Saturation strategy for grid forming in DFIG systems

Abstract

The present invention relates to a method to control a grid-forming controller (20) of a converter in a wind turbine (3), where the wind turbine (3) connected to an electrical grid (8), wherein the method includes a saturation strategy for the grid forming controller in which a saturation is applied to the rotor current set points I* r = (ir d*, ir q*). The present invention further relates to a controller adapted to operate the method, and to a wind turbine or wind farm operated by the method.

Description

[0001] SATURATION STRATEGY FOR GRID FORMING IN DFIG SYSTEMS

[0002] The present invention relates to a method for operating a power converter in a grid forming operating mode.

[0003] BACKGROUND

[0004] The electrical power provided by an electrical power generation system is usually fed into a power grid which is an interconnected network for electricity delivery and distribution. Electrical grids vary in size and can cover whole countries or continents. In order to maintain the stability of a power grid, the generated electrical power has to fulfill stringent grid codes. Due to the increasing variety of renewable energy sources, e. g. wind turbines or solar cells, converters are used to convert the generated electrical power appropriately and to maintain the stability. The converters are designed to perform a grid-forming operation where the wind turbines at events (e. g. voltage or frequency changes ) generate a stable frequency and maintain grid voltage.

[0005] Doubly fed induction generators (DFIG) are used in many wind turbines and includes electric generators, where both the field magnet windings and armature windings are separately connected to equipment outside the machine. Such DFIG have additional features which allow them to run at speeds slightly above or below their natural synchronous speed.

[0006] Due to the natural behavior of grid forming control, the saturation strategy used in the current control (grid feedingbased control) is not sufficient to ensure that power limits are not exceeded.2025P00097WO

[0007] SUMMARY OF THE INVENTION

[0008] The obj ect of the invention is achieved by the independent claims. The dependent claims describe advantageous developments and modifications of the invention.

[0009] This includes to introduce a method to control a grid-forming controller of a converter in a wind turbine, where the wind turbine connected to an electrical grid, wherein the method includes a saturation strategy for the grid forming controller in which a saturation is applied to the rotor current set points.

[0010] The grid forming controller may be implemented in the rotorside converter (RSC) controller of a DFIG system.

[0011] The outputs of the grid-forming controller may be the rotor current setpoints and may be applied to the current controllers as rotor saturation current set points generated as a modification to the rotor current limits I*rby the saturators.

[0012] The method may include control anti-windup loops consisting of PI controllers that compares the difference between the saturated and the non-saturated set points.

[0013] The method may include a first and second anti-windup loop, where the first of the loops compares the first coordinates values, and the second of the loops compares the second coordinate values.

[0014] Both the first and the second anti-windup loops may operate as a plus input of the respective first or second component values as outputted from a respective saturator, and be2025P00097WO

[0015] compared via a negative input of the similar respective first or second non-saturated rotor set points as outputted from the grid-forming controller.

[0016] The grid forming controller may adapted to react to a fault on the grid, such as a deviation in voltage or frequency.

[0017] The grid forming controller may be based on a Virtual Synchronous Machine (VSM), droop control, power synchronization control ( PSC), virtual synchronous machine control (VSM), direct power control (DPC), and / or virtual oscillator control (VOC).

[0018] In addition to the rotor current limits, the saturators may be inputted the respective saturation limits.

[0019] To drive the rotor current setpoints out of saturation, the first axis loop will act on the reactive power set point of the grid forming control, while the second axis loop acts on the active power set point.

[0020] The saturation limits may be being calculated dynamically.

[0021] In case of a detected or registered fault, then the rotor current saturation limits may be calculated using the DFIG electrical equations in steady state; uL= RsIs + j osi|xs, and l|js — LsIs T LMIr •

[0022] The positive maximum powers (upper limits ) may be used for calculating the negative rotor current limits, while negative maximum powers (lower limits ), may be used for calculating the positive rotor current limits.

