically selected from one or more appropriately configured databases and taking into account their specification data, further ones resulting operating parameters, such as power losses, determined and Appropriate preventive measures, for example, appropriately designed coolants, especially heat sinks, from one or more preconfigured databases be selected, whereby an optimized design or conditioning of the entire Power electronic circuit or the circuit structure is achieved.

Experimental inputs can be used as input variables Measured variables recorded and / or stored or by specifying corresponding ones Boundary conditions generated within the process, such as Load currents, harmonics and supply voltages then serve automatically recorded, retrieved from a data store and / or manually via a corresponding input device can be entered.

The determined, optimized for the respective circuit and the given Input parameters matched operating parameters as well as the specification data of the selected Components are then compared to the default values.

If the data match within predetermined tolerances, then both the Operating parameters as well as a parts list of the selected components and Components together with their specification data by means of a corresponding device for Displayed and output if necessary, for example via a printer or Plotter, as well as stored on a data storage device.

If the data do not match within the predetermined tolerances, they are Modify default values and / or the circuit structure and that to go through the inventive method again.

Due to its modular structure, the method according to the invention allows known systems in an advantageous manner, the analysis and determination also individual parameters and operating parameters and thus an optimized coordination, Design or conditioning of individual circuit components or Control mechanisms, so that with minor modifications already existing or real existing circuits do not necessarily include all process steps or processing modules have to go through.

In addition to the off-line application area for planning, analysis and review and / or design purposes, that is, already recorded, stored Input variables are analyzed and processed and, as a result, corresponding ones Determines operating parameters and, if necessary, suitable components and Components selected, there is also the option of using the on-line method in the To use within the framework of a regulation.

Compared to the off-line area, on-line use becomes experimental Input variables are recorded, analyzed, processed and directly using suitable interfaces evaluated and there are for the respective circuit and the given The operating parameters resulting from input variables are determined or ascertained. The through the Determined operating parameters given setpoints or manipulated values are conveyed suitable processing modules prepared and a control device for generation appropriate control variables to achieve an optimized operating state transmitted.

The increasing power requirements in both the private and industrial sectors as well as the development and use of increasingly powerful and powerful Machines and drive systems and the associated occurrence always stronger power electronic and electrical loads, as well as caused by them Harmonics, particularly in the energy supply sector, but also in traffic engineering to a number of adverse side effects. So perform harmonics caused by nonlinear loads for example to a Impairment of the supply currents or voltages or their quality and thus the Quality of service, what for consumers connected to the same electrical supply network the occurrence of malfunctions and / or device failures or -may mean defects.

One way to deal with the problem is to use active Filter systems that are operated here like a regulated countercurrent source.

Accordingly, active filter elements connected in parallel can be found in many technical Areas for regulation and / or compensation by non-linear loads caused current harmonics and the resulting reactive currents in order to ensure that the supply currents when nonlinear loading occurs maintain their sinusoidal shape and continue to have a power factor of One remains.

Power electronic controlled by means of pulse width modulation Voltage intermediate circuit converters, such as those used for active filter systems, show for usually an AC circuit, one DC voltage intermediate circuit and regulator for switching the converter or converter on.

The range of services and the interaction of the individual Processing modules of the method according to the invention are described below using an example active filter system to compensate for nonlinear loads Harmonics set out.

The performance spectrum of the method according to the invention allows one active filter system on the AC side, for example the determination of Voltage harmonics of the converter, current harmonics, current rms values as well as absolute current peaks and also allows based on the aforementioned Measured value quantities a reconstruction of the current waveform or the current profile an alternating current or fundamental oscillation cycle.

The converter module of a voltage intermediate circuit converter, in particular also an active filter, has power electronics in a known manner Components such as IGBTs (Insulated Gate Bipolar Transistors) and / or IGCTs (Integrated Gate Commutated Thyristors) and semiconductor diodes. Regarding the The converter module allows the method according to the invention to determine the Average current value, rms current value, absolute current peak value and Current flow of the respective component and switching and / or Structural losses. The determination and calculation of any power losses that occur the thermal development and conditioning of the components used as well as the entire structure, including the provided heat sink.

The DC link has a filter capacitor with resistors for voltage adjustment. On the DC side, this enables inventive method the determination and determination of current harmonics, Voltage harmonics, one for limiting the current ripple to one predetermined value required capacitor and more. Beyond that it according to the invention with comparatively little technical and time expenditure possible to modify the DC link, for example Add an inductively and capacitively tuned filter circuit for suppression dominant current harmonics in the current of the DC link or through Adding another voltage intermediate circuit converter for the drive an AC motor or by adding a DC chopper to the drive a DC motor.

The modulation voltage of the voltage intermediate circuit converter is below Taking into account the transfer function of the control loop of the corresponding Application calculated. Different modes of action of an analog and a digital control as well as different control strategies for the comparison Structure and conditioning of the control loop should be considered.

To calculate the individual current parameters of the respective components and To be able to determine, it is necessary to determine the time course of the current reconstruct or reproduce. The waveform within one of the full period of the current and / or voltage fundamental Replicate time span.

The method according to the invention advantageously uses both the advantages of the time as well as the frequency space, with on-line, i.e. direct use, for example in the context of a control device, when determining the corresponding Operating parameters for reasons of time to take into account the influences of the Switching frequency is dispensed with.

In the case of an off-line application, however, it is recommended with a view to being as accurate as possible Determination of the optimal conditioning of an existing or in the company operating circuit requires an accurate analysis occurring or expected harmonic components in electricity and / or Voltage profiles, for example caused by switching a converter, perform.

The method according to the invention opens up a variety of areas of application, for example in connection with active line-side converters (active front-end converter) or STATCOM's (static compensators). With active network-side Power converters are the currents of the line-side converter depending on the on the To determine the DC side occurring load. The load power on the DC side must be known, so that the supply of a corresponding current can be withdrawn. In the case of a STATCOM, the load currents are processed and only compensates for the blind component contained.

Uninterruptible power supplies (UPS) are very tight with active filters related. The only difference is that the UPS Basic vibration active part of the load current is provided in addition to the harmonic components becomes. For this reason, there is no need to consuming the load currents to process. Rather, the current of the active filter is the load current equate.

In the case of drive controls, the input choke is the active filter and the supply voltages through the corresponding components and To replace courses of the respective drive.

The implementation or implementation of the method according to the invention can be done in hard-wired form using analog components and control elements as well using digital processors or in the form of a Data processing program using digital control elements, suitable interfaces for Data or signal input and output and a correspondingly designed Data processing device or as a combination of the aforementioned options and Components are made.

Some of the further explanation and explanation of the invention are given Drawings and working examples.

Further advantageous embodiments of the invention are the figure descriptions and the dependent claims.

Show it

Fig. 1 scheme of a compensation circuit with an active filter

Fig. 2 construction diagram of a three-phase active filter

Fig. 3 room pointer representation of the eight switching options of the converter

Fig. 4 Generalized block diagram of a voltage intermediate circuit converter with pulse width modulation, control device ac and dc circuit.

Fig. 5 pulse width modulation method and course of the switching waveform of the reference voltage of the second phase

Fig. 6 flow chart of the step-by-step analysis, simulation and conditioning of a converter circuit for an active filter

Fig. 7a sampled waveform of the individual phases of the supply voltage

Fig. 7b sampled waveform of the individual phases of the load current

Fig. 8 sampled waveform of the individual phases of the current of the active filter

Fig. 9 Sampled waveform of the reference voltages of the three phases of the active filter

Fig. 10 sampled, standardized reference voltages of the three phases of the active filter

Fig. 11 digital reconstruction of the switching signals for controlling the active filter, based on a sine / triangle modulation

Fig. 12 harmonic components of the supply voltage of the first phase for the first 50 harmonic orders

Fig. 13 harmonic components of the converter output voltage of the first phase for the first 50 harmonic orders

Fig. 14 Complete harmonic spectrum of the converter output voltage of the first phase

Fig. 15 harmonic components of the current of the active filter of the first phase for the first 50 harmonic orders

Fig. 16 Complete harmonic spectrum of the current of the active filter of the first phase

Fig. 17 harmonic components of the load current of the first phase for the first 50 orders

Fig. 18 Harmonic spectrum of the supply current of the first phase, adjusted by active filtering

Fig. 19 Remaining harmonic spectrum of the supply current of the first phase up to the 50th harmonic order, without fundamental wave

Fig. 20 Reconstructed current curve of the first phase of the active filter

Fig. 21 Reconstructed voltage during the first phase of the active filter

Fig. 22 Reconstructed current flow of the first phase through the upper diode of the power converter

Fig. 23 Reconstructed current flow of the first phase through the upper IGBT of the converter

Fig. 24 Reconstructed current flow of the first phase through the lower diode of the converter

Fig. 25 Reconstructed current flow of the first phase through the lower IGBT of the converter

Fig. 26 On-line, without taking into account the influences of the switching frequency determined or reconstructed current flow of the first phase through the upper diode of the converter

Fig. 27 On-line, without taking into account the influences of the switching frequency determined or reconstructed current flow of the first phase through the upper IGBT of the converter

Fig. 28 Reconstructed course of the current flowing across the capacitor of the DC link

Fig. 29 Harmonic components of the current flowing across the capacitor of the DC link for the first 50 harmonic orders

Fig. 30 Complete harmonic spectrum of the current flowing across the capacitor of the DC link

Fig. 31 On-line, without taking into account the influences of the switching frequency determined or reconstructed course of the current flowing through the capacitor of the DC voltage intermediate circuit

Fig. 32 the course of the time-integrated current flowing through the capacitor of the DC intermediate circuit

Fig. 33 Equivalent circuit diagram for the problem of first switching on the active filter

Fig. 34 the course of the resistance R _p, the time t _max, as well as the current value i _dcmax as a function of the damping factor ξ

The mode of operation, possibilities and advantages of the method according to the invention are used below to compensate for non-linear loads caused current harmonics and the resulting reactive currents used control loop with active filter system.

