Model predictive control for electric machine having multiple sets of windings

Abstract

A method for controlling an electrical converter (14) comprises: determining an average flux reference (I) from a reference torque (Tref) and an estimated torque (II) determined from a measured current (isabc, I, isabc, II) and a measured voltage (vsabc, I, vsabcll) and / or a reconstruction voltage (III) in a winding system (20a, 20b); determining at least one differential flux reference (IV) from the at least one differential current reference (V); determining a winding flux reference (VI) for each winding system (20a, 20b) from the average flux reference (I) and the at least one differential flux reference (IV) by adding and / or subtracting the at least one differential flux reference (IV) to the average flux reference (I); a switching command (uabc, I, uabc, II) for each converter branch (14a, 14b) is determined from a winding flux reference (VII) for a winding system (20a, 20b) and an estimated winding flux (VIII), the winding system (20a, 20b) being connected to the respective converter branch (14a, 14b), the estimated winding flux (VIII) being determined from a measured current (isabc, I, isabc, II) and a measured voltage (vsabc, I, vsabcll) and / or a reconfiguration voltage (III) in the winding system (20a, 20b), and wherein the winding flux reference (VII) and the estimated winding flux (VIII) are provided to a dedicated model predictive controller (52) for the respective converter branches (14a, 14b), the dedicated model predictive controller (52) determining switching commands (uabc, I, uabc, II); and applying a switching command (uabc, I, uabc, II) to each converter branch (14a, 14b).

Description

Technical Field

[0001] The present invention relates to the field of high power converter control. In particular, the present invention relates to a method, a computer program, a computer readable medium and a controller for controlling an electric converter. The present invention also relates to an electric converter. Background Art

[0002] Torque control of motors with multiple sets of three-phase windings is usually achieved via vector current control in a rotating reference frame consistent with the rotor flux. This linear control method focuses on the control of the fundamental current component and is applicable to converters modulated via carrier-based pulse width modulation (CB-PWM). By sampling with CB-PWM at the appropriate time point, only the current fundamental component is sampled without the use of a low-pass filter.

[0003] However, in medium-voltage drives, an optimized pulse pattern (OPP) is often employed to minimize current ripples at the (very low) permissible switching frequency. With an OPP modulated converter, the fundamental current component cannot be distinguished from the ripples via proper sampling. Instead, a low-pass filter must be applied to the measured current. Low-pass filtering limits the controller gain and has an adverse effect on the drive dynamics. Another implication of a linear controller is that the switching angle of the OPP must vary smoothly and without any discontinuities within the applied modulation index range. This is an additional restriction on the OPP calculation and, in most cases, results in a suboptimal OPP with higher current ripples.

[0004] The above mentioned problems are solved by using Model Predictive Pulse Mode Control (MP3C) for torque and flux control in an OPP modulated converter. MP3C controls the transient flux along its reference trajectory which is determined by the applied OPP of the converter connected to the machine terminals. The reference trajectory includes fluctuations and therefore no filtering is required to extract the fundamental component. This results in fast torque control. Furthermore, the switching angles of the applied OPP are allowed to have some discontinuities in the modulation index.

[0005] Nevertheless, the currently adopted MP3C has been designed for a motor having only one set of three-phase windings and cannot be directly applied to a motor having two or more sets of three-phase windings.

[0006] EP 2 469 69 2 A1 and US 8766570 B2 describe a flux-based MP3C method that uses OPP control to track a pre-calculated stator flux trajectory. EP 15177 656 A1 describes how to incorporate the neutral point potential into MP3C.

[0007] WO 2019 / 137916 A1 describes a model predictive control based on an online determined pulse width modulation switching pattern. Summary of the invention

[0008] The object of the invention is to improve the control of an electric converter having at least two converter branches connected to at least two winding systems of an electric machine. A further object of the invention is to reduce the total harmonic distortion and the losses of such an electric converter.

[0009] These objects are achieved by the subject matter of the independent claims. Further exemplary embodiments are evident from the dependent claims and the following description.

[0010] One aspect of the invention relates to a method for controlling an electrical converter. The electrical converter may be a low voltage or medium voltage converter suitable for handling voltages up to 5 kV or more. The electrical converter comprises at least two converter branches, which at one end are connected in parallel to at least two winding systems of an electrical machine. At the other end, each converter branch may be connected to a winding of at least one transformer. The electrical machine may be a generator, in particular a generator of a wind turbine.

[0011] According to one embodiment of the invention, the method comprises: determining an average flux reference from a reference torque and an estimated torque, for example in a rotating reference frame, the reference torque and the estimated torque being determined from measured currents and measured and / or reconstructed voltages in a winding system. For torque control, an average flux reference is considered. An optimal reference amplitude and angle for the average flux reference can be determined for a desired reference torque. The average reference flux can be an average of all reference fluxes of the winding system. The average flux reference can be provided as an amplitude and an angle. The reference torque can be provided by an external control loop that controls the torque of the motor. The estimated torque is determined from three-phase currents measured in the winding system and an estimated stator flux. Alternatively, the estimated torque can be determined from three-phase currents, an estimated rotor flux and an estimated machine inductance. The motor can have a number N of winding systems. The winding system can be a three-phase winding.

[0012] According to an embodiment of the invention, the method comprises determining at least one differential flux reference from at least one differential current reference, for example in a rotating reference frame. The at least one differential current reference may be provided by an external control loop.