[0023] The stator current limits may be calculated using the equations for the stator active power and reactive power to come to the equations for the stator current, Isd= (Qsusq+ Psusd) / ( (usd)2+ (usq)2) and Isq= ( Psusq- Qsusd) / ( (usd)2+ (usq)2).By using that Ls= L + Los, and by manipulation of the equations for DFIG in steady state and for stator current, coming to the following calculations of the first and second axis rotor current s respectively, Ird= ( U^ / OLM )>( LS / LM ) ( Psusd+ Qsusq) / ( ( usd)2+ ( usq)2), and Irq= - ( U^ / QLM ) + ( LS / LM ) ( Qsusd- Psusq) / ( ( usd)2+ ( usq)2), which equations may be used to dynamically to calculate the rotor current limits.

[0024] The measured active power and reactive power limit may be applied to the first axi s rotor current limit, and the measured reactive power and the active power limit respectively may be applied to the second axis rotor current limit calculation in the two equations, to arrive to the equations used to calculate the rotor current limits,

[0025] I J-rdlT im ( uSpO Sq / COLM ) ( Ls / LM ) ( Pmeas USpoSd_l“QiimuSpoSq) / ( ( uSpO Sd)2+ ( uSpO Sq)2), and

[0026] I rqlim ( USPOEP / QLM ) 9" ( LS / LM ) ( Qme as USpO sd

[0027]

[0028] PlimUSposq) / ( ( usposd)2+ ( usposq)2) •

[0029] The present invention further relate s to the controller adapted to operate the method according to any of the embodiment s.

[0030] The present invention further relate s to a wind turbine or wind farm operated by the method according to the any of the embodiments.

[0031] BRIEF DESCRI PTION OF THE DRAWINGS

[0032] Embodiments of the invention are now described, by way of example only, with reference to the accompanying drawings, of which:2025P00097WO

[0033] 5

[0034] Figure 1 shows a schematic and typical electrical supply and distribution system supplied by power produced by a plural of wind turbines in a wind farm.

[0035] Figure 2 illustrates an embodiment solution according to the present invention, where a saturation is applied to the current set points.

[0036] Figures 3 illustrates another embodiment calculation of rotor current limits used in the embodiment solution seen in figure 2.

[0037] Figure 4 illustrates a DFIG equivalent electrical circuit.

[0038] Figures 5 illustrates another embodiment calculation of rotor current limits used in the embodiment solution seen in figure 2.

[0039] Figure 6 illustrates is a more detailed illustration of an embodiment calculationsof the rotor current limits IrDQmax in the FLL frame.

[0040] The illustration in the drawings is in schematic form. It is noted that in different figures, similar or identical elements may be provided with the same reference signs.

[0041] DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a schematic illustration of an electrical supply and distribution system 1. The power is supplied a renewable power source, which could be in the form of a collection of wind turbines 3, also referred to as a wind farm 2, or wind park. Alternatively, or additionally, the renewable power sources could include or be photovoltage devices etc.

[0043] The individual wind turbines 3 each are formed with a tower 4 and a nacelle 5 which comprises a hub and a rotor with blades2025P00097WG

[0044] 6 to catch the wind. A generator is connected to the rotor to convert a rotation into electrical power.

[0045] The generated power is feed to distributions systems 7 including substations, transformers etc. From the distribution systems 7 the power is feed to the electrical distribution grid 8, or utility grid, to be provided as power for end users 9.

[0046] Due to the natural behavior of grid forming control, it has been observed the saturation strategy used in present wind turbine control from time to time to be insufficient to ensure that the of the wind turbines 3 power limits, respective the active power limit Pmaxand reactive power limit Qmax, are not exceeded. This is in particular the situation for grid feeding-based control, where the control comprises a reaction to a fault on the grid (e. g. utility grid 8 ), such as a deviation in voltage or frequency.

[0047] The solution according to the present invention is to introduce a new saturation strategy for the grid forming controller, in which the saturation is applied to the rotor current set points I*rsat= (irsatd*, irsatq* ) in case the grid forming controller is implemented in the rotor-side converter (RSC) controller in the DFIG.

