Integration of Offshore Wind Farms using HVDC Technologies: Results from the BEST PATHS Project

Size: px

Start display at page:

Download "Integration of Offshore Wind Farms using HVDC Technologies: Results from the BEST PATHS Project"

Jane Reynolds
5 years ago
Views:

May 28 June 1, 2018 Leuven, Belgium EES-UETP HVDC and HVDC Grids for Future Transmission Integration of Offshore Wind Farms using HVDC Technologies: Results from the BEST PATHS Project Dr. Carlos E.

1 May 28 June 1, 2018 Leuven, Belgium EES-UETP HVDC and HVDC Grids for Future Transmission Integration of Offshore Wind Farms using HVDC Technologies: Results from the BEST PATHS Project Dr. Carlos E. UGALDE-LOO Cardiff University, Wales, UK 1 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 2 4. Simulation Results 5. Real-Time Demonstrator 6. Experimental Results 7. On-Going (Simulation and Experimental) Work 8. Conclusions and Next Steps

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.

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 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: GW** 3 * **

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. 4

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,

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. 5

6 Introduction (4) > 100 km Substation Wind farm 6

7 Introduction (5) Substation Wind farm 7

8 Introduction (6) HVAC HVDC Substation Wind farm 8

9 The BEST PATHS Project BEyond State-of-the-art Technologies for re-powering Ac corridors & multi-terminal Hvdc Systems Key Figures Budget of 62.8M, 56% co-funded by the European Commission under the 7 th Framework Programme for Research, Technological Development and Demonstration (EU FP7 Energy). Duration: 01/10/ /09/2018 (4 years). Composition: 5 large-scale demonstrations, 2 replication projects, 1 dissemination project. 9 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.

10 The BEST PATHS Project (2) BEyond State-of-the-art Technologies for re-powering Ac corridors & multi-terminal Hvdc Systems 10

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 using scaled models. 4. To use the validated models to simulate a real grid with OWFs connected in HVDC. 11

12 BEST PATHS Demo #1 (2) HVDC equipment manufacturers provide black boxes? R&D Centres We intend to use open models TSOs Utilities & RES developers Detailed models Simulation & Validation Independent Manufacturers 12

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. Point-to-Point HVDC Link (Topology A) Offshore Grid #1 V ac_w1 WFC Offshore Onshore GSC P g1,q g1 Onshore AC Grid #1 V dc_g1 P w1 DC CABLE V ac_w1 * AC Voltage Control f w1 * θ w1 * V dc_g1 * V dc and Q Controller Q g1 * 13

14 Network Topologies (2) Three-Terminal HVDC System Offshore Grid #1 V ac_w2 WFC #2 Offshore Onshore P w2 AC Voltage Control θ w2 * DC NETWORK V ac_w2 * f w12 * Offshore Grid #1 V ac_w1 WFC #1 GSC #1 P g1,q g1 Onshore AC Grid #1 V dc_g1 P w1 V ac_w1 * AC Voltage Control f w1 * θ w1 * (V dc vs. P) and Q Controller V dc_g1 * Q g1 * 14

15 Network Topologies (3) Six-Terminal HVDC System with Offshore AC Links (Topology B) Offshore Grid #3 V ac_w3 WFC #3 Offshore Onshore GSC #3 P g3,q g3 Onshore AC Grid #3 V dc_g3 P w3 AC interlink V ac_w3 * AC Voltage Control θ w3 * f w3 * V dc_g3 * (V dc vs. P) and Q Controller Q g3 * Offshore Grid #2 V ac_w2 WFC #2 DC NETWORK GSC #2 P g2,q g2 Onshore AC Grid #2 V dc_g2 P w2 V ac_w2 * AC Voltage Control θ w2 * f w2 * V dc_g2 * (V dc vs. P) and Q Controller Q g2 * Offshore Grid #1 V ac_w1 WFC #1 GSC #1 P g1,q g1 Onshore AC Grid #1 V dc_g1 P w1 V ac_w1 * AC Voltage Control θ w1 * f w1 * V dc_g1 * (V dc vs. P) and Q Controller Q g1 * 15