[0013] According to one embodiment of the invention, the method comprises: determining at least one differential flux reference from at least one differential current. In general, there are N-1 differential flux references for controlling the differential (circulating) current between the winding system groups. Each of the differential flux references can be provided as an amplitude and an angle.

[0014] According to one embodiment of the invention, the method comprises: determining a winding flux reference for each winding system from an average flux reference and at least one differential flux reference by adding and / or subtracting an average flux reference from at least one differential flux reference. The winding flux reference for each winding system is determined based on an average flux reference and N-1 differential flux references. The average flux and the N-1 differential fluxes are calculated by multiplying the vector of the winding fluxes with the following matrix:

[0015]

[0016] The winding flux reference is calculated by multiplying the vector of the average and differential fluxes by the inverse of D. For example, for a machine with 3 winding systems, we have the matrix

[0017]

[0018] According to one embodiment of the invention, the method comprises: determining a switching command for each converter branch from a winding flux reference and an estimated winding flux of a winding system connected to the respective converter branch, wherein the estimated winding flux is determined from a measured current and a measured and / or reconstructed voltage in the winding system, and wherein the winding flux reference and the estimated winding flux are provided to a dedicated model predictive controller of the respective converter branch, which dedicated model predictive controller determines the switching command. Each winding system and / or converter branch uses a model predictive controller, i.e. a controller operating with a model predictive control method.

[0019] The method relies on a forecast model, a cost function, an optimization phase and a rolling policy. More specifically, a sequence of future control inputs is calculated over the forecast horizon such that the cost function is minimized by solving an optimization problem. In the sequence of control inputs, only the first element of the current time step is applied to the controlled system. At the next time step, new measurements and / or estimates are obtained and a revised sequence of future control inputs is calculated over the forecast horizon shifted by one time step. This results in a rolling back of the horizon policy.

[0020] MP3C is a special model predictive control method based on OPP. In MP3C, the winding flux vector is controlled along the winding flux reference trajectory using fast closed-loop control by modifying the switching moments of the OPP switching transitions within the prediction horizon. The switching transitions can be modified in time so that the predicted stator flux error, i.e. the difference between the reference stator flux and the estimated stator flux, is minimized at the end of the prediction horizon.

[0021] According to one embodiment of the invention, the method comprises: applying a switching command to each converter branch. Each converter branch comprises a power semiconductor controlled according to the switching command. The electrical converter may be a voltage source converter having at least one DC link capacitor. By manipulating the switch position of the converter, the voltage applied to the machine winding terminals is manipulated, which in turn allows controlling the current, flux and torque components of the machine.

[0022] The method generalizes the flux-based MP3C of the torque and flux of the motor to motors with more than one three-phase winding. To do this, the average and differential flux references are calculated and then converted to winding flux references for each winding system.

[0023] In the model predictive control of each winding system, the coupling with other three-phase winding systems can be considered or ignored to calculate the flux reference trajectory. In particular, for small resistance winding systems, such as medium voltage motors, the coupling can be ignored. The reference flux trajectory of each winding system can be calculated by a simple integration of the converter output voltage.

[0024] According to an embodiment of the invention, the switching command is determined from the OPP, which is loaded from a lookup table. The OPP may be pre-calculated (i.e., before the converter is operated) and stored in the lookup table with respect to the number of pulses and the modulation index. During converter operation, the number of pulses and the modulation index may be provided by an external control loop, and the corresponding OPP may be loaded from the lookup table. The OPP includes the switching moments, which encode when and how the converter switches to another switching state.

[0025] According to one embodiment of the invention, the switching command is determined online during the operation of the electrical converter. For example, the switching command, in particular the switching command sequence, can be calculated by pulse width modulation (e.g. carrier-based pulse width modulation or space vector modulation). For example, WO 2019 / 137916A1 describes the online calculation of the switching command based on pulse width modulation, which is incorporated herein by reference.

[0026] According to an embodiment of the invention, the switching instants of the switching commands are shifted forward and / or backward in time to minimize a flux error, which is the difference between an optimal winding flux trajectory and an estimated winding flux, the optimal winding flux trajectory being determined based on a winding flux reference. Using model predictive control, in particular MP3C, the switching instants can be modified to minimize the flux error. The switching instants can be modified by modifying their switching angles and / or switching times.

[0027] According to an embodiment of the invention, the switching commands for each converter branch are determined independently from the switching commands of at least one other converter branch. The controller performing the model predictive control method for each converter branch may operate independently, i.e. independently of the winding flux references of the other converter branches. The calculation of the optimal winding flux trajectory for each three-phase winding system may only consider the OPP of the specific three-phase winding system.

[0028] According to one embodiment of the invention, the average flux reference and at least one differential flux reference are determined in a rotating reference frame. The rotating reference frame may rotate at a fundamental frequency so that the reference becomes a DC quantity during steady state operation. The rotating reference frame may be a dq frame with the d axis aligned with the rotor flux.

[0029] According to an embodiment of the present invention, a winding flux reference is determined in a rotating reference frame and converted to a stationary reference frame. This can simplify the determination of winding flux errors and corresponding corrections. For processing by a model predictive control method, it may be necessary to provide a winding flux reference in a stationary reference frame.

[0030] According to one embodiment of the invention, the angle of the winding flux reference is calculated in the mirror reference frame.

