Post on 02-Feb-2021
1
Dr. Carlos E. UGALDE-LOOCardiff University, Wales, UK
Integration of Offshore Wind Farms using HVDC
Technologies: Results from the BEST PATHS Project
EES-UETP
HVDC and HVDC Grids for Future Transmission
May 28 – June 1, 2018
Leuven, Belgium
Additional credit to:
Daniel ADEUYI, Sheng WANG, (Cardiff University); Salvador CEBALLOS, (Tecnalia, Spain);
Salvatore D’ARCO and Gilbert BERGNA (SINTEF, Norway); Mireia BARENYS (GAMESA, Spain);
Max PARKER and Stephen FINNEY (University of Strathclyde, Scotland);
Andrea PITTO and Diego CIRIO (RSE SpA, Italy);
Jakob GLASDAM (Energinet, Denmark); and Íñigo AZPIRI (Iberdrola, Spain).
2
Outline of the Presentation
1. Introduction
2. The BEST PATHS Project
3. BEST PATHS Demo 1:
a) Network Topologies
b) Key Performance Indicators
c) The ‘Open Access’ Toolbox
4. Simulation Results
5. Real-Time Demonstrator
6. Experimental Results
7. On-Going (Simulation and Experimental) Work
8. Conclusions and Next Steps
3
Introduction
• Wind energy will be the most widely adopted renewable energy source (RES) by 2050 to contribute towards the abatement of green house gas emissions.
• Europe’s installed wind capacity is 168.7 GW* (18% of EU’s total installed power generation capacity). In the UK:
• Operational in 2018:
o Onshore: 12.1 GW (1523 projects, 7057 turbines);
o Offshore: 7.11 GW (33 projects, 1832 turbines).
o Total: 19.214 GW**
* https://windeurope.org/wp-content/uploads/files/about-wind/statistics/WindEurope-Annual-Statistics-2017.pdf** https://www.renewableuk.com/page/UKWEDhome
4
Introduction (2)
• HVAC technology is mature and suitable for subsea transmission at typical voltages up to 150 kV and distances up to 100 km.
• HVDC has better control capabilities, lower power losses and occupies less space compared with HVAC.
• A ‘Business as Usual’ approach to improve infrastructure will not be sufficient to meet policy objectives at reasonable cost.
• Operators and manufacturers are now considering HVDC solutions over HVAC for offshore power transmission systems:
o A higher quality and more reliable wind resource with higher average wind speeds is farther away from shore, and thus,
o Long distances to shore.
o Above 150 kV and beyond 100 km HVAC is not practical due to capacitance and hence charging current of submarine cable.
5
Introduction (3)
• Voltage source converter (VSC) based schemes are becoming the preferred option over line commutated converter (LCC) alternatives due to their decoupled power flow control, black-start capability and control flexibility.
• MTDC grids will facilitate a cross-border energy exchange between different countries and will enable reliable power transfer from offshore wind farms (OWFs).
• The interactions between wind turbine (WT) converters and different VSC types in a meshed topology need further investigation.
6
Introduction (4)
SubstationWind farm
> 100 km
7
Introduction (5)
SubstationWind farm
8
Introduction (6)
SubstationWind farm
HVDC
HVAC
9
The BEST PATHS Project
Key Figures
• Budget of €62.8M, 56% co-funded by the European Commission under the 7th Framework Programme for Research, Technological Development and Demonstration (EU FP7 Energy).
• Duration: 01/10/2014 – 30/09/2018 (4 years).
• Composition: 5 large-scale demonstrations, 2 replication projects, 1 dissemination project.
Key Aims
• Through the contribution of 40 leading research institutions, industry, utilities, and transmission systems operators (8), the project aims to develop novel network technologies to increase the pan-European transmission network capacity and electricity system flexibility.
BEyond State-of-the-art Technologies for re-Powering Ac corridors & multi-Terminal Hvdc Systems
10
The BEST PATHS Project (2)
BEyond State-of-the-art Technologies for re-Powering Ac corridors & multi-Terminal Hvdc Systems
11
BEST PATHS Demo #1
• Objectives:
1. To investigate the electrical interactions between the HVDC link converters and the wind turbine (WT) converters in OWFs.
2. To de-risk multivendor and multi-terminal HVDC (MTDC) schemes.
3. To demonstrate the results in a laboratory environment usingscaled models.
4. To use the validated models to simulate a real grid with OWFs connected in HVDC.
12
BEST PATHS Demo #1 (2)
HVDC equipment manufacturers provide ‘black boxes’
We intend to use ‘open models’R&D Centres
TSOs
Utilities & RES developers
Independent Manufacturers
?