[0048] Figure 2 shows a block diagram of an embodiment of the solution according to the present invention.

[0049] The grid forming (GFM) controller 20 may be based on any known techniques, such as the well-known Virtual Synchronous Machine (VSM), droop control, power synchronization control ( PSC), virtual synchronous machine control (VSM), direct power control (DPC), and virtual oscillator control (VOC), etc. In the following it is exemplified via a Virtual Synchronous Machine.

[0050] The outputs of the GFM controller 20 are the set points of the rotor current setpoints I*r= (ird*, irq* ), here expressed2025P00097WG

[0051] in dq-coordinates. The calculated rotor current setpoints I*rare applied to the saturators 24.

[0052] The saturators 24 output rotor saturation current set points I*rsat= (irsatd*, irsatq* ) being feed as input to the current controllers 26 controlling the rotor current.

[0053] The control anti-windup loops 31d, 31q consist of two PI controllers 22 that uses the difference between the saturated setpoints I*rsat and the non-saturated setpoints I*r. The first of the loops compares 31d the first of the coordinates ird*, irsatd*, such as d-coordinates, and the second of the loops 31q compares 31q the second of the coordinates irq*, irsatq*, such as q-coordinates. Both anti-windup loops 31d, 31q in the illustrated embodiment operate as a plus input of the respective first or second component of the saturated setpoint I*rsat as outputted from the respective saturator 24 and is compared via a negative input of the similar respective first or negative the non-saturated setpoints I*ras outputted from the GFM controller 20. This then is feedback to the respective power reference Q*, P* inputs to the GFM controller 20. In the illustrated embodiment the first component anti-windup loop 31d is feed k to the reactive power reference Q* and the second component anti-windup loop 31q is feed to the active power reference P*.

[0054] Thus, in the dq-system, to drive both current setpoints (I*r, I*rsat) out of saturation, the d-axis loop will act on the reactive power set point of the grid forming control Q*, while the q-axis one will act on the active power set point P*.

[0055] In addition to the (non-saturated) rotor current limits I*r= (ird*, irq* ), the saturators 24 are inputted the respective saturation limits Idiim= ( irdP°sllm, irdneqiim) and Iqiim= (irqp°sllm, irqnegllmj. The fir st setof saturation limits Idnm= (irdposiim, irdneqnm) give a respective maximum positive value irdP°snmog negative irdneqnmvalues in the first coordinate, e. g. the d-2025P00097WG

[0056] coordinate. The second set of saturation limits Iqum= (irqp°sllm, irqnellm) give the respective maximum positive value irqnegllm ancj negative irqneqllmvalues in the second coordinate, e. g. the q-coordinate.

[0057] The saturation limits Idum, Iqiim according to the present invention are not static, but are being calculated dynamically 30.

[0058] Input to the calculation 30 of the saturation limits Idum, Iqnminclude a range of different input such as variables and fixed parameters, where the variables can be determined by measurements, by estimation e. g. via digital twins, etc. In the illustrated embodiment of figure 2 these include the active and reactive power limits Plim, Qiim- The two inputs usdand usqare the respective first d and second q axis stator voltages (measured or registered or estimated by other means like digital twins etc. ). The illustration illustrates the setup for the calculation is for the positive-sequence voltage usdand usq, similar calculations may be done for the negative sequency voltage, though this is not illustrated as the setup is similar.

[0059] The input parameters Piec, Qiec are predefined safety parameters which may be set based on grid code from the grid ( 8 ) and / or wind farm (2 ) or other grid ( 8 ) safety and standard requirements etc., like for active and reactive power calculations according to IEC-61400-21 standards. They may be fixed values or calculated based on different input. Further inputted values are instantaneous power measurements, respective active measured power Pmeasand reactive measured power Qmeas • Though indicated as measured, they could also be estimated or registered by other means.

[0060] Finally, a fault is inputted, being e. g. '0' in case of no fault and '1' in case of a detected or registered fault.