In Fig. 1, a circuit configuration is shown in which a non-linear load is fed from a three-phase AC voltage source 11 12. An active filter 10 for compensating harmonics caused by nonlinear loads is connected in parallel with the load. The exemplary course of the supply voltage, the load current, the filter current or current of the active filter 10 and the compensated supply current is shown in each case.

The construction of an active three-phase filter 10 with converter 20 , with semiconductor valves 21 , inductors L _f , ohmic resistors R _f and two capacitors 22 is shown in FIG. 2. The voltage across the capacitors 22 is regulated to maintain a constant virtual DC voltage u _dc . The network-side voltages u _s1 , u _s2 and U _s3 are determined and corresponding switching signals are generated in such a way that the voltage difference between u _s1 and u _f1 , u _s2 and u _f2 and between u _s3 and u _f3 assumes the desired value. The indexing 1 , 2 , 3 identifies the variables assigned to the respective phase. By regulating the voltages at the converter 20 , the current flowing through the semiconductor valves 21 can be regulated and the active filter 10 can accordingly be operated as a kind of regulated current source. The semiconductor valves 21 used are preferably IGBTs (Insulated Gate Bipolar Transistors). Two semiconductor valves 21 are assigned to each phase, each of which can assume two different switching positions.

If one of the two semiconductor valves 21 of the respective phase shown in FIG. 2 is in the switching state OPEN, the other semiconductor valve 21 is in the switching state CLOSED and vice versa. If the upper of the two semiconductor valves 21 shown in FIG. 2 of a phase in the switching state OPEN corresponds to the lower one in the switching state CLOSED, this switching combination of the two semiconductor valves 21 is identified by a +. If the opposite is the case, i.e. the upper valve is in the CLOSED switching state and the lower valve is in the OPEN switching state, this switching combination is identified with a -.

In view of the three phases present, this results in 2 ³ possible switching combinations of the total of six semiconductor valves 21 of the converter 20 . These eight switching options of the converter 20 are shown in FIG. 3 in the context of a room pointer representation. Starting from the center of a hexagon or hexagon, each switching combination of the three phases is illustrated by its own switching vector, the length of which corresponds to the voltage value u _{dc of} the DC voltage circuit. The combinations +++, ---, + -, ++ -, - + -, - ++, - + and + - + occur as switching vectors. The vectors +++ and --- are zero-length vectors. Since in a three-phase system the summation of all currents or voltages always results in the value zero and therefore only two independent variables occur, both currents and voltages can be from a three-phase static reference system with three corresponding to the individual phases and 120 ° against each other twisted axes intersecting in a common origin are transferred or projected onto a two-dimensional static Cartesian reference system with an α, β axis. All three phase quantities, for example the reference voltages, taken together result in the α-, β-reference system resulting and rotating counter-clockwise with the angular frequency ω, which corresponds to the angular velocity of the fundamental voltage oscillation, resulting and rotating reference voltage vector u * _f , from which together with the respective active switching vector the inductive voltage drop u _Lf at the choke and thus the switching ripple can be derived, cf. Fig. 3.

To generate the switching signals for actuating the semiconductor valves 21 , cf. Fig. 2, the power converter 20 is a pulse width modulation (PWM) is used.

A power electronic voltage intermediate circuit converter controlled by means of pulse width modulation, as shown in FIG. 4, has an alternating current or voltage circuit 40, a direct voltage intermediate circuit 41 and controller 42 for switching the converter or converter 20 in a known manner.

The method according to the invention and the associated device are modular constructed and allow due to their individual processing modules known methods and devices advantageously a simulation, analysis and coordination or conditioning of individual circuit components and / or control mechanisms.

Basically, a control device 42 used to control an active filter 10 generates three reference voltages corresponding to the individual phases, which are used as the basis for the switching signal generation by means of the PWM method. In the context of the PWM method, as shown in FIG. 5, the three reference voltages u * _f1 , u * _f2 and u * _f3 corresponding to the individual phases are compared with a triangular voltage 50 , the frequency of which corresponds to the switching frequency of the converter 20 . The course of the triangular voltage 50 and of the three reference voltages u * _f1 , u * _f2 and u * _f3 is plotted in FIG. 5 over a period of two sampling cycles over time. If the value of the respective reference voltage exceeds the actual value of the delta voltage 50, then the semiconductor valves 21 , cf. 2, the respective phase the switching combination + A, the reference value but is below the actual value of the triangular voltage 50, so 21 take the semiconductor valves of the respective phase the switching combination -. A. The above behavior applies to all three phases. This results in 50 different switching vectors, such as +++, ++ -, + - or ---, depending on the relationships between the reference voltages and the applied delta voltage, cf. Fig. 5. Since the value of the delta voltage is always smaller than that of the reference voltages u * _f1 , u * _f2 and u * _f3 in the time interval between points A and B, the two semiconductor valves 21 of the three phases are each in the switching state + and form accordingly the switching vector +++. If the value of the delta voltage 50 now exceeds the reference voltage u * _f3 , then the semiconductor valves 21 of the third phase form the switching combination at point B - and the switching vector ++ - results. The course of the switching waveform of the second phase 51 can thus be derived and follows a square-wave voltage, as shown in FIG. 5.

As can be seen on the basis of the flow diagram shown in FIG. 6, in one or more process preparatory steps V, the selection or detection or determination of the power electronic circuit structure corresponding to the respective task or problem and the associated processing or process modules is carried out. The input variables to be used as the basis for the procedure and the default values to be met are also recorded and determined.

The input variables to be recorded can be, for example experimentally acquired measured data, either directly, on-line via a suitable Interface can be read and processed, or indirectly, offline by one Data storage on which they were previously saved can be called up. Also there is the possibility to create or generate artificial data sets to process.

In preparation, it must also be stated whether the method according to the invention is online or off-line. line should be used, since this means that the selection of those to be used is proportional Processing modules is determined or influenced.

• - die Spannungswellenform der Versorgungsspannung an der Stelle, an der das aktive Filter 10 in den Schaltungsaufbau implementiert wird,
• - die Stromwellenform der Lastströme, die kompensiert werden sollen, wobei stets die maximal auftretende Last angenommen werden sollte, oder alternativ hierzu die Effektivwerte der Grund- und Oberschwingungsanteile der Lastströme,
• - die Frequenz f der Versorgungsspannung,
• - die höchste zu kompensierende Oberschwingungsordnung,
• - die maximal zulässige Stromwelligkeit, in Prozentangabe oder als Absolutwert,
• - der maximal zulässige Welligkeitswert der Gleichspannung, in Prozentangabe,
• - der gewünschte maximale Modulationsindex,
• - die maximale Umgebungstemperatur,
• - Spezifikationsdaten der eingesetzten IGBT's und der ihnen zugehörigen elektrisch antiparallel geschalteten Dioden

In the case of an active filter system, for example, the following input variables can be used for the comprehensive determination of optimized operating parameters and a corresponding, optimized design or conditioning of the present circuit:

• the voltage waveform of the supply voltage at the point where the active filter 10 is implemented in the circuit structure,
• the current waveform of the load currents which are to be compensated, the maximum load which should always be assumed, or alternatively the effective values of the fundamental and harmonic components of the load currents,
• - the frequency f of the supply voltage,
• - the highest harmonic order to be compensated,
• - the maximum permissible current ripple, in percentages or as an absolute value,
• - the maximum permissible ripple value of the DC voltage, in percent,
• - the desired maximum modulation index,
• - the maximum ambient temperature,
• - Specification data of the IGBTs used and the associated diodes connected electrically in anti-parallel

• - Referenzströme i*_f des aktiven Filters 10
• - Schaltfrequenz f_sw des aktiven Filters 10
• - Induktivität L_f der Eingangsdrossel
• - Spannungs- u_dc sowie Stromwert i_dc des zugehörigen Gleichspannungzwischenkreises
• - Kapazität C_dc des Gleichspannungzwischenkreises
• - beim off-line Einsatz eine Spektralanalyse der Spannungsoberschwingungen des eingesetzten Stromrichters
• - beim off-line Einsatz eine Spektralanalyse der Stromoberschwingungen des aktiven Filters 10
• - Rekonstruktion der Spannungs- und Stromwellenform des aktiven Filters 10
• - Schaltsignale zum Ansteuern des Stromrichters 20
• - Individuelle Ströme der eingesetzten Halbleiterbauelemente
• - Verlustleistungen
• - Auslegung oder Konditionierung eines vorzusehenden Kühlkörpers

After recording the default values, the circuit structure and the corresponding processing modules corresponding to the respective problem or task, as well as the corresponding input variables, the following actions or processing steps are carried out step by step in the corresponding processing modules, for example, and the following, optimized for the respective circuit and the predetermined input variables Operating parameters determined:

• - Reference currents i * _{f of} the active filter 10
• - Switching frequency f _{sw of} the active filter 10
• - Inductance L _{f of} the input choke
• - Voltage u _dc and current value i _{dc of} the associated DC voltage intermediate circuit
• - Capacitance C _{dc of} the DC link
• - In the case of off-line use, a spectral analysis of the voltage harmonics of the converter used
• - When used off-line, a spectral analysis of the current harmonics of the active filter 10
• - Reconstruction of the voltage and current waveform of the active filter 10
• - Switching signals for controlling the converter 20
• - Individual currents of the semiconductor components used
• - power losses
• - Design or conditioning of a heat sink to be provided

In addition, depending on the respectively determined optimized Operating parameters and an automated taking into account the entered default values Selection of suitable, or alternatively in the predetermined circuit structure Components to be implemented in addition from one or more, correspondingly pre-structured and configured databases to achieve a best possible conditioning and thus performance of the respective power electronic circuit.