Detailed models
Simulation & Validation
13
Network Topologies
➢System configurations have been implemented in Simulink
• A number of topologies has been modelled, simulated and analysed.
• The topologies considered constitute likely scenarios to be adopted for the transmission of offshore wind energy in future years.
• Full details available in Deliverable D3.1 of the BEST PATHS project.
GSC
Pw1
Pg1,Qg1Onshore
AC Grid #1
DC CABLE
Vdc and Q
Controller
AC Voltage
Control
Vdc_g1
Vdc_g1*fw1*|Vac_w1*|
Vac_w1
Offshore
Grid #1WFC Offshore Onshore
Qg1*
θw1*
➢Point-to-Point HVDC Link (Topology A)
14
Network Topologies (2)
➢Three-Terminal HVDC System
Pw1
DC
NETWORK
AC Voltage
Control
fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
Offshore Onshore
θw1*
GSC #1Pg1,Qg1
Onshore
AC Grid #1
(Vdc vs. P) and Q
Controller
Vdc_g1* Qg1*
Vdc_g1
AC Voltage
Control
fw12*Vac_w2*
Vac_w2
WFC #2
θw2*
Pw2
Offshore
Grid #1
15
Network Topologies (3)
GSC #1
Pw1
Pg1,Qg1Onshore
AC Grid #1
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g1
Vdc_g1*fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
Qg1*θw1*
GSC #2
Pw2
Pg2,Qg2Onshore
AC Grid #2
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g2
Vdc_g2*fw2*Vac_w2*
Vac_w2Offshore
Grid #2WFC #2
Qg2*θw2*
GSC #3
Pw3
Pg3,Qg3Onshore
AC Grid #3
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g3
Vdc_g3*fw3*Vac_w3*
Vac_w3
Offshore
Grid #3WFC #3
Qg3*θw3*
DC
NETWORK
Offshore Onshore
AC interlink
➢Six-Terminal HVDC System with Offshore AC Links (Topology B)
16
Network Topologies (4)
GSC #1
Pw1
Pg1,Qg1Onshore
AC Grid #1
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g1
Vdc_g1*fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
Qg1*θw1*
GSC #2
Pw2
Pg2,Qg2Onshore
AC Grid #2
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g2
Vdc_g2*fw2*Vac_w2*
Vac_w2
Offshore
Grid #2WFC #2
Qg2*θw2*
GSC #3
Pw3
Pg3,Qg3Onshore
AC Grid #3
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g3
Vdc_g3*fw3*Vac_w3*
Vac_w3
Offshore
Grid #3WFC #3
Qg3*θw3*
DC
NETWORK
Offshore Onshore
DC interlink
➢Six-Terminal HVDC System with Offshore DC Links (Topology C)
17
Network Topologies (5)
➢Twelve-Terminal HVDC System with
Offshore DC Links (Topology D)
GSC #4
Pg3,Qg3
(Vdc vs. P) and Q
Controller
Vdc_g4
Vdc_g4* Qg4*
GSC #5
Pg2,Qg2
(Vdc vs. P) and Q
Controller
Vdc_g2
Vdc_g2* Qg2*
GSC #6
Pg1,Qg1
Onshore
AC Grid #B
(Vdc vs. P) and Q
Controller
Vdc_g6
Vdc_g6* Qg6*
DC
NETWORK
Offshore Onshore
Pw1
AC Voltage
Control
fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
θw1*
Pw2
AC Voltage
Control
fw2*Vac_w2*
Vac_w2
Offshore
Grid #2WFC #2
θw2*
Pw3
AC Voltage
Control
fw3*Vac_w3*
Vac_w3
Offshore
Grid #3WFC #3
θw3*
Pw4
AC Voltage
Control
fw4* Vac_w4*
Vac_w4
Offshore
Grid #4
θw4*
Pw5
AC Voltage
Control
fw5* Vac_w5*
Vac_w5
Offshore
Grid #5WFC #5
θw5*
Pw6
AC Voltage
Control
fw6* Vac_w6*
Vac_w6
Offshore
Grid #6WFC #6
θw6*
GSC #1
Pg6,Qg6
(Vdc vs. P) and Q
Controller
Vdc_g1
Vdc_g1* Qg1*
GSC #2
Pg5,Qg5
(Vdc vs. P) and Q
Controller
Vdc_g5
Vdc_g5* Qg5*
GSC #3
Pg4,Qg4
Onshore
AC Grid #A
(Vdc vs. P) and Q
Controller
Vdc_g3
Vdc_g3* Qg3*
DC
NETWORK
Offshore Onshore
WFC #4
(100 km)
(5 k
m)
(10 km)
(10 km)
(10
km)
(10
km)
(100 km)
18
Key Performance Indicators
KPI.D1.1 – AC/DC interactions: power and harmonics
Steady state Power quality WT ramp rates
• To assess the suitability of the models and proposed HVDC network topologies, converter configurations and control algorithms, a set of KPIshave been defined.