[0061] The rotor current saturation limits Irdiim, Irqumare calculated 34 using the DFIG electrical equations in steady state:2025P00097WO

[0062] UL= RsIs + j Wsl|Js, and l|xsLsIs+ LM Ir ( 1 )

[0063] Here Rsand is stator resistance and Isis stator current, each of these inputted to the calculations 34 respectively including a first and second coordinate inputs. Lmis the mutual inductance, IR the rotor current. Ls is the inductance and Isthe current of the stator. The input Qsis the rotation frequency of the field and the input l|xsis flux.

[0064] Figure 3 shows a block diagram of an embodiment calculation of the rotor current limits Ir^iim = Irdum, Irqumshowing positive Ir^iim and negative limits Irnegiim calculated independently. The equations used within the block are the same in both calculations. However, a couple of inputs change. The positive maximum powers (upper limits ), Pposiimand Qposiim, are used for calculating the negative rotor current limits Irnegiim, while negative maximum powers (lower limits ), Pnegumand Qnegnmare used for calculating the positive rotor current limits IrP^iim.

[0065] As it is well known for power sources connected to the grid, when the respectively positive and negative Power refers to if power is supplied to the grid or absorbed from the grid.

[0066] The rotor current limits Irdiim, Irqnmare calculated based on stator active and reactive power limits Pnm, Qnm-

[0067] First, stator current limits are calculated using the equations for the active power Psand reactive power Qs:

[0068] Ps= usdIsd+ usgIsg(2 )

[0069]

[0070] Qs= -UsdIsq+ UsgIsd(3 )

[0071] Where usis the stator voltage and isis the stator current seen for respective in the first axis d and second axis q.2025P00097WO

[0072] 10

[0073] From ( 2 ) and ( 3 ), the following stator current equations may be obtained:

[0074] Isd= ( Qsusq+ Psusd) / ( ( Usd)2+ ( Usq)2) ( 4 )

[0075]

[0076] Isq= ( Psusq- Qsusd) / ( ( Usd)2+ ( usq)2) ( 5 )

[0077] Fig. 4 shows an equivalent electrical circuit of the DFIG, which in the steady-state leads to DFIG electrical equations ( 1 ) •

[0078] Since Ls= LM + Los, and by manipulation of the equations ( 1 ) ( 4 ) and ( 5 ), it is pos s ible to come to the following calculations of the f irst Irdand second Irqaxis rotor current s respectively,

[0079] Ird= ( U^ / QLM ) - ( LS / LM) ( PsUsd+ Qsusq) / ( ( usd)2+ (up)2), ( 6 )

[0080] Irq= - ( usd / oLM) + ( LS / LM) ( Qsusd— Psusq) / ( ( usd)2+ (up)2)t( y)

[0081] As it can be observed in ( 6 ) and ( 7 ), both first ( e. g. d) and second ( e. g. q) axi s of the rotor current depend on both active and reactive power. To avoid this coupling, the measured active power Pmeasand reactive power limit Qlimis applied to the first axis ( e. g. d-axis ) rotor current limit Irdnm. Further, the measured reactive power Qmeasand the active power limit Plimrespectively is applied to the second axis ( e. g. q-axis ) rotor current limit Irqlimcalculation. The se limitations give the two equations ( 8 ) and ( 9 ) to calculate the rotor current limits Irdnmand lAtm respectively:

[0082] ( Uspos^ / Op ) ( LS / LM ) ( PmeasUSposd+QlimUsposq) / ( ( USpO sd)2+ ( Usposq)2) ( 8 )

[0083] I rqlim ( USpO sd / (j)LM ) T ( LS / LM) ( Qme as USpO sd

[0084]

[0085] PlimUSposq) / ( ( USposd)2+ ( USposq)2) ( 9 )2025P00097WQ

[0086] 11

[0087] Note that the input usposdand usposqused in equations ( 8 ) and ( 9) are the first and second axis (dq) positive sequence of the stator voltage uspos. Also, it is important to notice that when a fault is detected (fault = 1 ), the instantaneous power measurements Pmeas, Qmeas are substituted by their iec predefined safety parameters Piec, Qiec, which may be calculated based on grid code from the grid ( 8 ) and / or wind farm (2 ) or other grid ( 8 ) safety and standard requirements etc., like for

[0088] active and reactive power calculations according to IEC-61400-21 standards.