Taking into account the specification data of the selected component (s), in particular of semiconductor components and components, in particular cables and Bus connectors or contactors are now determined based on the Operating parameters and the known specification data further component and Circuit-specific parameters such as peak current compatibility, power loss, Heat development or heat loss to be determined and taken into account accordingly, for example by selecting and implementing suitable coolants or Heatsink.

An optimized determination, dimensioning and selection of suitable precharging resistors for vibration damping and suppression of the inrush current when the active filter 10 is switched on is possible in this way, taking into account the operating parameters determined to date and the respective component specification.

The determined frequency spectra as well as all the current and voltage profiles determined and reconstructed simulated within the scope of the method according to the invention, for example the active filter 10 and the dc-side capacitor 22, are graphically represented and output or displayed and optionally stored on a data memory.

The optimized operating parameters determined in the individual processing modules are then compared with the default values. Exists within predetermined tolerances a match between the optimized Operating parameters and the default values, the process ends with the display of the determined Operating parameters and a parts list of the selected components and the component specifications and / or output, for example on a Printer or plotter and / or storage on a data storage device.

However, there is no match within the predetermined tolerances existing differences are shown and the procedure if necessary after modification of the operating parameters and / or the circuit structure Run through again as needed.

The step-by-step determination of the various operating parameters, which are optimized for a circuit with an active filter 10 for harmonic compensation and the input variables, and the selection of suitable components by the respective processing modules involved in the method are described in detail below with reference to FIG. 6.

After selection or detection or determination 60 of the corresponding power electronic circuit structure and the associated processing or process modules, the input variables to be used as the basis for the process and default values in a process preparatory step V, as shown in FIG. 6, in a first process step S1 based on a recorded three-phase load current, cf. Fig. 7b which here serves as an input signal or size, in a processing module 61, the reference currents i * _f for controlling the active filter 10 is determined.

For this purpose, the active component of the fundamental current oscillation of the load current is determined by means of a processing module 61 with one or more control and / or processing elements and subtracted from the load current i _l . The resulting currents are then inverted, which results in the reference currents i * _f for driving the active filter 10 .

It should be noted that in the case of an analog control device, the recorded data quantity should correspond to ten times the switching frequency f _{sw of} the converter 20 , whereas in the case of a digital control 42 it only has to correspond to the simple sampling frequency of the control. In the case of digital control, this means that, for example, for each phase of a 50 Hz load current, with an exemplary switching frequency f _{sw of} the converter 20 of 6450 Hz and a corresponding sampling frequency of the control 42 , which is twice the switching frequency f _{sw of} the converter 20 , ie corresponds to 12900 Hz, a data record of 258 data points must be available. Since in a three-phase system one of the three phases results from the other two, it is sufficient to record the required amount of data only for two of the three phases.

The above applies to both the load currents and the supply voltages u _s .

In the event that the amount of data recorded exceeds the required amount, a new sampling cycle is carried out in the case of digital control, so that the resulting data record or the resulting amount of data finally corresponds to the sampling frequency of the controller 42 .

If no experimental data regarding load current i _l and supply voltage u _{s are} available or available, they can also be artificially generated or processed using appropriately implemented processing modules with knowledge and use of the root mean square values of the supply voltage u _s or load current i _L and their harmonic components. be generated.

A Fast Fourier Transformation (FFT) is used to determine the fundamental current oscillation of the load current i _l in the processing module 61 , which satisfies the following equation,


where k corresponds to the order of the harmonic, n the order of the sampling, I _l the load current, i _l (n) the load current of the nth sampling and N the total number of samples taken and the number of harmonics.

The individual phases of the load current i _{1 used as} an example are shown in FIG. 7.

All that is sufficient is a calculation of the real part of the Current fundamental oscillation of the load current.

To calculate the real component of the fundamental current vibration, both its imaginary component and any harmonic components are set to zero or assigned the value zero. In order to obtain the active components of the basic oscillation of the load current i _l1re in the respective time range, an inverse Fourier transformation according to equation 2 is now carried out.


where k corresponds to the order of the harmonic, n the order of the sampling, I _l the load current, i _l (n) the load current of the nth sampling and N the number of samples taken and the number of harmonics.

Since the input values on which the inverse Fourier transformation is based only from real numerical values are formed, the determined results only show real numerical values.

The reference currents i * _f (n) of the nth sampling of the active filter 10 result accordingly

i * _f (n) = - (i _l (n) - i _l1re (n)) 1 ≤ n ≤ N, equation 3,

The reference currents i * _{f determined in this way} and the compensation currents i _{f of} the active filter 10 formed from them are shown in FIG. 8.

The optimal switching frequency f _{sw of} the active filter 10 is now determined in a second processing module 62 . This results from the frequency of the highest order current harmonic to be compensated. According to theoretical considerations, it is possible to regulate or compensate harmonics up to a frequency half the switching frequency f _sw . Consequently, the switching frequency f _sw must be twice the frequency of the maximum harmonic to be compensated. In the following example a switching frequency f _sw of 6.45 kHz was chosen.

In the case of a digital control 42 , which is to be assumed here for all further considerations as an example, the supply voltage u _{s is applied} in a third processing module 63 with a phase advance of, for example, 1.5 sampling periods. In this case, time delays and dead times of the control 42 and the scanning are taken into account by the phase feed.

To compensate or compensate for or correct any delays, the supply voltage u _s (n) of the nth scan is determined on the basis of the voltage values u _s (n + 1) of the next scan and u _s (n + 2) of the scan after next in accordance with equation 4.


where n denotes the order of the samples or sampling points and u _s (n) the respective sampled supply voltage u _s .

The exemplary time profile of the supply voltage u _s prepared in the aforementioned manner is shown in FIG. 7a.

In a fourth processing module 64 , both the inductance L _{f of} the input choke of the active filter 10 is determined based on a current ripple peak value î _well to be determined or specified in advance, and also its permissible resistance value R _f . To calculate the peak value of the current ripple î _well , the absence of any load currents i _l is initially assumed or assumed. Under this condition and a negligible ohmic resistance R _{f of} the choke of the active filter 10 , the reference voltage u * _{f of} the converter corresponds to the supply voltage u _s .

The difference vector between the supply voltage u _s or the reference vector and the switching vector corresponds to the space vector of the voltage drop at the choke of the active filter 10 , cf. Fig. 3. This voltage drop is largely responsible for the current ripple.

Theory and simulations show that the peak value of the current ripple î _well is reached when the actual voltage value of a phase of the supply voltage u _{s is} close to zero. As from FIG. 3, the second phase of the supply voltage reaches a value of zero at an angular position of π / 6.

From the aforementioned points of view, the three phases of the supply voltage result as follows when the reference vector has an angular position of π / 6. For ωt = π / 6:


where u _s1 , u _s2 , u _{s3 denote} the supply voltages of the respective phase, U _s the effective value (rms value) of the supply voltage per phase and ωt the position indicator of the supply voltage in the three-phase static or stationary, non-rotating 123 reference system. The 123 reference system has three axes that correspond to the individual phases and are rotated by 120 ° with respect to one another and intersect in a common origin.

Under the same condition, for the three phases of the reference voltage u * _f and taking the _{DC link voltage} u _dc into account:

Accordingly, the switching times or the time durations t _{sw of} the respective switching states can be exemplified using the sine / triangle modulation, cf. . Determine Figure 5, as follows:


where u * _f1 , u * _f2 , u * _f3 the reference voltages of the respective phase, U _s the effective value (rms value) of the supply voltage per phase, ωt the space vector position of the supply voltage, u _dc the dc voltage value of the DC link, f _sw the switching frequency and t _sw0 , t _sw1 , t _sw2 , t _{sw7 denote} the switching times or time periods of the individual switching vectors 0, 1, 2 and 7 in the three-phase static or stationary, non-rotating 123 reference system.

Depending on the modulation _method , the switching time t _{sw0 of} the zero vector changes, but the sum remains constant over all switching times. The switching times t _sw1 and t _{sw2 are} retained regardless of the respective modulation _method .

This results in the peak value of the current ripple of the second phase selected as an example:


where î _f2 the peak value of the current ripple of the second phase, U _s the effective value (rms value) of the supply voltage per phase, u _dc the dc voltage value of the DC _link , f _sw the switching frequency, L _f the inductance of the input _choke and t _sw2 the duty cycle in designate three-phase static or stationary, non-rotating 123 reference system.

The peak value of the current ripple of the second phase î _f2 essentially corresponds to the summation of the current ripples of all three phases, so that the inductance L _{f of} the input choke is therefore described by the following relation:


where î _{well is} the default value of the permissible peak value of the current ripple, f _sw the switching frequency, L _f the inductance of the input choke and U _s the effective value (rms value) of the supply voltage per phase.

At this point it should be noted that the determined inductance L _{f of} the choke or inductor depends not only on the switching frequency f _{sw of} the supply voltage u _s but also on the desired maximum permissible value of the current ripple î _well , but at zero voltage _crossing regardless of the DC voltage value u _{dc of} the dc -Circle 41 is.

With an effective value of the supply voltage U _s of 230 V per phase, a switching frequency f _sw of 6.45 kHz and an exemplary current ripple î _well of 15 A, this results in an inductance L _{f of} the choke of 485 µH, which is rounded up to 500 µH becomes.

The peak value of the current ripple î _well can be viewed, for example, as a percentage of the current measured for the IGBT.

The upper limit of the permissible resistance value R _{f of} the choke is determined on the basis of the switching frequency f _{sw of} the active filter 10 . The maximum time interval within which a switching vector can be switched corresponds to half the switching time or duration. Within this time interval, the current should have a curve that is as linear as possible, so that the control 42 can detect the nominally correct value. Depending on this criterion, the throttle has a resistance value R _f of

R _f << 2f _sw L _f .

In the example on which this is based, the resistance R _{f of} the choke should accordingly be less than 6.45 Ω, which in practice is fulfilled for almost all chokes or induction coils.