[0031] According to one embodiment of the invention, at least one differential flux reference is determined from at least one differential current reference by multiplying with an estimated stator winding inductance of each winding system. The estimated stator winding inductance of a winding system may be the self-inductance of the winding system subtracted from the mutual inductance with other winding systems.

[0032] According to an embodiment of the invention, the method further comprises: determining at least one differential current from the measured currents in at least two windings, and determining a differential current error by subtracting the differential current from a differential current reference. The differential current error may be low pass filtered and / or PI controlled, and then added to the differential current reference, from which a differential flux reference may then be calculated. In this way, differential current errors that may occur due to incorrect differential flux calculations caused by mismatched parameter values ​​may be corrected.

[0033] According to an embodiment of the invention, the amplitude of the average flux reference may be low pass filtered. This may dampen oscillations and may reduce fluctuations in the winding flux reference value.

[0034] According to one embodiment of the invention, the magnitude of the average flux reference is determined from a lookup table based on the reference torque. The corresponding lookup table may be pre-calculated before the converter is operated.

[0035] According to one embodiment of the invention, the amplitude of the average flux reference is limited within predefined limits. The amplitude of the average flux reference may be limited so that specific limits are observed before synthesizing the winding flux reference. This ensures that the maximum possible converter output voltage is not exceeded.

[0036] According to one embodiment of the invention, the angle of the average flux reference is determined from a lookup table according to the magnitude of the average flux reference and the reference torque. The corresponding lookup table may be pre-calculated before the converter is operated.

[0037] According to an embodiment of the invention, the torque error as the difference between the reference torque and the estimated torque is low-pass filtered and / or PI-controlled before the angle of the average flux reference is determined from the look-up table and then added to the reference torque. In this way, torque errors that may occur due to mismatched parameter values, i.e. differences between actual and considered values ​​of machine model parameters, can be corrected.

[0038] According to an embodiment of the invention, when the amplitude of a winding flux reference of a winding system exceeds a limit, the amplitude of the average flux reference and / or the amplitude of the differential flux reference is limited so that the amplitude of the recalculated winding flux reference remains below the limit. The limit of the winding flux reference may be a fixed value. The angle of the average flux reference and the differential flux reference may remain constant. A coefficient k between 0 and 1, such as determined in the following formula (14), may be applied to the amplitude of the average flux reference and / or the differential flux reference.

[0039] According to one embodiment of the present invention, each converter branch is configured to convert a three-phase electrical quantity having a fundamental voltage, current and frequency into a three-phase electrical quantity having another fundamental voltage, current and frequency. For example, each converter branch may include a rectifier, a DC link and an inverter. The inverter and the optional rectifier may be controlled based on a switch command.

[0040] According to an embodiment of the invention, in contrast to the motor, each converter branch is connected to a secondary winding of a transformer. The secondary winding and the primary winding may have a common core. The transformer may also have a primary winding wound on the common core. The primary winding may be used to connect the converter system to a power grid. A common transformer with multiple sets of secondary windings or multiple series transformers may be used. The series transformers may each have a set of secondary three-phase windings.

[0041] Other aspects of the invention relate to a computer program for controlling an electric converter, which, when executed by a processor, is adapted to perform the steps of the method described herein, and to a computer readable medium storing the computer program. The winding flux reference determination and the model predictive control may be executed as software on one or more processors.

[0042] The computer readable medium may be a hard disk, a USB (Universal Serial Bus) storage device, a RAM (Random Access Memory), a ROM (Read Only Memory), an EPROM (Erasable Programmable Read Only Memory) or a FLASH memory. The computer readable medium may also be a data communication network, such as the Internet, that allows downloading program code. In general, the computer readable medium may be a non-transitory or a temporary medium.

[0043] Another aspect of the invention relates to a controller for controlling an electrical converter. The controller is adapted to perform the method described herein. The controller may include one or more processors, wherein the method is executed as software. The method may also be implemented at least partially or entirely in hardware, such as a digital signal processor or a field programmable gate array.

[0044] Another aspect of the invention relates to an electrical converter controlled as described herein.

[0045] According to an embodiment of the invention, an electrical converter comprises at least two converter branches connected in parallel to at least two winding systems of an electrical machine, and a controller as described herein.

[0046] In general, it has to be understood that features of the methods described above and below may be features of the computer programs, computer-readable media, controllers and electrical converters described above and below, and vice versa.

[0047] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The subject matter of the invention will be explained in more detail hereinafter with reference to exemplary embodiments shown in the drawings.

[0049] Figure 1 A converter system according to an embodiment of the invention is schematically shown.

[0050] Figure 2 A motor with two sets of three-phase windings is schematically shown.

[0051] Figure 3 A controller according to an embodiment of the present invention is schematically shown, and a method according to an embodiment of the present invention is shown.

[0052] Figure 4 The schematic diagram in more detail shows Figure 3 part of the controller.

[0053] Figure 5 The schematic diagram in more detail shows Figure 3 Another part of the controller.

[0054] The reference symbols used in the drawings and their meanings are listed in summary form in the reference symbol list. In principle, identical components are provided with the same reference symbols in the figures. DETAILED DESCRIPTION

[0055] Figure 1 A converter system 10 is schematically shown, which may be a wind energy conversion system, such as a gearless wind energy conversion system. The converter system 10 comprises a rotating electrical machine 12, at least two converter branches 14a, 14b and a transformer 16. The at least two converter branches 14a, 14b constitute an electrical converter 14. The converter system 10 with two converter branches 14a, 14b may be regarded as a double conversion line system.