• Full details available in Deliverable D2.1 of the BEST PATHS project.
KPI.D1.2 – AC/DC Interactions –Transients & Voltage Margins
Normal operation Extreme operation
KPI.D1.3 – DC Protection Performance / Protection & Faults
Protection selectivity Peak current and clearance time
KPI.D1.4 – DC Inter-array Design
Inter-array topology
Power unbalance
Fault tolerance
Motorising capability
KPI.D1.5 – Resonances
AC systems oscillation Internal DC resonance
KPI.D1.6 – Grid Code Compliance
Active and reactive power
Fault ride-through
19
The ‘Open Access’ Toolbox
• A set of models and control algorithms has been developed, simulated and assessed.
• Their portability as basic building blocks will enable researchers and designers to study and simulate any system configuration of choice.
20
The ‘Open Access’ Toolbox (2)
• The models and control algorithms have been published in the BEST PATHS website as a MATLAB ‘Open Access’ Toolbox: http://www.bestpaths-project.eu/.
• The toolbox was originally presented last year in the 13th IET Conference on AC and DC Power Transmission (ACDC2017):
o CE Ugalde-Loo, et al., “Open Access Simulation Toolbox for the Grid Connection of Offshore Wind Farms using Multi-Terminal HVDC Networks”, 13th IET ACDC17, Manchester, UK, 2017.
http://www.bestpaths-project.eu/
21
The ‘Open Access’ Toolbox (3)
• A user manual is also provided, together with the published models and accompanying examples.
• Full details of the models available in Deliverable D3.1 of the BEST PATHS project.
22
The ‘Open Access’ Toolbox (4)
➢ Converter Stations
➢ AC Grid
• Averaged and switched models for an MMC.
• The combined averaged-switched model consists of two blocks:
o Power electronics block,
o Low level controller block: circulating current reference generation, circulating current controller, nearest level control modulation strategy & sub-module voltage regulator.
• AC network adapted from the classical nine-bus power system.
➢ DC Cable
• The DC cable section has been modelled as a one-phase, frequency-dependent, travelling wave model.
• It is based on the universal line model (ULM), which takes into account the frequency dependence of parameters.
23
The ‘Open Access’ Toolbox (5)
➢ Wind Farm
• It accurately represents the behaviour of an aggregated OWF.
• To avoid large simulation times and undesirable computer burden, simplifications have been carried out in the electrical system:
o The converter of the a wind turbine generator (WTG) is modelled with averaged-model based voltage sources.
o A current source represents the remaining WTGs of the OWF. The current injection of the first WTG is properly scaled to complete the rated power of the whole OWF.
• The detailed WTG contains
o a permanent magnet synchronous generator model;
o Averaged models: machine- and grid-side converters, including filters and DC link;
o An LV/MV transformer and internal control algorithms.
24
The ‘Open Access’ Toolbox (6)
➢ High Level Controller
• It considers converter operation in three control modes.
• The aim is to cover the main control needs for different system configurations.
id1*, iq1*
High Level Controller
Vdc – Q
Mode
Selector
P – Q AC Island
Mode 1 Mode 2 Mode 0
id2*, iq2*
Outer
Loop
Inner Current
Loopabc-dq
Vdq
idq
Vdq*
dq-abc
Vabc
iabc
PLL θ
vabc*vabc*
θ
Half-Bridge
MMC
L
± VDC
Vabciabc
AC Voltage
Control
fwf*
|Vwf*|
θwf*
Low Level Controller
Current Limiter
DC
Grid
AC
System
idq*
P – Vdc
Droop
Select
Vdc – P
Droop
Select
Imax
dq-φ θ
θ*
θwf*
25
The ‘Open Access’ Toolbox (7)
➢ High Level Controller
o Mode 0: Vac voltage control: The converter sets the voltage and frequency.