Then, taking into account the previously determined operating parameters, in particular the inductance L _f and the resistance value R _f , and any default values, for example cost-relevant aspects, suitable input chokes or coils are selected from a correspondingly preconfigured database 64 b.

In a second method step S2, the optimal voltage _value u _{dc of} the DC link is determined.

The determination of the minimum permissible DC voltage value is of crucial importance for the calculation of the required reference voltage values u * _f for driving the active filter 10 .

For this purpose, the reference voltage u * _{f of} the active filter 10 is determined according to equation 10 by means of a fifth processing module 65 .


where u * _f the reference voltage of the active filter 10 , u _s the supply voltage, L _f the inductance of the input choke and di _f / dt the time derivative of the current i _{f of} the active filter 10 .

The time derivative of the current h of the active filter 10 is given by:


where di _f / dt the time derivative of the current of the active filter 10 , i _f the current of the active filter, n the order of the respective sampling and f _sw the switching frequency.

Because of a delayed sampling and a delay caused by the control process, the load current i _l is also subjected to a phase advance of, for example, 1.5 sampling periods. A sampling period denotes the time period between two successive samples. The rate of change of the load current i _l to be expected with a delay of 1.5 sampling periods is determined in accordance with equation 10 via a load current sampling of the next i _l (n + 1) and the next but one i _l (n + 2) order.

The determined reference voltages u * _{f of} the three phases are plotted in FIG. 9 over the order n of the sampling carried out.

Using a further, sixth processing module 66 , the necessary voltage _value u _{dc of} the DC voltage intermediate circuit 41 of the active filter 10 is then determined in accordance with equation 12, taking into account space vector modulation.


where u _{fmax denotes} the maximum peak _{voltage value of} the converter 20 to be switched, as shown in FIG. 9, and u _dc the minimum permissible voltage value of the dc circuit 41 .

As a result, there is a minimum allowable for the example selected here DC link voltage of 747 V, but rounded up to a smooth value of 750 V. becomes.

If the method is used on-line, that is to say directly in the context of a control system, there is no time-based consideration and analysis of individual current and / or voltage harmonics in the frequency domain caused by the switching process, and in a second method step S2 it is only achieved by means of a corresponding processing module 67 Approximately, an effective current _{ripple value} I _{lrip of} the choke or coil of the active filter 10 is determined according to equation 13, based on a space vector modulation.

Here U _s denote the effective value (rms value) of the supply voltage per phase, u _dc voltage value of the DC link 41 , L _f the inductance of the input choke and f _sw the switching frequency.

For the example presented here, there is accordingly an effective one Current ripple value of 4.85 A.

The current value i _{dc of} the DC or DC voltage intermediate circuit 41 is then calculated.

For this purpose, the converter or converter voltages and currents are either in Frequency space or used in the time-based reference space.

In the frequency domain, the direct current calculation is based on the mathematical model of a convolution, whereas in the time-based reference space, a calculation according to equation 14 can be carried out.

Here, i _dc denote the DC circuit-side current of the active filter 10 , u _dc the dc voltage value of the DC voltage intermediate circuit 41 and u _f1 , u _f2 , u _f3 the voltages of the active filter 10 in the three-phase static or stationary, non-rotating 123 reference system.

The previously in the sixth processing module 66 , cf. Fig. 6, determined DC voltage value u _dc is assumed to be constant. Even if it were not constant, this would compensate for a corresponding regulation 42 by normalizing the reference values u * _f by means of the current DC voltage value. For this reason, the current values i _{dc are} always close to the calculated quantities, the exemplary course of which is shown in FIG. 31.

The averaged i _avg and effective i _rms direct current _values result from equation 15.


where i _{avg is} the _mean current value, i _{rms is} the rms current value, n is the order of the sampling, N is the total number of samplings carried out and i _{dc is} the dc-side current of the active filter 10 .

When using the current _{ripple value} I _{lrip of} the choke and observing the energy conservation law, the current _ripple I _{Cdcrip of} the DC _link 41 results in:


with the effective value (rms value) U _{s of} the supply voltage per phase and the voltage value u _{dc of} the DC voltage circuit 41 .

For example, a value of 4.48 A results for the current ripple and consequently a value of 25.16 A for the total effective current of the capacitor 22 of the DC voltage intermediate circuit 41 .

In the case of off-line, that is to say indirect application or use of the method according to the invention, it is advisable, in order to ensure the highest possible accuracy, to consider relevant variables in the frequency domain and to take account of harmonics which occur as a result of switching the converter 20 , and it is to be ascertained here that the current ripple can be determined by analyzing the Current harmonics of the active filter 10 is determined.

For this purpose, in the second process step S2, based on the online process, alternatively to carry out the following processing steps for the on-line method or processing modules.

First, in the second method step S2, the voltage harmonics of the converter 20 of the active filter 10 are determined by means of an eighth processing module 68 , wherein it should be noted that the reference voltages u * _f can be processed in the same way as in a real converter 20 , ie that the behavior of a real converter 20 is to be simulated as realistically as possible here. In most cases, the space vector modulation method is used for this, the reference values u * _{f being} normalized by means of the DC voltage value u _{dc in} order to achieve a scaling of the reference values u * _f between -1 and +1. Each sampled reference voltage value u * _{f of} the space vector modulation can be newly determined according to the following matrix combination:

The standardized, digital reference voltages u * _f determined according to equation 17 are shown in FIG. 11.

For a spectral analysis of the voltage harmonics, the switching angles or angular positions of the space vector during switching are first determined, it being assumed that the semiconductor valves 21 of the converter 20 exhibit ideal behavior. Compared to a trapezoidal switching curve with finite switch-on and switch-off times, ideal switching behavior arithmetically leads to the worst possible harmonics. If the switching angle α _n is determined with sufficient accuracy, the harmonic components determined are also determined with sufficient accuracy and the achievable accuracy is only limited by the resolution or the performance of the data processing system used. The normalized triangular voltage 50 used in pulse width modulation, cf. Fig. 11 is here presupposed as running synchronously with the fundamental wave of the supply voltage, a cosinusoidal curve was adopted by the fundamental component of the supply voltage.

The switching times t _n result from the reference voltages u * _f according to FIG. 11 and equation 18:


with the switching frequency f _sw and the reference voltage u * _f (n) of the sampling with the order n.

The corresponding angles α _n result in


with the switching frequency f _sw , the fundamental frequency f and the reference voltage u * _f (n) of the sampling with the order n.

Neglecting the dc components, the time-dependent voltage u _f (t) of the active filter 10 can be developed as a Fourier series according to equation 20.


with the coefficients


with the harmonic order k and the scanning order n, where the Fourier coefficients of a three-phase converter 20 can also be modified as follows:

By means of the above-mentioned modification, any dc components, which are for the sake of simplicity were initially neglected, eliminated.

At this point it should be noted that with the pulse width modulation method it is not necessary to calculate the harmonic components to infinity. Rather, it is advantageous, depending on the accuracy sought, to set an upper limit on the order Nh on which the determination is to be based. In the example given here, this is 25.5 times the switching frequency f _sw .

The voltage harmonics result from the determined Fourier coefficients according to equation 23.

U _f (k) = A _k - jB _k Equation 23

Now the current harmonic components I _f (k) of the active filter 10 or its converter 20 in the frequency domain are determined by means of a further, ninth processing element 69 . For this purpose, the supply voltage is first Fourier transformed to obtain the harmonic components of the supply voltage u _s , cf. Fig. 12 and 13.

Taking into account the previously determined voltage harmonics U _f (k) and the voltage harmonics of the transformed supply voltage U _s , the current harmonics I _f (k) of the active filter 10 result in the frequency domain according to the following relation:

The compensation currents i _{f of} the active filter 10 have a fundamental component which corresponds to the reactive component of the load current i _l . It should be noted here that some load-related harmonic components are still preserved in the supply current, but their amplitude is very small. The reason for this is an amplitude error in the sampled signal curve that occurs with every harmonic. This can also be observed in practice. When using frequency-based control mechanisms, however, this can be influenced and reduced.

The current and voltage harmonics determined in the frequency or phase space are then transformed back in a tenth processing module 70 by means of an inverse Fourier transformation into the time-based reference space, thereby reconstructing their waveform in the time-based reference space and calculating corresponding rms values i _rms , cf. Fig. 20 and 21.

Alternatively, the effective rms values i _rms can also be calculated in the frequency domain.

Taking into account the determined effective values i _rms , the copper windings of the input choke of the active filter 10 , the busbar webs and the cables to be used are then conditioned and selected in terms of material and cross section. The effective value-dependent selection takes place automatically from one or more correspondingly preconfigured databases 70 b.

Due to a limited number of harmonics that were used for the inverse Fourier transformation, isolated sharp edges can be observed in the converter voltages u _f . However, the results obtained meet all practical applications in terms of their accuracy.

The DC _value i _dc is again calculated in an eleventh processing module 71 . For this purpose, the converter voltages u _f and currents i _{f are used} either in the frequency domain or in the period.

In the frequency domain, the direct current calculation is based on the mathematical model of a convolution, whereas in the period or time-based reference space, a calculation is carried out according to the following equation:

The DC voltage value u _dc determined in the sixth method module 66 is assumed to be constant. Even if it were not constant, this would compensate the control 42 by normalizing the reference values u * _f using the current DC voltage value.

The converter voltage u _f determined by means of the voltage harmonics of the converter 20 is used to calculate the current value i _{dc of} the DC voltage intermediate circuit 41 . This means that the calculation of the direct current i _{dc is} carried out effectively in the frequency domain. In order to save or shorten the computing time, however, the direct current i _dc is transferred to the time-based reference space or period. The averaged and effective DC currents are calculated according to Equation 26.

For the example given here, the off-line method accordingly results effective DC current value of 28.58 A.

The harmonic spectrum of the current i _dc in the DC link 41 is shown in FIGS. 29 and 30.