[0056] The rotating electrical machine 12 may be a generator, such as a synchronous generator or a motor. In the following, it is assumed that the rotating electrical machine 12 is a synchronous generator without loss of generality. The rotating electrical machine 12 may be connected to a wind turbine 18 .

[0057] The rotating electrical machine 12 may have two or more independent winding systems 20 (see Figure 2 Each of the winding systems 20a, 20b is connected to one of the converter branches 14a, 14b. The first winding system 20a is connected to the first converter branch 14a and generates a current having a component i Sa，I ,i Sb，I ,i Sc，I The three-phase stator current i Sabc，I , the three-phase stator current i Sabc，I The first converter branch 14a converts the stator current i Sabc，I Convert to have component i a， I,i b，I ,i c，I The output current i Sabc，I . Output current i Sabc，I Power is supplied to the secondary winding 22a of the transformer 16. The secondary winding 22a may be star-connected.

[0058] Similarly, the second winding system 20b is connected to the second converter branch 14b and generates a voltage having a component i Sa，I I,i Sb，I I,iS c，II The three-phase stator current i Sabc，II , the three-phase stator current i Sabc，II The second converter branch 14b converts the stator current i Sabc，II Convert to have component i a，I I,i b，I I,i c，II The output current iabc，II . Output current iab c，II The power is supplied to the secondary winding 22b of the transformer 16. The secondary winding 22a may be connected in delta.

[0059] Each converter branch 14a, 14b may consist of a rectifier 24, a DC link 26 and an inverter 28. Both the rectifier 24 and the inverter 28 may be two-level or, as described herein, three-level converters, or they may have more voltage levels. The rectifier 24 and the inverter 28 may be back-to-back converters and / or 3LNPC (three-level neutral point clamped) converters. The machine-side rectifier 24 converts the AC voltage of the corresponding winding system 20a, 20b into a DC voltage v dc，I or v dc，II , the DC voltage v dc，I or v dc，II The power is supplied to the DC link 26 and the grid-side inverter 28. The inverter 28 converts the DC voltage v dc，I or v dc，II Converted into a DC output voltage with corresponding three-phase output current, the components of which are i a，I ,i b，I ,i c，I or a，I I,i b， I I,i c，II .

[0060] The transformer 16 also includes a primary winding 30 connected to a grid 32. The transformer 16 may be a sawtooth / delta sawtooth transformer.

[0061] The electrical converter 14 and at least two converter branches 14a, 14b are controlled by a controller 34. The controller 34 may include a processor and a computer-readable medium. The computer-readable medium may store a computer program for executing the control method described herein. In addition, the controller 34 may include a circuit, such as an FPGA, an ASIC and / or a DSP, which is configured to execute the control method. The controller 34 may be configured to modulate and / or switch the rectifier 24 and the inverter 28. In particular, the rectifier 24 and / or the inverter 28 may be switched with an OPP. The method described below may be applied to the rectifier 24, which is an inverter if we have an electric motor instead of a generator.

[0062] Figure 2 The electric machine 12 is shown in more detail and includes a stator 36 and a rotor 38. The stator includes at least two stator winding systems 20a, 20b, which are geometrically displaced relative to each other by a fixed angle β. Each stator winding system 20a, 20b has a three-phase winding, and the stator current of the three-phase winding is i Sabc，Iand i Sabc，II The rotor 38 is in the moving position θ R ω R Rotates at a speed of Figure 2 In, the rotating dq reference frame is also described.

[0063] The electric machine 12 can be a synchronous generator with permanent magnets, but can also be a synchronous generator with a field winding 40 or even an asynchronous generator. The three-phase winding system 20a, 20b, also called the armature winding, can be in the stator 36 (as shown) or in the rotor 38. In the latter case, the field winding 40 or the permanent magnets are placed in the stator 36. Since the method described herein can also be used for electric motors, the term "electric machine" is generally used. In particular, hereinafter, the term "stator current" can refer to the winding current, which can also be provided by the winding system 20a, 20b as part of the rotor 38.

[0064] Figure 3 A block diagram is shown illustrating the controller 34 and a method for controlling the electrical converter 14 . Figure 3 The control scheme of a motor 12 with two winding systems 20a, 20b is shown. The main purpose is to control the torque of the motor. In wind turbine applications, the reference torque T ref Provided by the wind turbine operator's external control loop. The purpose is to use torque to control and maximize the power produced by the wind turbine at a specific speed and wind speed. In other applications where speed control is required, the reference torque T ref Provided by speed controller.

[0065] The secondary control task is to control the differential current i between the two winding systems 20a, 20b Sdq，diff , which is important for certain operating modes (e.g. thermal operation test of generators). The differential current i Sdq，diff By subtracting the stator current i in the dq reference frame in block 46 Sdq，I and i Sdq，II is calculated. The measured stator current I Sabc，I and i Sabc，II Transformed to the stationary αβ reference frame by block 42, i.e. Sαβ，I and i Sαβ，II , and is converted to the dq reference frame by block 44.

[0066] Differential Current Reference It can be provided in a synchronously rotating dq reference frame that is aligned with the rotor flux and consists of a d-axis component and a q-axis component.