16 Network Topologies (4) Six-Terminal HVDC System with Offshore DC Links (Topology C) Offshore Grid #3 V ac_w3 WFC #3 Offshore Onshore GSC #3 P g3,q g3 Onshore AC Grid #3 V dc_g3 P w3 V ac_w3 * AC Voltage Control θ w3 * f w3 * V dc_g3 * (V dc vs. P) and Q Controller Q g3 * Offshore Grid #2 V ac_w2 WFC #2 DC NETWORK GSC #2 P g2,q g2 Onshore AC Grid #2 V dc_g2 P w2 V ac_w2 * AC Voltage Control θ w2 * f w2 * DC interlink V dc_g2 * (V dc vs. P) and Q Controller Q g2 * Offshore Grid #1 V ac_w1 WFC #1 GSC #1 P g1,q g1 Onshore AC Grid #1 V dc_g1 P w1 V ac_w1 * AC Voltage Control θ w1 * f w1 * V dc_g1 * (V dc vs. P) and Q Controller Q g1 * 16

17 Network Topologies (5) Twelve-Terminal HVDC System with Offshore Onshore Vdc_g6 GSC #6 Pg6,Qg6 Onshore AC Grid #B Offshore DC Links (Topology D) (Vdc vs. P) and Q Controller Vdc_g6* Qg6* DC NETWORK GSC #5 Pg5,Qg5 Vdc_g5 (100 km) Offshore Grid #3 Vac_w3 WFC #3 (10 km) WFC #6 Vac_w6 Offshore Grid #6 (Vdc vs. P) and Q Controller Vdc_g5* Qg5* Pw3 Pw6 GSC #4 Pg4,Qg4 Vac_w3* AC Voltage Control θw3* fw3* (10 km) (5 km) AC Voltage Control fw6* θw6* Vac_w6* Vdc_g4 Offshore Grid #2 Vac_w2 WFC #2 (10 km) WFC #5 Vac_w5 Offshore Grid #5 Vdc_g4* (Vdc vs. P) and Q Controller Qg4* Pw2 Vac_w2* AC Voltage Control θw2* fw2* (10 km) AC Voltage Control fw5* θw5* Vac_w5* Pw5 Offshore Onshore Vdc_g3 GSC #3 Pg3,Qg3 Onshore AC Grid #A Offshore Grid #1 Vac_w1 WFC #1 WFC #4 Vac_w4 Offshore Grid #4 (Vdc vs. P) and Q Controller Vdc_g3* Qg3* Pw1 AC Voltage Control AC Voltage Control Pw4 DC NETWORK GSC #2 Pg2,Qg2 Vac_w1* θw1* fw1* fw4* θw4* Vac_w4* Vdc_g2 (100 km) (Vdc vs. P) and Q Controller Vdc_g2* Qg2* GSC #1 Pg1,Qg1 Vdc_g1 17 Vdc_g1* (Vdc vs. P) and Q Controller Qg1*

18 Key Performance Indicators To assess the suitability of the models and proposed HVDC network topologies, converter configurations and control algorithms, a set of KPIs have been defined. Full details available in Deliverable D2.1 of the BEST PATHS project. KPI.D1.1 AC/DC interactions: power and harmonics Steady state Power quality WT ramp rates KPI.D1.4 DC Inter-array Design Inter-array topology Power unbalance Fault tolerance Motorising capability KPI.D1.2 AC/DC Interactions Transients & Voltage Margins Normal operation Extreme operation KPI.D1.5 Resonances AC systems oscillation Internal DC resonance KPI.D1.3 DC Protection Performance / Protection & Faults Protection selectivity Peak current and clearance time KPI.D1.6 Grid Code Compliance Active and reactive power Fault ride-through 18

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. 19

The Open Access Toolbox (2) The models and control algorithms have been

on AC and DC Power Transmission (ACDC2017): o CE Ugalde-Loo, et al.

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: 20 The toolbox was originally presented last year in the 13 th 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, 13 th IET ACDC17, Manchester, UK, 2017.

21 The Open Access Toolbox (3) A user manual is also provided, together with the published models and accompanying examples. 21 Full details of the models available in Deliverable D3.1 of the BEST PATHS project.