For the conditioning of the capacitor (s) 22 of the DC voltage intermediate circuit 41 , the determined current value i _{dc is} first integrated over time in a further processing module 72 to determine the current ripple in both the on-line and off-line method. Since an integration is a type of summation process, the ripples caused by switching the converter 20 are largely averaged out, see FIG. 32.

The peak-to-peak ripple value of the integrated direct current is 0.0358 As.

The capacitor or capacitors 22 of the DC voltage intermediate _circuit 41 are now conditioned in such a way that the peak-to-peak ripple _value u _{dcpp is} limited to a predetermined percentage of the nominal DC voltage _value u _dc .

The total capacitance C _dc of the capacitor (s) 22 or of the dc circuit 41 used is accordingly:

For example, if the peak-to-peak value u _{dcpp is} to be limited to 0.5% of a nominal DC voltage value of 750 V, the capacitance of the capacitor 22 is 0.0095 F according to equation 28.

Another relevant parameter that has to be taken into account when conditioning and selecting a corresponding capacitor 22 , in addition to the required capacitance C _dc , is the rms current value that has already been determined. In the example given here 28.58 A.

After a suitable capacitor 22 has been selected from a preconfigured database 73 b, ESR effects or effective series resistance effects, DC ripples and losses can be determined taking into account its specification data.

Due to the fact that the reconstructed voltage curves themselves have certain errors, their further use, for example for determining individual currents flowing through the semiconductor components used, in particular the IGBTs and the diodes of the semiconductor valves 21 electrically connected in parallel, would likewise lead to erroneous results.

For the off-line approach, a determination of the individual currents flowing through the respective semiconductor components of the active filter 10 is therefore carried out in a further processing module 74 based on a digital reconstruction of the switching signals 110 , cf. 11, with a value of one or zero, α for the control of the active filter 10 while maintaining an identical number of sampling points as that of the inverter voltage and reconstructed using the switching angular positions _n conducted. In time-based reference space.

If the switching signals 110 , cf. 11, once reconstructed. It is relatively easy to determine the 20 currents flowing through the individual semiconductor valves 21 of the converter. If the switching signal 110 is high, at the value one, the upper IGBT of the corresponding phase is switched on, that is to say in the switching state OPEN and the lower one is off, that is to say in the switching state CLOSE. The same applies in the opposite case. If the upper IGBT is switched on or on, in the switching state OPEN, and a positive current flows, the upper diode (IGBT) is switched in the forward direction, a negative current flows, then this flows through the upper IGBT. The same logic is used when locating the currents flowing through the lower IGBT and the lower diode when the switching signal is at low or at the value zero.

The temporal current profiles reconstructed in this way are shown in FIGS. 22 to 25.

For online use, an alternative, simplified view is again carried out or taken into account in the processing module 74 , without taking into account the influences caused by the switching frequency. If the reference voltage is positive, the upper IGBT of the respective phase is switched on, and if the reference voltage is negative, the lower IGBT. If the upper IGBT is on and the current is positive, the upper diode conducts and if the current is negative then the current flows through the upper IGBT. Corresponding considerations apply to the determination of the current through the lower IGBT or the lower diode if the reference voltage is negative. The resulting temporal current profiles are shown in FIGS. 26 and 27.

The averaged and effective currents based on a total of N samples are determined according to Equation 29.

The values resulting here by way of example can be found in the following Table 1. Table 1

In the example given here, the peak current value flowing through the semiconductor valves 21 is 73.77 A.

Since both the voltage value u _{dc of} the DC voltage intermediate circuit 41 and the peak value of the compensation current of the required power electronic component are now known, the same component is now selected automatically from a preconfigured database 74 b, taking into account the aforementioned operating parameters and any default values.

It should be noted that some manufacturers look at their specification data regarding the current compatibility of your products to a housing temperature of 25 ° C, others, however, refer to 85 ° C.

Currents that are temporarily permissible in the semiconductor components are often significantly larger as currents flowing continuously or over a longer period of time. A short-term, permissible current of, for example, two milliseconds is usually twice as high here big as a continuously flowing stream.

A matter of increasing or increasing the efficiency or the benefit of the respective component can be used. Power losses that occur as well as coolant provided for cooling the respective component also the permissible current measurement.

The specification data of the semiconductor components of the type SEMIKRON SkiiP 342 GD 120-314 CTV used as examples in the active filter system can be found in the following table. Table 2

For the selected IGBTs, the expected switching and forward losses are calculated in a further processing module 75


determined.

The power losses of the individual diodes are calculated according to


certainly.

Here V _TO denotes the voltage _threshold value, r _TO the dynamic equivalent resistance value, V _CC the collector-emitter supply voltage, I _C the continuous collector current, V _CCref the reference value of the collector-emitter supply voltage, I _Cref the reference value of the continuous collector current, V _{CE (O)} the collector-emitter threshold voltage (static), I _Crms the effective value (rms value) of the collector current, E _on the energy loss during the switch-on period or phase, E _off the energy loss during the switch-off period or phase, P _swIGBT the switching losses in the IGBT , P _swDIODE the switching losses in the diode, P _condIGBT the _forward losses in the IGBT and P _condDiode the _{forward losses} in the diode.

Switch-on losses generally occur in the case of a diode Neglecting switch-off losses, the losses occurring at the IGBTs still are significantly higher than diodes. The thermal resistance between The barrier layer and housing is significantly larger for diodes than for IGBTs. For this reason it is advisable to always determine and check both junction temperatures, which of the two limits the compensation current upwards.

The junction temperatures can be calculated online and based on them a limitation of the peak current value can be initiated or carried out.

An off-line determination of the upper limit of the peak current value and one Determination of the permissible time interval for which this upper limit is taken or may also be allowed.

In most cases, the maximum permissible in stationary operation Junction temperature at 125 ° C and the housing temperature at 80 ° C.

The given equations for the calculation of occurring losses offer a Accuracy that is considered sufficient for most applications. If necessary however, a more exact calculation model can also be used here.

The resulting power losses of both the IGBTs and the electrically antiparallel connected diodes can be seen in the following table. Table 3

Here too, depending on the operating temperatures to be observed, the geometric dimensions or a corresponding thermal equivalent circuit diagram and any specifications that have been made, selection of suitable coolants or bodies is carried out automatically from a correspondingly preconfigured database 75 b.

The heat sink selected as an example based on the available data is referred to as P16, its thermal resistance to the environment is 0.036 ° C / W. This requires the use of a SKF 16B-230-01 blower. Assuming an ambient temperature of 45 ° C, this results in a housing temperature or heat sink temperature close to the component of

T _S = 174.96 × 0.036 + 45 = 51.3 ° C.

The junction temperature of the IGBTs and the diodes is accordingly

T _jIGBT 21.76 × 0.09 + 51.3 = 53.25 ° C,

T _{j diode} = 7.4 × 0.25 + 51.3 = 53.15 ° C,

where T _{jIGBT is} the junction _{temperature of} the IGBT and T _j diode is the junction _{temperature of} the diode.

When the converter 20 is switched on for the first time, the electrical system formed in this way is equivalent to a three-phase diode rectifier. For this reason, vibrations could occur between the input choke and the capacitor 22 on the DC voltage intermediate circuit 41 , which can lead to the occurrence of overvoltages in the capacitor 22 , which can damage it sustainably.

To prevent this and to dampen vibrations at times is a Preload resistance is required, which is also used to suppress the Inrush current ensures.

The diodes contained in the IGBT's have a significantly lower one Impulse compatibility as conventional diodes, as usually used for rectifiers Come into play.

In the case of a conventional system with a diode rectifier, the Pre-charging resistance only the task of suppressing the inrush current. In one Diode rectifiers can only do two phases at a time carry electricity at the same time. All three phases can only in the commutation phase be live at the same time, which is the case with the considerations presented here is neglected.

The transfer function between the input current i _dc and the voltage u _dc is represented by equation 32.

Comparing Equation 32 with an ordinary second order differential equation, we get


with the frequency of the vibration ω _n to be damped, the resistance value of the precharge resistor R _p , the inductance of the input choke L _f and the capacitance of the capacitor of the dc circuit C _dc .

The damping factor ξ to be taken into account must be greater than or equal to one. The approximate expression for the current in the time-based reference space is:

By deriving equation 33 and equating it to zero, both the maximum direct current _value i _dcmax and the time t _max at which this value is obtained. The course of the resistance value R _p , the time t _max and the current value i _dcmax as a function of the damping _factor ξ is shown in FIG. 34.

If a precharge resistance of 10 ohms is selected from a preconfigured database 73 b, the inrush current is in the range of 28 A and thus within acceptable limits.

The following table shows a comparison of the optimized operating parameters ascertained according to the invention by means of on-line methods, that is, without taking into account harmonics caused by the switching and the harmonic components, which are determined by means of off-line methods, that is, taking into account the harmonic components caused by switching of the converter 20 . Table 4

The optimized operating parameters determined in the individual processing modules 60 to 75 are then compared with the default values by means of a further module 76 . If there is agreement between the optimized operating parameters and the default values within predetermined tolerances, the method ends with display 77 of the determined operating parameters and a parts list of the selected components and components as well as the component specifications and / or output, for example on a printer or plotter and / or storage on a data storage.

However, there is no match within the predetermined tolerances existing differences are shown and the procedure if necessary after modification of the operating parameters and / or the circuit structure Run through again as needed.

In Fig. 7a, the supply voltages u _s1 , u _s2 and u _s3 on which the method is based as an example as input variables, and in Fig. 7b the load currents i _l1 , i _l2 , i _{l3 of} the three phases in amperes (A) are dependent the order or number of samples taken.

In FIG. 8, the compensation determined or filter currents I _f1, I _{f2, f3} i of the three phases of the active filter 10 used for harmonic compensation in dependence on the order or number of sampling carried forth.

In Fig. 9, the determined reference voltages u * _{f1, f2} u, u _f3 shown the scanning performed * of the three phases of the active filter 10 in response to the order or number.