Offshore
AC Grid
Vdc
ω = 2πf
Vwf,a =Msin(ωt +θ)
Vwf,b = Msin(ωt+θ+2π/3)
Vwf,c=Msin(ωt+θ-2π/3)
fwf* = const
θwf* = const
f
θ
WFC
M
vabc
Vabcma
Pwf
ski3kp3+
|Vwf*|
+-
Switching
Control
vabc*
|Vwf|
26
The ‘Open Access’ Toolbox (8)
➢ High Level Controller
Mode 1: Vdc – Q control scheme with a P – Vdc droop
id*Vdc
Vdc,0*s
ki,V+kp,V
P*kpv
+-
P
id1*Vdc*
Outer Loop
Outer
LoopInner Current
Loop
Switching
Control
Vabc
Liabc
vabc*
abc-dq
vdq
P, Qabc-dq
idq
dq-abc
vdq*
Vdc MTDC system
PLL
θ
θ
Vdc
idq*
vabc
Main AC
Grid
ΔVdc,0
iq*
-1
vd
++ +
-id
vd*kp +
iq
vdInner Current Loop
ωL
vq*
+-
++
+
ωL
kp ++-
++
IQ*
q-axis
Limited iq*
Limited id*
iq* before limiting
id* before limiting
d-axis
ski
ski
Radius
Imax
Imax
Q*
Current
Limiter
Current
Limiter
idq1*
27
The ‘Open Access’ Toolbox (9)
➢ High Level Controller
Mode 2: P – Q control scheme with a Vdc – P droop
id*
Vdc
Vdc*
P0*
kvp
+-
POuter Loop
Outer
LoopInner Current
Loop
Switching
Control
vabc
Liabc
vabc*
abc-dq
vdq
P, Qabc-dq
idq
dq-abc
vdq*
Vdc MTDC system
PLL
θ
θ
Vdc
idq*
vabc
Main AC
Grid
ΔP*
iq*
-1
vd
++
+-
id
vd*kp +
iq
vdInner Current Loop
ωL
vq*
+-
++
+
ωL
kp ++-
++
q-axis
Limited iq*
Limited id*
iq* before limiting
id* before limiting
d-axis
ski
ski
Radius
Imax
Imax
Q*
Current
Limiter
P*1
vd
Current
Limiteridq2*
id2*
id2*
28
The ‘Open Access’ Toolbox (10)
o Universities include Aalborg University, KU Leuven, Fukui University of Technology, Imperial College London, Technical University of Denmark, University College of Dublin, Ensam, Technical University of Darmstadt, Technical University of Eindhoven, University College London, Pontifical Comillas University, King Fahd University of Petroleum and Minerals, Shanghai Jiao Tong University, Huazhong university, Florida State University, and Technical University Kaiserslautern.
o Research centres include KTH Royal Institute of Technology, the SuperGrid Institute, GridLab, IREC (Institut de Recerca en Energia de Catalunya) and L2EP (Laboratoire d’Electrotechnique et Electronique de Puissance, Lille).
o Companies include Siemens, Tractebel, Sarawak Energy, DNV GL, IBM Research, SP Energy Networks, TenneT Offshore, Nissin, Enstore, SCiBreak, and General Electric.
• Toolbox and user manual uploaded on BEST PATHS website on 14th February 2017.
• Presentation at 13th IET ACDC2017; advertisement via social media and on project website.
• 5,076 new users have been recorded on the website since the toolbox was uploaded.
• The toolbox has been downloaded by 119 different users (until 23rd
May 2018).
Purposes of use
Testing
Information
Research
Evaluation
Training
Type of organisation
University
Research centre
Company
http://bestpaths-project.eu/en/publications
29
Simulation Results
Example: KPI Assessment
• Simulation results for three topologies are presented.
• A subset of the KPIs is shown.
GSC
Pw1
Pg1,Qg1Onshore
AC Grid #1
DC CABLE
Vdc and Q
Controller
AC Voltage
Control
Vdc_g1
Vdc_g1*fw1*|Vac_w1*|
Vac_w1
Offshore
Grid #1WFC Offshore Onshore
Qg1*
θw1*
Pw1
DC
NETWORK
AC Voltage
Control
fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
Offshore Onshore
θw1*
GSC #1Pg1,Qg1
Onshore
AC Grid #1
(Vdc vs. P) and Q
Controller
Vdc_g1* Qg1*
Vdc_g1
AC Voltage
Control
fw12*Vac_w2*
Vac_w2
WFC #2
θw2*
Pw2
Offshore
Grid #1
GSC #1
Pw1
Pg1,Qg1Onshore
AC Grid #1
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g1
Vdc_g1*fw1*Vac_w1*
Vac_w1
Offshore
Grid #1
WFC #1
Qg1*θw1*
GSC #2
Pw2
Pg2,Qg2Onshore
AC Grid #2
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g2
Vdc_g2*fw2*Vac_w2*
Vac_w2
Offshore
Grid #2WFC #2
Qg2*θw2*
GSC #3
Pw3
Pg3,Qg3Onshore
AC Grid #3
(Vdc vs. P) and Q
Controller
AC Voltage
Control
Vdc_g3
Vdc_g3*fw3*Vac_w3*
Vac_w3
Offshore
Grid #3WFC #3
Qg3*θw3*
DC
NETWORK
Offshore Onshore
Topology 1
Topology 2
Topology 3
• Full details of KPI assessment for all defined topologies available in Deliverable D3.2 of the BEST PATHS project.