[0067] Reference torque T ref and differential current reference is sent to block 48 where a flux reference calculation is performed for each of the two stator winding systems 20a, 20b. Figure 4 The winding flux reference calculation block 48 is described in more detail.

[0068] Calculated winding flux reference (amplitude is and Angle and ) is then forwarded to a model predictive control block 50, which contains two MP3C controllers and / or loops 52, one for each winding system 20a, 20b. In addition to the winding flux reference determined by the winding flux reference calculation block 48, the estimated stator flux (amplitude and and the angle is and ) or more generally the winding flux is sent to the model predictive control block 50 and the MP3C internal controller 52, which modify the switching commands u of each converter branch 14a, 14b abc，I and u abc，II To track the winding flux reference. In addition, the neutral point potential v of the DC link 26 is measured or estimated. NP，I and v NP，II It can be provided to the model predictive control block 50 and the internal MP3C controller 52.

[0069] The estimator block 54 calculates the stator current i measured in the stationary αβ reference frame from Sαβ，I and i Sαβ，II and the measured stator voltage v Sαβ，I and v Sαβ，II Or the reconstructed stator voltage and The winding flux (amplitude is and and the angle is and ), rotor flux (angle is and torque ). Stator three-phase voltage v Sabc，I and v Sabc，II Usually not measured, but based on the switch position and the measured DC link voltage v dc，I and v dc，II Subsequently, they are transformed into the stationary αβ reference frame.

[0070] If an encoder is available, the rotor position and speed are also provided to the flux observer. If no encoder is available, the rotor position θ R and speed ω R is estimated by the estimator block 54 .

[0071] Figure 4 The winding flux reference calculation block 48 is shown in more detail. To reduce the number of symbols, only minus signs "-" are used in nodes. Where no subtraction sign appears, a plus sign "+" is implied. To calculate the winding flux reference (magnitude of and and the angle is and Their magnitudes and angles must be determined in the stationary αβ reference frame.

[0072] Block 56 is based on the reference torque T ref and estimated torque The average flux reference is calculated in the calculation, in particular, the magnitude of the average flux reference relative to the rotor flux angle and angle In order to control the torque of the motor 12 In the rotating dq reference frame aligned with the rotor flux, the average of the fundamental components of the two fluxes must have a specific amplitude and angle For some types of machines, such as asynchronous machines, the required amplitude and angle can be calculated analytically using a simple formula. However, for some other types of machines, such as synchronous permanent magnet machines, the required amplitude and angle are solutions to nonlinear optimization problems. In this case, the optimal values ​​of the amplitude and angle can be pre-calculated and stored in the lookup tables 58, 60.

[0073] First, the reference torque T ref Provided to the look-up table 58, defines the optimum average winding flux amplitude Typically, for a given torque, the winding flux amplitude that minimizes the stator current is is considered optimal. This approach is referred to as maximum torque per ampere MTPA). Alternatively, the lookup table 58 may define the stator flux or more generally the winding flux that minimizes the rotor losses, the total generator losses, or the total losses of the drive (generator 12 and converter 14).

[0074] The subsequent limiter 62 ensures compliance with the average flux reference amplitude The output of the limiter 62 is optionally low-pass filtered in an LP filter 64 and then forwarded to another lookup table 60 which defines the required average flux angle relative to the rotor flux angle. The required average flux angle can be interpreted as the required load angle. Lookup table 60 calculates the given average flux amplitude and reference torque T refThe required load angle. Note that the reference torque T at the input of the lookup table 60 ref can be modified by the PI controller 66, which converts the reference torque T ref and estimated torque The PI controller 66 corrects the torque error that may occur due to parameter value mismatch (difference between actual value and considered value of machine model parameter). This difference can also be low pass filtered by LP filter 68. Estimated torque Calculated from the product of the estimated mean flux and the mean current:

[0075]

[0076] The asterisks indicate the complex conjugate values ​​of the estimated mean fluxes. The arrows indicate the space vectors.

[0077] The average flux and current in the stationary reference frame are calculated as follows:

[0078]

[0079] in

[0080]

[0081] The winding flux that appears on the right side of equation (2) and The components of are calculated by two separate winding flux estimators, one for each winding system 20a, 20b. These winding flux estimators are part of block 54 ( Figure 3 ). Estimation of the winding flux is also necessary for the MP3C controller 52, Figure 5 Directly measure the winding current i appearing on the right side of equation (3) Sabc，I and i Sabc，II .

[0082] The differential current is controlled by controlling a Angle This is achieved by using a differential flux reference of , which is determined in block 70. From the differential current, the differential flux reference in the synchronous rotating dq reference frame is calculated in block 72 using the estimated stator winding inductance.

[0083]

[0084] The differential current i in the rotating dq reference frame Sdq，diff It is calculated from the measured current of each winding in the rotating system, and the formula is as follows:

[0085] i Sd，difff =i Sd，I-i Sd，II (9)

[0086] i Sq，diff =i Sq，I -i Sq，II (10)

[0087] Note that the product in (6) is not directly applied to the differential current reference component and The additional terms from the outputs of the two PI controllers 74 are added to them to obtain the filtered components and They are input to block 72. The input to these PI controllers 74 is the differential current error on the d-axis and q-axis, low-pass filtered by a low-voltage filter 76. The task of blocks 74, 76 is to correct the differential current error that may occur due to incorrect calculation of the differential flux due to mismatched parameter values.