[0089] Figures 5 and 6 shows some further embodiment details where fig. 5 shows an embodiment block diagram of the calculation 30 of the rotor current limits Irdqmax =

[0090]

[0091] Ird, Irqmax -

[0092] According to the illustrated embodiment, the parameters of the calculations as illustrated in fig. 3 of the Irdqmax, are referred to in FLL ( Frequency-Locked Loop) reference frame, where such parameters in the present are indicated by capital indexes, like 'DQ'.

[0093] First, stator current limits IsDQmax are calculated 32 based on a set power limits, respective a reactive power limit Qlimand active power limit Plim. The power limits Qlim, Plimin turn depend on the detection of a fault. If no fault is detected the switch 42 is set to ' O', and the power limits Qlim, Plimare as the maximum active Pmaxand reactive Qmaxpowers respectively according to the e. g. wind turbine 3 settings. In case of a detected fault the switch is set to '1' and the power Qlim, Plimlimits is the power references Q*, P*. Like for the rotor, the stator current limits IsDQmax includes respectively a first coordinate limit, IsDmax and a second coordinate limit IsDmax - The first coordinate stator current limits IsDmax further includes respectively a positive maximum IsDP°smaxand a negative maximum IrDne9maxvalue. In the same manner, the second coordinate stator current limits Is^ax further includes respectively a positive maximum IsQp°smaxand a negative maximum IrQne9maxvalue.Finally, rotor current limits are referred 36 to in the VSM reference frame (dq) using the angle difference between the angle of the VSM and the angle of the FLL (

[0094]

[0095] 3 = 0...,r.

[0096] Figure 6 is a more detailed illustration of an embodiment calculations 34 of the rotor current limits IrDQmax in the FLL frame. In the illustrated embodiment a difference is made based on the detection of an error. If no error is detected the switch 42 is on '1 ' and then in this connected mode ( steady state), the measured stator current ISDQis used in the electrical equations of the DFIG via a saturator 40 used to saturate the stator current ISDQto output the stator current limit IsDQlim. This stator current limit IsDQlimthen is feed as the stator current Isas input to the calculations 38 using the DFIG electrical equations in steady state according to the equations above.

[0097] In case of a detected fault the switch goes to ' O' and the maximum stator current limit IsDQposmax is used as stator current IsDQthe input to the calculations 38.

[0098] In either case the (measured) stator current usDQis inputted directly for the calculations 38.

[0099] Although the present invention has been described in detail with reference to the preferred embodiment, it is to be understood that the present invention is not limited by the disclosed examples, and that numerous additional modifications and variations could be made thereto by a person skilled in the art without departing from the scope of the invention.

[0100] It should be noted that the use of "a" or "an" throughout this application does not exclude a plurality, and "comprising" does not exclude other steps or elements. Also, elements described in association with different embodiments may becombined. It should also be noted that reference signs in the claims should not be construed as limiting the scope of the claims.

Claims

2025 P00097WG14Patent Claims1. Method to control a grid-forming controller ( 20 ) of a converter in a wind turbine ( 3 ), where the wind turbine ( 3 ) connected to an electrical grid ( 8 ), wherein the method includes a saturation strategy for the grid forming controller in which a saturation is applied to the rotor current set points I*r= ( ird*, i ) •2. Method according to claim 1, wherein the grid forming controller ( 20 ) is implemented in the rotor- side converter ( RSC ) controller of a DFIG system.

3. Method according to claim 1 or 2, wherein the outputs of the grid-forming controller ( 20 ) are the rotor current setpoints I*r= ( ird*, irq* ) and is applied to the current controllers ( 26 ) as rotor saturation current set points I*rsat = ( ir-satd*, irsatq* ) generated as a modification to the rotor current limits I*rby the saturators ( 24 ).