30
Simulation Results (2)
Example: KPI Assessment (continued…)
• The converter control performance is assessed when references for DC voltage and reactive power are changed to onshore converter GSC in Topologies 1 and 2 and GSC2 in topology 3.
o Reactive power changed from 330 MVAr to 165 MVAr at 1.5 s;
o DC voltage changed from 640 kV to 576 kV at 1.8 s.
Topology 1 Topology 2 Topology 3
➢ Assessment of KPI.D1.1 – Steady state error (SSE)
31
Simulation Results (3)
Example: KPI Assessment (continued…)➢ Assessment of KPI.D1.6 – Grid Code Compliance
• The AC fault ride-through capability of the systems is evaluated.
o A voltage dip at an onshore grid converter is applied at 1.5 s during 300 ms, reducing the AC voltage from 1 p.u. to 0.15 p.u.
Topology 1 Topology 2 Topology 3
32
Simulation Results (4)
Example: KPI Assessment (continued…)
➢ Assessment of KPI.D1.1 – Harmonics and SSE
• The THD of the AC voltage and the converter performance are evaluated during AC voltage regulation (offshore converter).
o The offshore AC voltage (rms) is changed from 1 p.u. (380 kV) to 0.9 p.u. (342 kV) at 1.5 s.
Topology 1 Topology 2 Topology 3
33
Simulation Results (5)
KPI Assessment Summary
• Full details of the models available in Deliverable D3.2 of the BEST PATHS project.
Steady State AC/DC Interactions
Transient AC/DC Interactions
Protection Performance
DC Inter-array Design
Resonances
Grid Code Compliance
1.1
1.2
1.3
1.4
1.5
1.6
KPI Description Status
Due to converter overloading and DC overvoltage during extreme conditions (e.g. AC faults). Overloading sustained for a very short time
34
Real-Time Demonstrator
➢ Built in the premises of SINTEF (Trondheim, Norway), it aims to:
• Provide experimental validation to the results obtained from simulations:
o Establish a correspondence between simulation and experimental setup on single components and at system level;
o Identify relevant scenarios to test in the laboratory;
o Perform experiments.
• Reduce risks of HVDC link connecting OWFs.
• Validate meshed HVDC grids with different VSC technologies.
• Foster new suppliers and sub-suppliers of HVDC technology.
➢ Facilities include:
• a four-terminal 50 kW HVDC grid with 3 VSC-based MMCs and 1 two-level VSC;
• a 20 kW synchronous generator;
• DC circuit breakers;
• a wind emulator;
• a real-time simulator system and control unit (OPAL-RT).
35
Real-Time Demonstrator (2)
• Further detail on the demonstrator available in Deliverable D8.1 of the BEST PATHS project.
36
Real-Time Demonstrator (3)
➢ National SmartGrid Laboratory (SINTEF)
37
Real-Time Demonstrator (4)
➢ MMC Power Cells Boards
38
Real-Time Demonstrator (5)
➢ MMC Assembling Stages
39
Real-Time Demonstrator (6)
➢ MMC Assembling Stages (2)
40
Experimental Results
➢ Matching converter parameters of demonstrator with those of
simulation models
• The matching process was based on experimental results from the demonstrator running in open loop connected to a resistive load.
• This way, the MMC arm current and submodule voltages would depend only on converter parameters and not on the control action.
• MMC parameters were iteratively matched, including arm inductance, arm resistance, and submodule capacitance.
• With component parameters matched, the delay between measurements of arm current and submodule voltage could be determined from experimental results.
• The main aim of this iterative exercise was to:
o Increase the accuracy of the simulation models.
o Obtain a highly reliable representation to perform offline tests.
o Help ensure adequate performance of test configurations.
o Identify adverse operating conditions via software.