[0088] In the winding flux calculation block 78, the average flux reference (amplitude is and The load angle is and ) and the differential flux reference (amplitude is The load angle is ) calculates the winding flux reference (amplitude is The load angle is ):

[0089]

[0090] The flux amplitude and angle of each winding system can be calculated from the d-axis and q-axis components:

[0091]

[0092] Calculated Amplitude and can exceed the specified limits. In this case, either the average flux or the differential flux must be reduced. Either the average flux (torque) or the differential flux (differential current) can take precedence. Normally, it is the former, so the differential flux amplitude needs to be reduced in such a way that the amplitude of the winding flux does not exceed the specified limits. The required reduction factor k is determined by the solution of a quadratic equation:

[0093]

[0094] In the above considerations, the differential flux reference The angle of remains unchanged, only its amplitude changes. In equation (14), the differential flux reference The amplitude of is multiplied by a factor k, where 0<k<1, to limit the winding flux reference amplitude and

[0095] If desired, the d-axis or q-axis components of the differential flux may also be prioritized. The component with the lowest priority may again be determined by the solution of the quadratic equation.

[0096] Finally, the winding flux reference angle in the stationary αβ reference frame is calculated and To do this, the rotor flux angle is estimated (or measured if an encoder is available) and the rotor flux angle is subtracted from the electrical geometric displacement angle of the No. 2 winding. Add to the previously calculated load angles for the No. 1 and No. 2 winding systems respectively.

[0097] like Figure 3 As shown, for each winding system 20a, 20b, the winding flux reference (amplitude is and Angle and ) and estimated winding flux (magnitude is and Angle and ) are sent to the MP3C controller 52, which can operate as described in EP 2 469 69 2A1 and US 8766570 B2.

[0098] Such MP3C controller 52 is as follows Figure 5 As shown. According to the flux reference amplitude and Estimated or measured stator angular frequency and the measured DC link voltage v dc，I and v dc，II , determine the appropriate modulation index m for the respective winding systems 20a, 20b and, in combination with the pulse number d, load the two converters with the corresponding OPPs, respectively, according to the following formula:

[0099]

[0100] Low pass filtered DC link voltage v dc，I and v dc，II can be used in the above equation. Note that for synchronous motor 12, the estimated stator frequency can be replaced by the measured or estimated rotor speed For the selection of a suitable OPP, the pulse number d is also required, which is chosen so that the switching frequency of the converter does not exceed its maximum value.

[0101] According to the modulation index m and the pulse number d, the mode loader 80 provides a switching angle vector A=[α1 α2 ...α d ] T and the corresponding single-phase switch position vector U = [u1 u2 ... u d ] T . In general, assuming a pulse pattern with quarter-wave and half-wave symmetry, the pulse number d is the number of switching angles within one quarter of the fundamental period of the three-level switching waveform. Using quarter-wave and half-wave and three-phase symmetry, a three-phase pulse pattern corresponding to the above vectors A and U can be easily constructed.

[0102] The reference flux trajectory can then be calculated in block 82 using the state-space model of the motor 12. For the general case of a convex synchronous motor, its state-space model is given by

[0103]

[0104] The system matrix is ​​defined as

[0105]

[0106] in

[0107]

[0108] Has weight

[0109] L Sav =(L Sd +L Sq ) / 2 (20) L Sdiff =(L Sd -L Sq ) / 2 (21)

[0110] M Sav =(M Sd +M Sq ) / 2 (22) M Sdiff =(M Sd -M Sq ) / 2 (23)

[0111] The output matrix of formula (16) is defined as

[0112]

[0113] For insignificant machines, the following statements hold:

[0114] L Sd =L Sq =L Sav (26)L Sdiff =0 (27)

[0115] M Sd =M Sq =M Sav (28)M Sdiff =0 (29)

[0116] Furthermore, it is evident from (19) that for a salient pole machine, the state space model parameters, especially the inductance, depend on the rotor angle. This means that the reference trajectory of a set of OPPs also depends on the rotor angle and we have an LTV (Linear Time Varying) state space model.

[0117] For example, in a medium voltage motor, the stator resistance R S may have very low values. Therefore, the first and last terms of equation (16) can be omitted, and the state space model can be approximated as

[0118]

[0119] As can be seen from the above formula, the flux of each winding system 20a, 20b can be simply calculated by integrating the corresponding winding voltage, regardless of whether the machine is significant or not. The winding voltage corresponding to the selected OPP is calculated by multiplying the OPP switch position by the half DC link voltage. Usually, only the switch position is used, because the calculated flux reference is independent of the drive parameters, the most obvious of which is the DC link voltage. The integral of the switch position (in the αβ reference frame) is divided by the modulation index and multiplied by the flux reference amplitude. Taking into account the flux reference angle at the same time, this restaurant reference can be calculated from the optimal trajectory.

[0120] In block 84, a switching table is constructed to store the upcoming switching transitions at each stage. More specifically, the vector of the single-phase switch positions

[0121]

[0122] Use n x The elements of the phase x∈{a, b, c} are used to construct the phase. The nominal switching angle is divided by the estimated stator angular frequency is converted to the nominal switching instant relative to the current instant. This results in the corresponding nominal switching instant vector

[0123]

[0124] U x and The switching table for phase x is constructed. In addition to the OPP switching transition, other switching transitions may be inserted into the switching table to achieve a faster torque response.