22 The Open Access Toolbox (4) Converter Stations 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 Grid AC network adapted from the classical nine-bus power system. 22 DC Cable The DC cable section has been modelled as a one-phase, frequencydependent, 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. 23

24 The Open Access Toolbox (6) High Level Controller High Level Controller It considers converter operation in three control modes. The aim is to cover the main control needs for different system configurations. Outer Loop iabc Vabc PLL abc-dq AC System Imax θ Mode 1 Mode 2 Mode 0 Vdc Q P Q AC Island Select id1 *, iq1 * idq Vdq Vabc P Vdc Droop Inner Current Loop θ Vdq* dq-abc vabc* iabc L Vdc P Droop Current Limiter idq* Select id2 *, iq2 * dq-φ θ* θ Half-Bridge MMC Mode Selector Low Level Controller AC Voltage Control θwf* ± VDC vabc* DC Grid fwf* θwf* Vwf* 24

25 The Open Access Toolbox (7) High Level Controller o Mode 0: V ac voltage control: The converter sets the voltage and frequency. Offshore AC Grid P wf WFC V wf V wf * + - f wf * = const ki3 kp3+ s θ wf * = const m a f θ M V abc v abc Switching Control * v abc ω = 2πf V wf,a =Msin(ωt +θ) V wf,b = Msin(ωt+θ+2π/3) V wf,c =Msin(ωt+θ-2π/3) V dc 25

26 The Open Access Toolbox (8) High Level Controller Mode 1: V dc Q control scheme with a P V dc droop Main AC Grid L i abc V dc MTDC system V dc Outer Loop i dq1 * P, Q Current Limiter V abc PLL θ abc-dq v dq i dq* Inner Current Loop abc-dq i dq v dq * v abc Switching Control dq-abc v abc * θ Vdc,0* P* + - P Vdc ki,v kp,v Imax id s kpv Vdc* ΔVdc,0 Outer Loop Q* -1 vd id1* IQ* Radius Current Limiter Limited iq* q-axis Limited id* iq* before limiting Imax id* before limiting d-axis id* iq* Inner Current Loop iq vd + ki - kp s ωl ωl + - ki kp s vd* vq* 26

27 The Open Access Toolbox (9) High Level Controller Mode 2: P Q control scheme with a V dc P droop Main AC Grid v abc L i abc V dc MTDC system V dc Outer Loop P, Q i dq2 * Current Limiter PLL θ abc-dq v dq i dq * Inner Current Loop abc-dq i dq v dq * v abc Switching Control dq-abc v abc * θ Outer Loop P0* Vdc* + - Vdc kvp ΔP* P P* Radius iq* before 1 limiting Imax id + Q* vd -1 vd id2* id2* Current Limiter Limited iq* q-axis Limited id* Imax id* before limiting d-axis id* iq* Inner Current Loop iq vd + ki - kp s ωl ωl + - ki kp s vd* vq* 27

28 The Open Access Toolbox (10) Toolbox and user manual uploaded on BEST PATHS website on 14 th February Presentation at 13 th 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 23 rd May 2018). Purposes of use Testing Information Research Evaluation Training Type of organisation University Research centre Company 28 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.

29 Simulation Results Example: KPI Assessment Simulation results for three topologies are presented. Topology 3 A subset of the KPIs is shown. Offshore Grid #3 Vac_w3 WFC #3 Offshore Onshore GSC #3 Pg3,Qg3 Onshore AC Grid #3 Vdc_g3 Pw3 Topology 1 Vac_w3* AC Voltage Control θw3* fw3* Vdc_g3* (Vdc vs. P) and Q Controller Qg3* Offshore Grid #1 V ac_w1 WFC Offshore Onshore GSC P g1,q g1 Onshore AC Grid #1 Offshore Grid #2 Vac_w2 WFC #2 DC NETWORK Vdc_g2 GSC #2 Pg2,Qg2 Onshore AC Grid #2 Vdc_g1 P w1 V ac_w1* AC Voltage Control f w1* θ w1* DC CABLE V dc_g1* V dc and Q Controller Q g1* Pw2 Offshore Grid #1 Vac_w2* Vac_w1 AC Voltage Control θw2* WFC #1 fw2* Vdc_g2* (Vdc vs. P) and Q Controller GSC #1 Qg2* Pg1,Qg1 Onshore AC Grid #1 Vdc_g1 Pw1 Topology 2 Vac_w1* AC Voltage Control θw1* fw1* Vdc_g1* (Vdc vs. P) and Q Controller Qg1* Offshore Grid #1 Vac_w2 WFC #2 Offshore Onshore 29 Pw2 Offshore Grid #1 Pw1 Vac_w2* Vac_w1 Vac_w1* AC Voltage Control WFC #1 fw12* AC Voltage Control fw1* θw2* θw1* DC NETWORK Vdc_g1 Vdc_g1* GSC #1 (Vdc vs. P) and Q Controller Qg1* Pg1,Qg1 Onshore AC Grid #1 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 ) Assessment of KPI.D1.1 Steady state error (SSE) 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 30