41
Experimental Results (2)
➢ Matching converter parameters of demonstrator with those of
simulation models (continued…)
MMC12
60 kVA
690V DC
Source
Inner Current
Loop
Power Amplifier 200 kVA
Real Time
Simulator
Iabc
id*
VabcVdc*
300V AC
Source
Vabc1* Vabc*
MMC12
60 kVA
690V DC
Source
AC Voltage
Control
Power Amplifier
200 kVA
Real Time
Simulator
Iabc
f*
V*θ*
11 Ω
Load
VabcVdc* Vabc*
• MMC with AC voltage control connected to a load resistance.
o The control schemes creates an AC voltage with a reference amplitude of 330 V and 50 Hz.
• MMC with inner current control connected to an islanded AC grid.
42
Experimental Results (3)
➢ Matching converter parameters of demonstrator with those of
simulation models (continued…)
18-level 12-level 6-level
AC
Vo
ltag
e [V
]
AC
Cu
rren
t [A
]
Arm
Cu
rren
t [A
]
MMC with AC voltage control connected to load resistance
43
Experimental Results (4)
➢ Matching converter parameters of demonstrator with those of
simulation models (continued…)
MMC with inner current control connected to islanded AC grid
18-level 12-level 6-level
RM
S A
C C
urr
ent
[A]
• Reference of d-axis current increased from 20 to 30 A at t = 2.5 s.
18-level 12-level 6-level
RM
S A
C C
urr
ent
[A]
44
Experimental Results (5)
➢ Experimental Validation for Topology A
• Sending converter (18-level half-bridge) uses a nearest level modulation (NLM) and operates in a P/Q mode.
• Receiving converter (6-level half-bridge) uses a phase disposition PWM (PD-PWM) and operates in a Vdc/Q mode.
• Both converters make use of a circulating current regulator and voltage balancing algorithms.
45
Experimental Results (6)
➢ Experimental Validation for Topology A (continued…)
• Reference currents at the sending end:
46
Experimental Results (7)
➢ Experimental Validation for Topology A (continued…)
• DC voltage. Performance upon changes in current reference id and changes in the DC voltage reference:
47
Experimental Results (8)
➢ Experimental Validation for Topology A (continued…)
• Upper and lower arm currents and voltages at the receiving end converter in steady-state.
48
On-Going Work
Point-to-Point System
Objective:
Evaluate the operation of the point-to-point link when the WF powervaries.
Procedure:
Change active power of the WF from 0 to 1 p.u. with ramp rate
limitation of 10 p.u./s.
49
On-Going Work (2)
Point-to-Point System (continued…)
0.9 1 1.1 1.2 1.3 1.4 1.5-50
0
50
Curr
ent
(A)
time(s)
19-Level AC current
Demonstrator
Simulation
0.9 1 1.1 1.2 1.3 1.4 1.5-50
0
50
Curr
ent
(A)
time(s)
7-Level AC current
0.5 1 1.5 2660
670
680
690
700
710
720
Voltage (
V)
time (s)
DC-bus Voltage Vdc
Demonstrator
Simulation
0.5 1 1.5 2-40
-30
-20
-10
0
Curr
ent
(A)
time (s)
DC-bus Current Idc
50
On-Going Work (3)
Point-to-Point System (continued…)
0.9 1 1.1 1.2 1.3 1.4 1.5-40
-30
-20
-10
0
10
20
Curr
ent
(A)
time (s)
19-Level Arm current
Demonstrator
Simulation
0.9 1 1.1 1.2 1.3 1.4 1.5-20
-10
0
10
20
30
40
Curr
ent
(A)
time (s)
7-Level Arm current
0.9 1 1.1 1.2 1.3 1.4 1.5660
680
700
720
740
Voltage (
V)
time (s)
19-Level Arm voltage
Demonstrator
Simulation
0.9 1 1.1 1.2 1.3 1.4 1.5660
680
700
720
740
Voltage (
V)
time (s)
7-Level Arm voltage
51
On-Going Work (4)
Three-Terminal System
Objective:
Evaluate the operation of a three-terminal system when the WF power varies.
Procedure:
- Set the power of the PQ node to -0.5 p.u (injecting power into the
grid).