[0125] The next step is to calculate the transient flux error, i.e. the winding flux reference Rather than estimating The difference between

[0126]

[0127] In addition, the common-mode flux error component ψ can be increased Serr，0 To control the neutral point potential, as described in EP 15 177 656 A1. To this end, the measured neutral point potential v of the corresponding DC link 26 is dc，I and v dc，II Input neutral point controller 86, determine the zero component flux of the winding flux reference trajectory

[0128] The mode controller 88 manipulates the switching moment of the OPP in order to reduce the winding flux error within a certain time interval (prediction range). This allows the winding flux vector to be controlled along the reference trajectory, achieving closed-loop control of the winding flux error. The result is a three-phase switching command u abc =[u a u b u c ] T , or, more generally, a sequence thereof. The acquired switching commands of the first sampling interval are applied to the electrical converter 14. At the next sampling instant, the switching sequence of the switching commands is recalculated according to the horizon retreat strategy.

[0129] Although the present invention has been illustrated and described in detail in the drawings and the above description, such illustration and description are to be considered illustrative or exemplary rather than restrictive; the present invention is not limited to the disclosed embodiments. Other variations of the disclosed embodiments may be understood and implemented by a person skilled in the art through a study of the drawings, the present disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or controller or other unit may perform the functions of several items listed in a claim. The fact that certain measures are listed in mutually different dependent claims does not indicate that a combination of these measures cannot be used for advantageous purposes. Any reference signs in the claims should not be interpreted as limiting their scope.

[0130] Glossary

[0131] A OPP primary single-phase switching angle for a given modulation index

[0132] d I , d II Number of OPP pulses in converter branches no.I and no.II

[0133] m I , m II Modulation index in converter branches no.I and no.II

[0134] The space vector of the average current in the stationary αβ coordinate system, whose real part is the alpha axis component and imaginary part is the beta axis component

[0135] i Sx，I ,i Sx，II Stator currents of winding systems no.I and no.II in phase x∈{a,b,c}

[0136] i Sαβ，I ,i Sαβ，II Stator currents of winding systems no.I and no.II in the stationary αβ coordinate system

[0137] L Sd , L Sq Stator winding self-inductance in H on the d-axis and q-axis

[0138] M Sd , M Sq Stator winding mutual inductance in H on the d-axis and q-axis

[0139] n x Number of entries in the switching table for phase x∈{a,b,c}

[0140] p Pole pairs

[0141] R S The stator resistance in Ω

[0142] T is the electromagnetic torque in Ws

[0143] The vector of nominal switching instants for phase x∈{a, b, c]

[0144] U OPP primary single-phase switch position for a given modulation index

[0145] U x The vector of single-phase switch positions for phase x∈{a, b, c}

[0146] u abc，I ,u abc，II The three-phase switch positions of converters no.I and no.II, u∈{-1, 0, 1}

[0147] v Sαβ，I , v Sαβ，IIStator voltage of winding no.I and no.II in the stationary αβ coordinate system

[0148] v NP，I , v NP，II Neutral point potential of converters no.I and no.II

[0149] v dc，I , v dc，II Total DC link voltage of converter no.I and no.II

[0150] β is the electrical angle in rad representing the geometric displacement of winding no. II

[0151] Reference load angle for average flux (angle between stator and rotor flux)

[0152] Reference load angle for differential flux

[0153] θ R The electrical angle of the rotor in rad

[0154] The space vector of the average flux in the stationary reference frame, whose real part is the alpha axis component and imaginary part is the beta axis component

[0155] ψ Rαβ Rotor flux in αβ coordinate system

[0156] ψ S，I , ψ S，II Winding flux amplitude for winding no.I and no.II

[0157] Winding flux angle in the stationary αβ coordinate system for windings no.I and no.II

[0158] ψ Sαβ，I , ψ Sαβ，II Winding flux in αβ coordinate system for windings no.I and no.II