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 31

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 32

33 Simulation Results (5) KPI Assessment Summary KPI Description Status 1.1 Steady State AC/DC Interactions Fully Met 1.2 Transient AC/DC Interactions o Partially met Due to converter overloading and DC overvoltage during extreme conditions (e.g. AC faults). Overloading sustained for a very short time <300ms and braking resistor prevents overvoltage. 1.3 Protection Performance Fully Met 1.4 DC Inter-array Design Fully Met 1.5 Resonances Fully Met 1.6 Grid Code Compliance o Partially met Due to steady-state error between actual and reference active power during frequency oscillations on the AC grid of Topology A & B. 33 Full details of the models available in Deliverable D3.2 of the BEST PATHS project.

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). 34

35 Real-Time Demonstrator (2) 35 Further detail on the demonstrator available in Deliverable D8.1 of the BEST PATHS project.

36 Real-Time Demonstrator (3) National SmartGrid Laboratory (SINTEF) 36

37 Real-Time Demonstrator (4) MMC Power Cells Boards 37

38 Real-Time Demonstrator (5) MMC Assembling Stages 38

39 Real-Time Demonstrator (6) MMC Assembling Stages (2) 39

40 Experimental Results 40 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 ) Power Amplifier 200 kva Vdc* 690V DC Source Real Time Simulator f* V* θ* MMC12 60 kva Vabc* AC Voltage Control Iabc Vabc 11 Ω Load 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 V AC Source Power Amplifier 200 kva 690V DC Source Vabc1* Real Time Simulator Vdc* id* MMC12 60 kva Vabc* Inner Current Loop Iabc Vabc MMC with inner current control connected to an islanded AC grid.

42 Arm Current [A] AC Current [A] AC Voltage [V] Experimental Results (3) Matching converter parameters of demonstrator with those of simulation models (continued ) MMC with AC voltage control connected to load resistance 18-level 12-level 6-level 42

RMS AC Current [A] Experimental Results (4) Matching converter parameters of demonstrator with those of simulation models (continued ) 12-level MMC

43 RMS AC Current [A] Experimental Results (4) Matching converter parameters of demonstrator with those of simulation models (continued ) 12-level MMC with inner current control connected to islanded AC grid 6-level Reference of d-axis current increased from 20 to 30 A at t = 2.5 s. 18-level 12-level 43

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 V dc /Q mode. Both converters make use of a circulating current regulator and voltage balancing algorithms. 44

45 Experimental Results (6) Experimental Validation for Topology A (continued ) Reference currents at the sending end: 45

46 Experimental Results (7) Experimental Validation for Topology A (continued ) DC voltage. Performance upon changes in current reference i d and changes in the DC voltage reference: 46

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. 47

48 On-Going Work Point-to-Point System Objective: Evaluate the operation of the point-to-point link when the WF power varies. Procedure: Change active power of the WF from 0 to 1 p.u. with ramp rate limitation of 10 p.u./s. 48

49 Current (A) Current (A) Current (A) Voltage (V) On-Going Work (2) Point-to-Point System (continued ) Level AC current DC-bus Voltage Vdc Demonstrator Simulation Demonstrator Simulation time(s) 7-Level AC current time (s) 0 DC-bus Current Idc time(s) time (s) 49

50 Current (A) Voltage (V) Current (A) Voltage (V) On-Going Work (3) Point-to-Point System (continued ) Level Arm current Level Arm voltage Demonstrator Simulation Demonstrator Simulation time (s) time (s) 40 7-Level Arm current Level Arm voltage time (s) time (s) 50

51 On-Going Work (4) Three-Terminal System Objective: Evaluate the operation of a threeterminal system when the WF power varies. Procedure: 51 - 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 Current (A) Current (A) Current (A) Current (A) Current (A) Current (A) Voltage (V) On-Going Work (5) Three-Terminal System (continued ) 7-Level Id Demonstrator Simulation time(s) 19-Level Id Demonstrator DC-bus voltage Vdc 720 Simulation time(s) 7-Level DC current time(s) 13-Level Id time(s) time(s) 19-Level DC current time(s) 13-Level DC current time(s)