- Change active power of the WF from 0 to 1 p.u. with ramp rate
limitation 10 p.u./s
52
On-Going Work (5)
Three-Terminal System (continued…)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 20
50
100
7-Level Id
time(s)
Curr
ent
(A)
Demonstrator
Simulation
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
-60
-40
-20
0
19-Level Id
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
-60
-40
-20
0
13-Level Id
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
680
700
720DC-bus voltage Vdc
time(s)
Voltage (
V)
Demonstrator
Simulation
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 20
50
7-Level DC current
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2-80-60-40-20
0
19-Level DC current
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2-80-60-40-20
0
13-Level DC current
time(s)
Curr
ent
(A)
53
On-Going Work (6)
Three-Terminal System (continued…)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
-50
0
50
7-Level Arm current
time(s)
Curr
ent
(A)
Demonstrator
Simulation
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
-50
0
50
19-Level Arm current
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
-50
0
50
13-Level Arm current
time(s)
Curr
ent
(A)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
650
700
750
7-Level Arm voltage
time(s)
Voltage (
V)
Demonstrator
Simulation
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
650
700
750
19-Level Arm voltage
time(s)V
oltage (
V)
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
650
700
750
13-Level Arm voltage
time(s)
Voltage (
V)
54
On-Going Work (7)
Three-Terminal System – TEST TWO
Objective:
Evaluate the operation of a three-terminal system when the power flow of the PQ node is reversed.
Procedure:
- Set the power of the WF to 0.5 p.u.
- Change the active power of the PQ node from -0.5 p.u to 0.5 p.u.
with ramp rate limitation 10 p.u./s.
55
On-Going Work (8)
Three-Terminal System – TEST TWO (continued…)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-100
0
1007-Level Id
time(s)
Curr
ent
(A)
Demonstrator
Simulation
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-100
0
10019-Level Id
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-100
0
10013-Level Id
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2600
650
700
DC-bus voltage Vdc
time(s)
Voltage (
V)
Demonstrator
Simulation
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-50
0
50
7-Level DC current
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-50
0
50
19-Level DC current
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2-50
0
50
13-Level DC current
time(s)
Curr
ent
(A)
56
On-Going Work (9)
Three-Terminal System – TEST TWO (continued…)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2
-50
0
50
7-Level Arm current
time(s)
Curr
ent
(A)
Demonstrator
Simulation
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2
-50
0
50
19-Level Arm current
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2
-50
0
50
13-Level Arm current
time(s)
Curr
ent
(A)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2600
650
700
7-Level Arm voltage
time(s)
Voltage (
V)
Demonstrator
Simulation
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2600
650
700
19-Level Arm voltage
time(s)V
oltage (
V)
1.5 1.6 1.7 1.8 1.9 2 2.1 2.2600
650
700
13-Level Arm voltage
time(s)
Voltage (
V)
57
On-Going Work (10)
Interactions of droop and power control characteristics
Conv2Pg2,Qg2 AC Grid
#2
(Vdc vs. P) and Q
Controller
Vdc_g2
Qg2*
DC
NETWORK
Conv3Pg3,Qg3AC Grid
#3
P and Q
Controller
Pg3*Qg3*
Vdc_g3
Conv1Pg1,Qg1
AC Grid
#1
P and Q
Controller
Qg1*
Vdc_g1
Vdc_g1* Vdc_g2*
• Different droop and power control strategies could have adverse interactions from multiple crossings of the control characteristics.
• Conv1 (in black): an exporting converter (inverter).
• Conv2 (in blue): an importing converter (rectifier), voltage
reference is given as 1 p.u. and the droop gain is 1.25 p.u.
• Conv3 (in red): an exporting converter (inverter).
• Bases are 650 V, 32500 W and 50 A.
58
On-Going Work (11)
Interactions of droop and power control characteristics
• The green power curve is an equivalent curve of adding those of Conv1 (black line) and Conv3 (red line).
• The droop curve is intersected by the power curve. This results in three different operation points.
Vdc
1 p.u. Idc
Vopt
Vopt1
Vopt2 Conv2
1 p.u.
Conv2
Power change
from 0 to
|32500| W
Conv3 +Conv1 Conv3 +Conv1
0 p.u.
Vdc
Conv3 Conv1
0.31 p.u. 0.69 p.u. Idc
Vdc
Conv3 +Conv1
1 p.u.
1 p.u.
Idc
59
On-Going Work (12)
Interactions of droop and power control characteristics
Power/current
reference (p.u.)
1.5 s Time (s)1.7 s
(1 p.u.)
(-0.31 p.u.)
(-0.69 p.u.)
Conv3
Conv1
Conv2
• Initial power referencesfor all converters are 0 p.u.
• From 1.5 to 1.7 s, ramp changes are applied to the power references. The power reference of Conv1is changed to -1 p.u. and the power references of Conv2 and Conv3 are changed to 0.69 p.u. and 0.31 p.u., respectively.