[0159] Maximum permissible amplitude of winding flux

[0160] ω R The electrical angular speed of the rotor in rad / s

[0161] ω S The electrical angular frequency of the stator in rad / s

[0162] ^ Symbol for estimator

[0163] * represents the symbol for the conjugate complex value

[0164] Reference numerals list

[0165] 10 Converter System

[0166] 12 Rotating motor

[0167] 14 Electric converter

[0168] 14a, 14b Converter branches

[0169] 16 Transformer

[0170] 18. Wind Turbines

[0171] 20a, 20b Winding system

[0172] 22a, 22b Secondary winding

[0173] 24 Rectifier

[0174] 26 DC Link

[0175] 28 Inverter

[0176] 30 Primary winding

[0177] 32 Power Grid

[0178] 34 Controller

[0179] 36 Stator

[0180] 38 Rotor

[0181] 40 Field winding

[0182] 42 abc to αβ conversion block

[0183] 44 αβ to abc conversion block

[0184] 46 Differential current calculation block

[0185] 48 Winding flux reference calculation block

[0186] 50 Model Predictive Control Block

[0187] 52 MP3C Controller

[0188] 54 Estimator Block

[0189] 56 Average flux calculation block

[0190] 58 Lookup Table

[0191] 60 Lookup Table

[0192] 62 Average Limiter

[0193] 64 Low pass filter

[0194] 66 PI controller

[0195] 68 Low pass filter

[0196] 70 Differential flux calculation block

[0197] 72 Internal differential flux calculation block

[0198] 74 PI controller

[0199] 76 Low pass filter

[0200] 78 Winding flux calculation block

[0201] 80 Mode Loader

[0202] 82 Flux reference calculation block

[0203] 84 Switch table creation block

[0204] 86 Neutral point controller

[0205] 88 Mode Controller

Claims

1. A method for controlling an electrical converter (14), the electrical converter (14) comprising at least two converter branches (14a, 14b), the at least two converter branches (14a, 14b) being connected in parallel to at least two winding systems (20a, 20b) of an electrical machine (12); The method comprises: From the reference torque (T ref ) and estimated torque Determine the average flux reference The estimated torque From the measured current (i Sabc，I ,i Sabc，II ) and the measured voltage (v Sabc，I , v sabc，II ) and / or reconstruction voltage be determined; From at least one differential current reference Determine at least one differential flux reference By referring to the average flux Adding and / or subtracting the at least one differential flux reference Coming from the average flux reference and the at least one differential flux reference Determining a winding flux reference for each winding system (20a, 20b) From the winding flux reference for the winding system (20a, 20b) and estimated winding flux Determine the switching command (u) for each converter branch (14a, 14b) abc，I ,u abc，II ), the winding system (20a, 20b) is connected to the corresponding converter branch (14a, 14b), wherein the estimated winding flux From the measured current (i Sabc，I ,i Sabc，II ) and the measured voltage (v Sabc，I , v Sabc，II ) and / or the reconstruction voltage is determined, and where the winding flux reference and the estimated winding flux is provided to a dedicated model predictive controller (52) for the respective converter branch (14a, 14b), the dedicated model predictive controller (52) determining the switching command (u abc，I ,u abc，II ); The switch command (u abc，I ,u abc，II ) is applied to each converter branch (14a, 14b).

2. The method according to claim 1, The switch command (u abc，I ,u abc，II ) is determined from an optimized pulse pattern (A, U), which is loaded from a look-up table (80); and / or The switch command (u abc，I ,u abc，II ) is determined online during operation of the electrical converter (14); and / or The switch command (u abc，I ,u abc，II ) of the switching transient is moved so that the flux error (ψ Serr,αβ0 ) is minimized, the flux error (ψ Serr,αβ0 ) is the optimal winding flux trajectory and the estimated winding flux The difference between the optimum winding flux trajectory and the winding flux reference is determined.

3. The method according to claim 1 or 2, wherein the switching command (u abc，I ,u abc，II ) said switching command from at least one other converter branch is independently determined.

4. The method according to any one of the preceding claims, The average flux reference and the at least one differential flux reference is determined in a rotating reference frame; and / or Where the winding flux reference is determined in the rotating reference frame and converted to the stationary reference frame in; and / or Where the winding flux reference The angles are calculated in the stationary reference frame.

5. The method according to any one of the preceding claims, wherein the at least one differential flux reference The at least one differential current reference is obtained by multiplying the estimated stator winding inductance for each winding system (20a, 20b) is determined.

6. The method according to any one of the preceding claims, From the measured current (i Sabc，I ,i Sabc，II ) determines at least one differential current (i Sdq,diff ), and by taking the differential current reference Subtract the differential current (i Sdq,diff ) to determine the differential current error; Low pass filtering (76) the differential current error and / or PI controlling (74) the differential current error; The low-pass filtered and / or PI-controlled differential current error is added to the differential current reference 7. The method according to any one of the preceding claims, The average flux reference The magnitude of the reference torque (T ref ) is determined from a look-up table (58); and / or The average flux reference The amplitude of is low pass filtered (64).

8. The method according to any one of the preceding claims, The average flux reference The magnitude of is limited to within predefined limits (62).

9. The method according to any one of the preceding claims, The average flux is referenced at an angle Based on the average flux reference The magnitude and the reference torque (T ref ) is determined from a look-up table (60), and / or The torque error is low-pass filtered (68) and / or PI controlled (66) and added to the reference torque (T ref ), the torque error is the reference torque (T ref ) and the estimated torque The difference between.

10. The method according to any one of the preceding claims, further comprising: When the winding flux reference of the winding system (20a, 20b) The magnitude exceeds the limit, limiting the average flux reference The magnitude and / or the differential flux reference The magnitude of the winding flux is recalculated to The magnitude of is kept below the limit.

11. The method according to any one of the preceding claims, wherein each converter branch (14a, 14b) is configured to convert a three-phase electrical quantity having a fundamental voltage, current and frequency into a three-phase electrical quantity having another fundamental voltage, current and frequency; wherein each converter branch (14a, 14b) comprises a rectifier (24), a DC link (26) and an inverter (28); and / or Opposite the motor (12), each converter branch (14a, 14b) is connected to a secondary winding (30) of a transformer (16).

12. A computer program for controlling an electric converter (14), said computer program being adapted to perform the steps of the method according to any one of the preceding claims when executed by a processor.

13. A computer readable medium in which the computer program according to claim 12 is stored.

14. A controller (34) for controlling an electrical converter (14), adapted to perform the method according to any one of claims 1 to 11.

15. An electrical converter (14), comprising: at least two converter branches (14a, 14b) connected in parallel to at least two winding systems (20a, 20b) of the electric machine (12); A controller (14) according to claim 14.

Images