4. Method according to any of the previous claims, wherein the method includes control anti-windup loops ( 31d, 31q) consisting of PI controllers ( 22 ) that compares the difference between the saturated I*rsat and the non-saturated set points I*r.

5. Method according to claim 4, wherein the method includes a first ( 31d) and second anti-windup ( 31q) loop, where the first (31d) of the loops compares the first coordinates values ( ird*, irSatd‘ ) ( and the second of the loops ( 31q ) compare s the second coordinate values ( irq*, irsatq* ).

6. Method according to claim 4, wherein both the f irst ( 31d ) and the second ( 31q) anti-windup loop operate as a plus input of the re spective first or second component values ( ird*, ir-satd* ), ( irq‘, irsatq* ) a s outputted from a respective saturator ( 24 ), and is compared via a negative input of the similar2025P00097WG15respective first or second non-saturated rotor set points (ird*, irq* ) as outputted from the grid-forming controller (20 ).

7. Method according to claim 3 or 6, wherein, in addition to the rotor current limits I*r= ( ird*, irq* ), the saturators ( 24 ) are inputted the respective saturation limits Idlim= (irdposlim, irdneqlim) and Iqlim= (irqposlim, irqneqlim).

8. Method according to any of the previous claims, wherein, to drive setpoints (I*r, I*rsat) out of saturation, the first axis loop (31d) will act on the reactive power set point (Q* ) of the grid forming control, while the second axis loop (31q) acts on the active power set point ( P* ).

9. Method according to any of the previous claims, wherein the saturation limits (Idlim, Iqlim) are being calculated dynamically (30 ).

10. Method according to claim 99, wherein, in case of a detected or registered fault, then the rotor current saturation limits (Irdlim, Irqlim) are calculated 34 using the DFIG electrical equations in steady state;uL= RsIs+ j ωsψs, and ψs= LsIs+ LMIr(1).

11. Method according to claim 10, wherein he positive maximum powers (upper limits ), Pposlimand Qposlim, are used for calculating the negative rotor current limits Irnegiim, while negative maximum powers (lower limits ), Pneglimand Qneglimare used for calculating the positive rotor current limits Irposlim.

12. Method according to claim 10 or 11, where stator current limits are calculated using the equations for the active power Psand reactive power Qsto come to the equations for the stator current,2025 P00097WQ16Isd= (Qsusq+ Psusd) / ((usd)2+ (usq)2) (4)Isq= (Psusq- Qsusd) / ((usd)2+ (usq)2) (5),which band by using that Ls= LM+ Lσs, and by manipulation of the equations ( 1 ) ( 4 ) and ( 5 ), coming to the following calculations of the first Irdand second Irqaxis rotor currents respectively,Ird= (usq / ωLM) - (LS / LM) (Psusd+ Qsusq) / ((usd)2+ (usq)2), (6) Irq= -(usd / ωLM) + (LS / LM) (Qsusd- Psusq) / ((usd)2+ (usq)2), (7)where claim ( 6 ) and ( 7 ) is used to dynamically to calculate the rotor current limits Irdlimand Irqlim.

13. Method according to claim 12, wherein the measured active power Pmeas and reactive power limit Qnmis applied to the first axi s ( rotor current limit Irdlim, and the measured reactive power Qmeas and the active power limit Plimrespectively is applied to the second axis rotor current limit Irqlimcalculation in the two equations ( 8 ) and ( 9 ) to calculate the rotor current limits Irdlimand Irqlimrespectively,Irdlim= (usposq / ωLM) - (LS / LM) (Pmeasusposd+ Qlimusposq) / ((usposd)2+ (usposq)2) (8)Irqlim= -(usposd / ωLM) + (LS / LM) (Qmeasusposd- Plimusposq) / ((usposd)2+ (usposq)2) (9)14. A controller adapted to operate the method according to any of the claims 1-13.

15. A wind turbine ( 3 ) or wind farm ( 2 ) operated by the method according to the any of the claims 1-13.

Images