In Fig. 10, and by means of the predetermined DC voltage value u _dc normalized reference voltages * _f1, u _{f2, f3} shown u * of the three phases of the active filter 10 for the pulse width modulation of space vector. Plotted here are the voltage values in volts as a function of the number or order of the scans carried out.

FIG. 11 shows the reconstruction of the digital switching signals 110 in the time-based three-phase static or stationary, non-rotating 123 reference system.

Basically, a control device 42 used to control the active filter 10 generates three reference voltages corresponding to the individual phases, which are used as the basis for the switching signal generation by means of the PWM method. In the context of the PWM method 11 will now, as shown in Fig. U of the respective reference voltage value * _f (n) of the n-th sample with the actual value of the normalized delta voltage 50, the frequency of the switching frequency f _sw of the converter 20 corresponds compared , The course of the triangular voltage 50 and the step-like course of the reference voltage 110 are plotted over time in FIG. 11. If the reference voltage value u * _f exceeds the actual value of the delta voltage 50 , a switching signal of the value one is generated, which corresponds to a voltage u _{f of} the active filter 10 from the voltage value of the DC voltage circuit u _dc . However, if the reference value is below the actual value of the delta voltage 50 , a switching signal of zero value is generated, which also corresponds to a voltage value of the active filter of zero. The number or order n of the samples taken per fundamental oscillation results from:


where f _sw denotes the switching frequency and f the fundamental frequency of the supply voltage u _s .

In Fig. 12, the fundamental and harmonic components are exemplary of the supply voltage U _S of the first phase in volts as a function of the harmonic order for the first 50 orders applied.

In contrast to this, as shown in FIG. 13, a whole spectrum of voltage harmonics is already formed due to the switching behavior of the converter 20 in the converter voltage u _{f in} the first phase. Shown here is the respective fundamental and harmonic amplitude in volts as a function of the harmonic order for the first 50 orders.

The harmonic spectrum of the voltage u _{f of} the first phase of the active filter 10 , determined completely according to the invention in the context of a realistic simulation, is shown in FIG. 14. The amplitudes of the respective harmonics in volts are plotted here as a function of their order.

Accordingly, the determined fundamental and harmonic components of the current i _{f of} the first phase of the active filter 10 are shown in FIG. 15. The first 50 harmonics are plotted here, the amplitude of which is given in amperes, depending on the respective harmonic order.

The harmonic spectrum of the current i _{f of} the first phase of the active filter 10 , determined completely according to the invention in the context of a realistic simulation, is shown in FIG. 16. The amplitudes of the respective harmonics in amperes are plotted here as a function of their order.

The first 50 harmonic components of the load current i _{l of} the first phase, which are caused and determined by non-linear loads 12 , are plotted in FIG. 17. The amplitudes of the respective harmonic in amperes are also shown here as a function of their order.

In Fig. 18 is adjusted by the active filtering the fundamental and harmonic spectrum of the supply current i _s set forth the first phase. The amplitudes of proportionately remaining harmonics of higher orders are very small thanks to the successful compensation by the active filtering. The amplitudes of the respective harmonics in amperes are plotted here as a function of the respective harmonic order.

The spectrum of the harmonics remaining in the supply current i _{s of} the first phase is shown in FIG. 19 with a higher resolution for the first 50 harmonics. The amplitude of the respective harmonics is plotted in amperes above the respective harmonic order.

FIG. 20 shows an example of the time profile of the current i _{f of} the first phase of the active filter 10 determined according to the invention. The individual current values are plotted in amperes above the order or number of the respectively associated or performed sampling.

FIG. 21 shows the voltage profile u _{f of} the phase-1 voltage of the active filter 10 , reconstructed in the time-based reference space. The absolute voltage values are plotted here as a function of the number or order of the scans carried out.

In Fig. 22, the off-line, that is, in consideration of the switching frequency components of the course of the current flowing through the upper diode of the power converter 20 current of the first phase, determined as a function of the scan order.

In Fig. 23, the off-line course determined is shown of the current flowing through the upper IGBT 20 of the converter current of the first phase as a function of the scan order.

In Fig. 24, the off-line course determined is shown flowing through the lower diode of the power converter 20 current of the first phase as a function of the scan order.

In Fig. 25, the off-line course determined is shown of the current flowing through the lower IGBT of the power converter 20 current of the first phase as a function of the scan order.

If the current profiles shown in FIGS. 22 to 25 are linked, the current profile of the first phase of the active filter 10 shown in FIG. 20 is obtained.

In contrast to FIG. 22, FIG. 26 shows the current of the first phase flowing through the upper diode of the converter 20 as a function of time, based on time and without taking into account any switching frequency harmonics.

In contrast to FIG. 24, FIG. 27 shows the current of the first phase flowing as a function of time using the on-line method without taking into account any harmonics through the upper IGBT of the converter 20 .

In Fig. 28, the off-line course is set forth of the current flowing through the capacitor 22 of the DC voltage intermediate circuit 41 current. The associated harmonic spectra are shown in FIG. 29 for the first 50 orders and in FIG. 30 in full. The amplitude or strength of the respective harmonic is plotted here in amperes above the harmonic order.

In contrast to FIG. 28, FIG. 31 shows the on-line curve of the current flowing through the capacitor 22 of the DC voltage intermediate circuit 41 as a function of time.

The integrated dc current required for determining the current ripple and thus for calculating the capacitance of the capacitor 22 used in the DC voltage intermediate circuit 41 is for off-line and on-line in FIG. 32 in units of As as a function of the order or number of scans carried out applied.

The equivalent circuit diagram for the first switching on of the active filter 10 is shown in FIG. 33.

In Fig. 34 in a logarithmic scale, the course of the resistance R _p, the time t _max, as well as the current value i _dcmax shown as a function of the damping factor ξ. List of reference symbols u _dc voltage _{value of} the DC link
u _dcpp ; peak-to-peak value of the dc voltage
i _dc DC circuit current of the active filter 10
u _s1 , u _s2 , u _s3 supply voltages in the three-phase static or stationary, non-rotating 123 reference system.
u _s generalized description of the supply voltage
U _s RMS value (rms value) of the supply voltage per phase
i _s generalized term for the supply current
i _l1 , i _l2 , i _l3 load currents in the three-phase static or stationary, non-rotating 123 reference system.
i _l generalized term for the load current
i _l1re real part of the fundamental wave of the load current
i _f1 , i _f2 , i _f3 currents of the active filter 10 in the three-phase static or stationary, non-rotating 123 reference system.
i _f generalized term for the current of the active filter 10
î _f2 absolute peak value of the phase 2 current of the active filter 10
k harmonic order
n order of sampling
i _avg average current
i _rms rms _current value
u * _f1 , u * _f2 , u * _f3 Reference voltages of the active filter 10 in the three-phase static or stationary, non-rotating 123 reference system.
u * _f generalized designation for the reference voltage of the active filter 10
u _f1 , u _f2 , u _f3 voltages of the active filter 10 in the three-phase static or stationary, non-rotating 123 reference system.
u _f generalized term for the voltage of the active filter 10
u _f (t) voltage of the active filter 10 as a function of time
u _fmax Maximum peak _{voltage value of} the converter 20 that can be switched
u _Lf voltage drop across the choke of the active filter 10 with the inductance L _f
t ₁ , t ₂ . , ., t _N switching times
α ₁ , α ₂ . , , , α _N switching angles or angular positions of the space vector corresponding to the switching times
A _k1 , A _k2 , A _k3 , B _k1 , B _k2 , B _k3 Fourier coefficients for the 1, 2 and 3 phases
B _k1 , B _k2 , B _k3 generalized designation for the Fourier coefficients
j √ - 1
ω Angular velocity of the supply voltage in rad / s
ρ or ωt Space vector position of the supply voltage in the three-phase static or stationary, non-rotating 123 reference system.
ξ damping factor
ω _n resonance frequency in rad / sec
L _f input inductance of the active filter 10
R _f input resistance of active filter 10
C _dc capacitance of the dc-side capacitor 22 of the active filter 10
C _sw Capacitance of the capacitor used to dampen the ripples caused by switching
s Laplace operator
f frequency of the fundamental oscillation of the supply voltage u _s
f _sw switching frequency
V _TO voltage threshold
r _TO dynamic equivalent resistance value
V _CC collector-emitter supply voltage
I _C continuous collector current
V _CCref Reference value of the collector-emitter supply voltage
I _Cref reference value of the continuous collector current
V _{CE (O)} collector-emitter threshold voltage (static)
i ² ti ² t value
I _Crms RMS value (rms value) of the collector current
E _on energy loss during duty cycle or phase
E _off energy loss during the switch-off period or phase
P _swIGBT switching losses in the IGBT
P _swDIODE switching losses in the diode
P _{condIGBT transmission losses} in the IGBT
P _{condDIODE conduction losses} in the diode
P _IGBTs total loss or total power loss in the IGBT's
P _Diodes total loss or total power loss in the diodes
R _thjs thermal resistance between the junction and the heat sink
R _thsa thermal resistance between heat sink and environment
R _thjsIGBT thermal resistance between the junction and heat sink of the IGBT
R _thjsDiode thermal resistance between the junction and the heat sink of the diode
T _jGBT junction _{temperature of} the IGBT
T _j diode junction _{temperature of} the diode
T _S heat sink temperature close to the respective component
T _A ambient temperature
R _p pre-charge resistance value
t _max time that corresponds to the maximum precharge current
i _dcmax Maximum direct current during pre-charging
I _F mean current of the respective component or component F
I _Fref Reference _current value of the component or component F
I _Frms RMS _current value of the component or component F
10 active filters
11 three-phase supply voltage
12 load source
20 active filter converter 10
21 semiconductor valves of the converter 20
22 capacitors of the DC voltage intermediate circuit 41
40 connected ac circuit of a voltage intermediate circuit converter 20 , cf. Fig. 4
41 connected DC circuit of a voltage intermediate circuit converter 20 , cf. Fig. 4
42 Regulation for controlling a voltage intermediate circuit converter 20 , cf. Fig. 4
50 normalized triangular voltage of pulse width modulation, cf. Fig. 5
51 switching waveform or switching signal of the second phase, cf. Fig. 5
60 Module for recording default values, input variables and circuit structure, cf. Fig. 6
61 processing module for determining the reference currents i * _f for driving the active filter 10 , cf. Fig. 6
62 processing module for selecting the switching frequency f _{sw of} the active filter 10 or its converter 20 , cf. Fig. 6
63 Processing module for compensation with regard to the supply voltage u _s occurring delays
64 processing module for determining the inductance L _{f of} the choke or input choke and the input resistance R _{f of} the active filter 10 , cf. Fig. 6
64 b preconfigured database or data memory with specification data of available chokes or choke types
65 processing module for determining the reference voltages u * _f for driving the active filter 10 , cf. Fig. 6
66 processing module for determining a necessary DC voltage value u _dc , cf. Fig. 6
67 Processing module for time-based online determination of the current in the dc circuit, the dc capacitor current and the current ripple of the dc current, cf. Fig. 6
68 Processing module for frequency-based off-line determination and analysis of the voltage harmonics generated by switching the converter 20 , cf. Fig. 6
69 Processing module for frequency-based off-line determination and analysis of the current harmonics generated by switching the converter 20 , cf. Fig. 6
70 processing module for the reconstruction of the current and voltage waveform in the time-based reference system and determination of the effective current values, cf. Fig. 6
70 b preconfigured database or data memory with specification data of available material classes with regard to the copper windings of the input choke of the busbar webs and the cables to be used
71 Processing module for the off-line determination of the dc capacitor current and the current ripple of the dc current, cf. Fig. 6
72 processing module for determining the optimal capacitance C _{dc of} the dc circuit or the dc capacitor, cf. Fig. 6
72 b preconfigured database or data memory with specification data of available capacitors, cf. Fig. 6
73 Processing module for determining an optimally matched precharge resistance to limit the inrush current and damping vibrations caused by switching on, cf. Fig. 6
73 b preconfigured database or data memory with specification data of available pre-charging resistors, cf. Fig. 6
74 processing module for determining the individual currents flowing via the respective semiconductor components (IGBTs and diodes) of the semiconductor valves 21 for on-line and off-line use.
74 b preconfigured database or data memory with specification data of available power electronic components (semiconductor valves 21 , IGBTs, diodes), cf. Fig. 6
75 processing module for determining the expected power losses of the selected components and conditioning of appropriate coolants or heat sinks, cf. Fig. 6
75 b preconfigured database or data memory with specification data of available coolants or heat sinks, cf. Fig. 6
76 Module for comparing the default values with the determined operating parameters, cf. Fig. 6
77 If the specifications display, output and storage of both the parts list, the specification data and the determined optimized operating parameters are met, cf. Fig. 6
110 digital switching signals, cf. Fig. 11
111 sampled, standardized reference voltage curve, cf. Fig. 11