• DC voltage in steady-state shifted.
• The current cannot be regulated effectively.
60
On-Going Work (13)
Interactions of droop and power control characteristics
• An additional test is performed, where the sign of all power references in the previous test is reversed.
• The multiple-crossing problem is avoided.
Vdc
1 p.u. Idc
Vopt
Conv1
1 p.u.
Conv1
Power change
from 0 to
|32500| W
Conv3 +Conv1 Conv3 +Conv1
61
On-Going Work (14)
Interactions of droop and power control characteristics
• I-V trajectory for Conv 2’s voltage and current.
• The voltage for the characteristic without multiple crossings (green curve)
stays around 650 V (from Vopt to Vopt_new). Following the transient regime, the
current stays at 50 A (1 p.u.) at one end and hence it is well regulated.
• The voltage for the characteristic with multiple crossings (yellow curve)
shifts from 650 to 610 V (Vopt2) in steady-state and also there are significant
oscillations. The current settles at −54.2 A (−1.084 p.u.) and hence not
accurately regulated at −50 A.
𝑉𝑜𝑝𝑡
𝑉𝑜𝑝𝑡 _𝑛𝑒𝑤
𝑉𝑜𝑝𝑡 2
62
Conclusions and Next Steps
• A set of models and control algorithms has been developed, simulated and assessed. These have been published as an ‘Open Access’ Toolbox.
• Network topologies constituting likely scenarios for the transmission of offshore wind energy have been proposed.
• To assess the suitability of the models, topologies and control algorithms, a set of KPIs have been defined.
• An experimental demonstrator for the integration of grid-connected OWFs using HVDC grids has been presented.
• Results demonstrating the capabilities of the demonstrator have been compared against simulation results. These show good agreement.
➢ Main Contributions of this Work
63
Conclusions and Next Steps (2)
➢ Main Contributions of this Work (continued)
• The main contribution of this work is the provision to TSOs, utilities, manufacturers and academic institutions with simulation and experimental tools to generate the necessary knowledge for the development, construction and connection of MTDC systems –aiming to help de-risking the use of MTDC grids for the connection of OWFs.
• Using the real-time experimental demonstrator, conduct tests for different system topologies representing future scenarios to validatesimulation results obtained using computational tools.
• Make the demonstrator available to interested parties for R&D activities.
➢ On-Going and Future Work
64
Conclusions and Next Steps (3)
➢ Some papers linked to this presentation
• Ugalde-Loo CE, et al., “Lessons Learnt from the BEST PATHS Project for the Integration of Offshore Wind Farms using Multi-Terminal HVDC Grids,” 47th CIGRE Session 2018, Paris, France, 26-31 August 2018, pp. 1-10.
• Parker M, Finney S, Holliday D, “DC protection of a multi-terminal HVDC network featuring offshore wind farms,” Energy Procedia, vol. 142, Dec. 2017, pp. 2195-2201.
• Azpiri I, Ciapessoni E, Cirio D, Glasdam J, Lund P, Pitto P, “Grid code compliant controllers for multi-terminal HVDC grids aimed to integrate wind power: assessing their impact on the operational security of a real-world system,” Energy Procedia, vol. 142, Dec. 2017, pp. 2165-2170.
• Ciapessoni, et al., “Assessing the impact of multi-terminal HVDC grids for wind integration on future scenarios of a real-world AC power system using grid code compliant open models,” 2017 IEEE Power Tech, Manchester, UK, 18-22 June 2017, pp. 1-6
• Ugalde-Loo CE, et al., “BEST PATHS Project: Real-Time Demonstrator for the Integration of Offshore Wind Farms using Multi-Terminal HVDC Grids,” Offshore Wind Energy (OWE 2017), London, UK, 6-8 June 2017, pp. 1-11.
• Ugalde-Loo CE, et al., “Open access simulation toolbox for the grid connection of offshore wind farms using multi-terminal HVDC networks,” 13th IET International Conference on AC and DC Power Transmission (ACDC 2017), Manchester, UK, 14-16 February 2017, pp. 1-6.
65
Questions?
The authors gratefully acknowledge the financial support provided by the EU FP7 Programme through the project “BEyond State of the art Technologies for re-Powering AC corridors & multi-Terminal HVDC Systems” (BEST PATHS), grant agreement number 612748.
Dr Carlos UGALDE
Ugalde-LooC@cardiff.ac.uk
Cardiff University, Wales, UK