Claims (42)

1. A method for the simulation, analysis and conditioning of power electronic circuits, in particular systems based on voltage intermediate circuit converters ( 20 ), wherein by means of a data processing device with data storage, input device and display device, at least one interface for data transmission and for communication with at least one database and several linked to one another processing modules a) in a preparatory method step V, the input variables to be used as the basis for the method, the default values to be taken into account as well as a power electronic circuit structure corresponding to the respective task or problem, and the associated processing modules are selected, b) in one or more method steps, the detected input variables are automatically processed and analyzed based on the predetermined circuit structure by means of the selected processing modules, and operating parameters which are matched to the respective circuit and the predetermined input variables are determined, c) depending on the operating parameters optimized for the respective circuit and the predetermined input variables and taking into account the default values from one or more correspondingly preconfigured databases, suitable components and components for the optimized construction or an optimized realization or completion of the detected power electronic circuit are selected, d) the operating parameters and specification data of the selected components and components, which are optimized and matched to the respective circuit and the predetermined input variables, are compared with the default values and e) if they match, within predetermined tolerances, the determined operating data, as well as a parts list of the selected components including the associated data sheets or specification data are output and, if necessary, stored on a data memory, whereby an optimized conditioning of an already existing or under construction power electronic circuit or the circuit structure is achieved or achieved. 2. The method according to claim 1, characterized in that under Consideration of the specification data of the selected component (s) and Components and the operating parameters already determined further parameters, in particular power losses, heat development as well as electricity and / or Voltage tolerances can be determined. 3. The method according to claim 1 or 2, characterized in that at mismatch between the optimized operating parameters, the Specification data of the selected components as well as components and the default values, within the predetermined tolerances, a corresponding note and a Request for modification of the default values and / or the circuit structure as well as the procedure is repeated. 4. The method according to any one of claims 1 to 3, characterized in that experimentally recorded and / or artificially generated as input variables Data are used. 5. The method according to any one of claims 1 to 4, characterized in that an incomplete data set of input variables by means of a corresponding Processing modules supplemented and completed by artificially generated data becomes. 6. The method according to any one of claims 1 to 5, characterized in that that the processing of the input variables and determination of the optimized Operating parameters are carried out proportionately both in the frequency domain and in the time-based reference space. 7. The method according to any one of claims 1 to 6, characterized in that that in the default values also economic aspects, especially installation and or maintenance costs, both of the individual components to be selected and components as well as the entire circuit are detected and taken into account. 8. The method according to any one of claims 1 to 7, characterized in that it is both on-line, i.e. directly within the framework of a regulation, and off-line and thus indirectly, for the simulation, analysis and conditioning of power electronics Circuits is used. 9. The method according to any one of claims 1 to 8, characterized in that in the on-line use time-dependent on a spectral analysis and processing of Harmonic components are dispensed with in the frequency domain. 10. The method according to any one of claims 1 to 9, characterized in that when using on-line using appropriate processing modules and Interfaces for data transmission the optimized determined Operating parameters are available as setpoint specifications for a controller for further processing be put and transmitted. 11. The method according to any one of claims 1 to 10, characterized in that for analysis, simulation and conditioning of systems based on voltage intermediate circuit converters ( 20 ), in particular active filter systems, with AC or voltage circuit ( 40 ), DC voltage intermediate circuit ( 41 ), a controller ( 42 ) for switching the converter ( 20 ), an input resistor and a choke is used. 12. The method according to any one of claims 1 to 11, characterized in that in the analysis based on voltage intermediate circuit converters ( 20 ) based active filter systems preferably the waveform or the profile of the supply voltage u _s and the waveform or the profile of the load current i _l or alternatively the effective values of the fundamental and harmonic components of the load current are used as input data. 13. The method according to any one of claims 1 to 12, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems, the frequency f of the supply voltage, the highest harmonic frequency to be compensated or the switching frequency f _{sw of} the active Filters ( 10 ), the maximum permissible current ripple, the maximum permissible ripple of the DC voltage u _{dc of} the dc circuit ( 41 ), the maximum desired modulation index and the maximum permissible ambient temperature are used as default values. 14. The method according to claim 1 to 13, characterized in that in the simulation, analysis and conditioning based on voltage intermediate circuit converter ( 20 ) based active filter systems, the determination of the reference currents i * _{f of} the active filter ( 10 ) based on the load current detected as an input variable i _l takes place, the active component of the fundamental current oscillation of the load current being determined, subtracted from the load current and the resulting currents being inverted, which results in the reference currents i * _f for driving the active filter ( 10 ). 15. The method according to any one of claims 1 to 14, characterized in that in the simulation, analysis and conditioning based on voltage intermediate circuit converter ( 20 ) based active filter systems for determining the reference currents i * _f for driving the active filter ( 10 ) a) the current fundamental of the load current i _l by means of a Fast Fourier Transformation (FFT) according to


is determined, where k corresponds to the order of the harmonic, n the order of the sampling, I _l the harmonic component of the load current, i _l (n) the load current of the nth sampling and N the total number of samples taken, b) the real component i _l1re of the current _{fundamental oscillation} of the load current is determined and an inverse Fourier transformation according to in order to obtain the active components of the fundamental oscillation of the load current


is carried out, k corresponding to the order of the harmonic, n the order of the sampling, I _l the load current, i _l (n) the load current of the nth sample and N the number of samples taken, c) the determined active components of the fundamental oscillation of the load current i _l1re (n) from the load current i _l (n) according to
i * _f (n) = - (i _l (n) - i _l1re (n)) 1 ≤ n ≤ N,
subtracted, the resulting values are inverted and thus the reference currents i * _f for driving the active filter ( 10 ) are formed. 16. The method according to any one of claims 1 to 15, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems, the switching frequency f _sw for switching the active filter ( 10 ) or the converter ( 20th ) is chosen twice as high as the frequency of the maximum harmonic to be compensated. 17. The method according to any one of claims 1 to 16, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converters ( 20 ) based active filter systems in the case of digital control of the supply voltage, a phase feed of preferably 1.5 sampling periods is applied, a determination of the supply voltage of the nth scan based on the voltage values u _s (n + 1) of the next and u _s (n + 2) of the overnight scan according to


is carried out and n denotes the order of the samples or sample points and u _s the respective sampled supply voltage. 18. The method according to any one of claims 1 to 17, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems, the determination of the inductance L _{f of} the input choke of the active filter ( 10 ), based on the specifying the allowable peak current ripple î _well as


is carried out, where î _{well denotes} the default value for the peak value of the permissible current ripple, f _sw the switching frequency, L _f the inductance of the input choke and U _s the effective value (rms value) of the supply voltage per phase. 19. The method according to any one of claims 1 to 18, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems, the maximum permissible resistance value R _{f of} the choke based on the switching frequency f _sw and the inductance of the input choke of the active filter ( 10 ) is determined according to R _f << 2f _sw L _f : 20. The method according to any one of claims 1 to 19, characterized in that in the simulation, analysis and conditioning on voltage intermediate _circuit converters ( 20 ) based active filter systems for determining the required DC voltage value u _dc the reference voltage u * _{f of} the active filter ( 10 ) according to


is determined, with u * _f the reference voltage of the active filter ( 10 ), u _s the supply voltage, L _f the inductance of the input choke, di _f / dt the time derivative of the current of the active filter ( 10 ), i _f the current of the active Filters ( 10 ), n the order of the respective sampling and f _sw denote the switching frequency and, taking into account space vector modulation, the voltage _value u _{dc of} the dc circuit ( 41 ) of the active filter ( 10 ) according to


is determined, u _{fmax denoting} the maximum peak _{voltage value of} the converter ( 20 ) to be switched and u _dc the required voltage _{value of} the dc circuit ( 41 ). 21. The method according to any one of claims 1 to 20, characterized in that in an off-line use in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems, the current ripple by analyzing the current and voltage harmonic spectrum of the active Filters ( 10 ) is determined in the frequency domain, the behavior of a real power converter being simulated as realistically as possible using a space vector modulation method. 22. The method according to claim 21, characterized in that for obtaining and analyzing the voltage harmonics and deriving the time profile a) the voltages u * _{f of} the active filter ( 10 ) prepared for a space vector modulation in accordance with


taking into account the DC voltage value u _dc to achieve a scaling of the reference values between -1 and +1 are formed and standardized, b) the switching times t _n and switching angle or angular positions α _{k of} the space vector when switching according to


and


are determined using the switching frequency f _sw , the fundamental frequency f and the previously determined reference voltage u * _f (n) of the scanning with the order n, c) the determination of the time profile of the voltage of the active filter ( 10 ) the time-dependent voltage value u _f (t) of the active filter over the Fourier series


with the harmonic order k, the space vector position ωt and the coefficients


takes place, the Fourier coefficients in accordance with a three-phase converter


be modified and the voltage harmonics according to
U _f (k) = A _k - jB _k
are formed, where j is the √ -1 equivalent. 23. The method according to claim 22, characterized in that for analysis and for obtaining the current harmonics of the active filter ( 10 ) or its converter ( 20 ) the supply voltage U _s (k) Fourier-transformed and taking into account the determined voltage harmonics U _f (k) the current harmonics I _f (k) of the active filter ( 10 ) according to


are determined, where j is the √ -1 , f the frequency of the supply voltage, L _f the inductance of the choke of the active filter ( 10 ) and k the order of the respective harmonic. 24. The method according to any one of claims 21 to 23, characterized in that the current and / or voltage harmonics determined in the frequency or phase space are transformed into the time-based reference space by means of an inverse Fourier transformation and the current mean and effective values are in accordance


be calculated. 25. The method according to claim 21 to 24, characterized in that, taking into account the determined effective values, the copper windings of the input choke of the active filter ( 10 ), the busbar webs and the cables to be used are conditioned and a selection of suitable components from one or more correspondingly preconfigured is automated Databases ( 64 b, 70 b, 72 b, 73 b, 74 b and 75 b) is performed. 26. The method according to any one of claims 21 to 25, characterized in that a calculation of the direct current value i _dc in the time-based reference space using the converter voltages u _f and currents i _f according to the individual phases


is carried out. 27. The method according to any one of claims 1 to 20, characterized in that the on-line, ie direct use as part of a regulation in the simulation, analysis and conditioning based on DC link converters ( 20 ) based active filter systems, time-based on a consideration and Analysis of individual current and / or voltage harmonics in the frequency domain is dispensed with. 28. The method according to any one of claims 1 to 20 or 27, characterized in that in an on-line use approximately an effective current _{ripple value} I _lrip the choke or coil of the active filter ( 10 ) according


is determined, U _s denoting the RMS value (rms value) of the supply voltage per phase, u _dc voltage value of the DC circuit ( 41 ), L _f the inductance of the input choke and f _sw the switching frequency. 29. The method according to any one of claims 1 to 20 and 27 or 28, characterized in that in an on-line application, the calculation of the current value i _{dc of} the DC or DC voltage intermediate circuit ( 41 ) in the time-based reference space using the voltage value u _dc DC voltage intermediate circuit ( 41 ) and the voltages u _f1 , u _f2 , u _f3 and currents i _f1 , i _f2 , i _{f3 of} the active filter ( 10 ) in the three-phase static or stationary, non-rotating 123 reference system according to


is carried out. 30. The method according to any one of claims 1 to 20 and 27 to 29, characterized in that in the case of on-line use, the calculation of the current _{ripple value} I _{Cdcrip of} the DC _link ( 41 ), using the current _ripple value I _{lrip of} the choke, the effective value (rms Value) U _{s of} the supply voltage per phase and the voltage value u _{dc of} the DC voltage circuit ( 41 ) according to


is carried out. 31. The method according to any one of claims 1 to 30, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converter ( 20 ) based active filter systems for conditioning the capacitor or capacitors ( 22 ) or for determining the capacitance C _dc of DC voltage intermediate circuit ( 41 ) according to


by integrating the current value i _dc over time t and resulting in the peak-to-peak ripple _value u _dcpp to a predetermined percentage of the nominal DC voltage _value u _dc . 32. The method according to any one of claims 1 to 31, characterized in that in the simulation, analysis and conditioning based on voltage intermediate circuit converters ( 20 ) based active filter systems, taking into account the switching angle positions in the time-based reference space, a reconstruction of the switching signals for controlling the active filter ( 10 ) takes place in such a way that they have a value of one or zero and have the same number of sampling points as the reconstructed converter voltage. 33. The method according to any one of claims 1 to 32, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converters ( 20 ) based active filter systems, the currents flowing through the individual semiconductor valves ( 21 ) of the converter ( 20 ) when off- line use can be determined by means of the reconstructed switching signals and, in the case of on-line use, based on the sign of the reference voltages. 34. The method according to any one of claims 1 to 33, characterized in that in the simulation, analysis and conditioning on DC _link converters ( 20 ) based active filter systems with knowledge of both the voltage value u _{dc of} the DC link ( 41 ) and the peak value of the compensation current a required power electronic component, the same component is automatically selected from at least one preconfigured database, taking into account the aforementioned operating parameters and corresponding default values. 35. The method according to any one of claims 1 to 34, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converters ( 20 ) based active filter systems, the switching and forward losses to be expected with regard to the IGBTs used in accordance


V _{CC is} the collector-emitter supply voltage, I _{C is} the continuous collector current, V _{CCref is} the reference value of the collector-emitter supply voltage, I _{Cref is} the reference value of the continuous collector current, V _{CE (O) is} the collector-emitter threshold voltage (static), I _Crms the effective value (rms value) of the collector current, E _on the energy loss during the switch-on period or phase, E _off the energy loss during the switch-off period or phase, P _swIGBT the switching losses in the IGBT and P _condIGBT the _forward losses in the IGBT. 36. The method according to any one of claims 11 to 35, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converters ( 20 ) based active filter systems, the power losses of the individual diodes according


can be determined, where V _{TO is} the voltage _threshold , V _{CC is} the collector-emitter supply voltage, I _{F is} the mean current of the component, V _{CCref is} the reference value of the collector-emitter supply voltage, I _{Cref is} the reference value of the continuous collector current, E _{on is} the energy loss during the switch-on period or phase, E _off the energy loss during the switch-off period or phase, P _swDIODE the switching losses in the diode and P _condDiode the _forward losses in the diode. 37. The method according to any one of claims 1 to 36, characterized in that in the simulation, analysis and conditioning on voltage intermediate circuit converters ( 20 ) based active filter systems, the junction temperatures of the semiconductor components used are calculated on-line and based thereon a limitation of the peak current value is carried out becomes. 38. The method according to any one of claims 1 to 37, characterized in that depending on the operating temperatures to be observed, geometric specifications or dimensions and, if applicable, financial specifications automates one Selection of suitable coolants or bodies from one or more preconfigured Databases. 39. The method according to any one of claims 1 to 38, characterized in that in the simulation, analysis and conditioning based on voltage intermediate circuit converters ( 20 ) based active filter systems for avoiding and damping when vibrations are suddenly switched on, a correspondingly conditioned precharge resistor is determined. 40.Device for the simulation, analysis and conditioning of power electronic circuits, in particular active filter systems based on voltage intermediate circuit converters ( 20 ), which has a data processing device with data memory, input device and display device and interfaces for data acquisition and data output and for communication with at least one database, several of which , based on a predetermined circuit structure and the respective problem or task selected, interconnected processing modules are available and set up for this, both on-line, ie directly, in particular as part of a regulation, and off-line, i.e. indirectly, a) further process and analyze recorded input variables and to determine operating parameters which are optimized for the respective circuit and predetermined input variables, b) depending on the operating parameters resulting from the processing, optimized for the respective circuit and the predetermined input variables, and taking into account default values, suitable components and components for the purpose of an optimized construction or an optimized realization or completion of the predetermined power electronic circuit from one or more select according to preconfigured databases, c) to determine the resulting operating parameters or parameters, in particular power losses, taking into account the specification data of the respectively selected components and components, d) to compare the determined operating parameters and specification data of the selected components and components, which are optimized for the respective circuit and the predetermined input variables, and e) if there is agreement, within predetermined tolerances, to output the determined operating data as well as a parts list of the selected components including the associated data sheets or specification data and, if necessary, to save them on a data memory, thereby optimizing the conditioning of an already existing or under construction power electronic circuit or the circuit structure is made possible. 41. Apparatus according to claim 40, characterized in that at least a processing module analog control and / or transmission elements, in particular Has integral, proportional-integral controllers, multipliers and integrators. 42. Device according to one of claims 40 or 41, characterized characterized in that at least one processing module digital processors and components having.