53 Current (A) Voltage (V) Current (A) Voltage (V) Current (A) Voltage (V) On-Going Work (6) Three-Terminal System (continued ) 7-Level Arm current 7-Level Arm voltage Demonstrator Simulation time(s) 19-Level Arm current time(s) 13-Level Arm current time(s) 700 Demonstrator 650 Simulation time(s) 19-Level Arm voltage time(s) 13-Level Arm voltage time(s) 53

54 On-Going Work (7) Three-Terminal System TEST TWO Objective: Evaluate the operation of a threeterminal system when the power flow of the PQ node is reversed. 54 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 Current (A) Current (A) Current (A) Current (A) Current (A) Current (A) Voltage (V) On-Going Work (8) Three-Terminal System TEST TWO (continued ) Level Id Demonstrator Simulation time(s) 19-Level Id DC-bus voltage Vdc Demonstrator Simulation time(s) 7-Level DC current time(s) 19-Level DC current time(s) 13-Level Id time(s) 13-Level DC current time(s) time(s) 55

56 Current (A) Voltage (V) Current (A) Voltage (V) Current (A) Voltage (V) On-Going Work (9) Three-Terminal System TEST TWO (continued ) 7-Level Arm current 7-Level Arm voltage 50 Demonstrator Simulation time(s) 19-Level Arm current Demonstrator Simulation time(s) 19-Level Arm voltage time(s) 13-Level Arm current time(s) 13-Level Arm voltage time(s) time(s) 56

57 On-Going Work (10) Interactions of droop and power control characteristics 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). AC Grid #3 P g3,q g3 Conv3 V dc_g3 Bases are 650 V, W and 50 A. AC Grid #1 P g1,q g1 Q g3 * P and Q Controller Conv1 P g3 * DC NETWORK Conv2 P g2,q g2 AC Grid #2 V dc_g1 V dc_g2 57 Q g1 * P and Q Controller V dc_g1 * V dc_g2 * (V dc vs. P) and Q Controller Q g2 *

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). Vdc Vdc Conv3 Conv1 Conv3 +Conv1 1 p.u p.u p.u. Idc 1 p.u. Idc The droop curve is intersected by the power curve. This results in three different operation points. Vdc Conv3 +Conv1 Power change from 0 to W Vopt1 Conv3 +Conv1 1 p.u. Conv2 Vopt Vopt2 Conv p.u. 1 p.u. Idc

59 On-Going Work (12) Interactions of droop and power control characteristics Power/current reference (p.u.) (1 p.u.) Conv2 DC voltage in steady-state shifted. The current cannot be regulated effectively. 1.5 s 1.7 s Time (s) (-0.31 p.u.) (-0.69 p.u.) Conv1 Conv3 Initial power references for 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 Conv1 is 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. 59

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. Vdc Conv3 +Conv1 Power change from 0 to W Conv3 +Conv1 1 p.u. Vopt Conv1 Conv1 1 p.u. Idc The multiple-crossing problem is avoided. 60

61 On-Going Work (14) Interactions of droop and power control characteristics I-V trajectory for Conv 2 s voltage and current. V opt _new V opt V opt 2 61 The voltage for the characteristic without multiple crossings (green curve) stays around 650 V (from V opt to V opt_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 (V opt2 ) in steady-state and also there are significant oscillations. The current settles at 54.2 A ( p.u.) and hence not accurately regulated at 50 A.

62 Conclusions and Next Steps Main Contributions of this Work 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 gridconnected OWFs using HVDC grids has been presented. Results demonstrating the capabilities of the demonstrator have been compared against simulation results. These show good agreement. 62

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. On-Going and Future Work Using the real-time experimental demonstrator, conduct tests for different system topologies representing future scenarios to validate simulation results obtained using computational tools. Make the demonstrator available to interested parties for R&D activities. 63

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, 47 th CIGRE Session 2018, Paris, France, August 2018, pp 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 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 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, 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 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, February 2017, pp. 1-6.

Questions? Dr Carlos UGALDE Ugalde-LooC@cardiff.ac.

65 Questions? Dr Carlos UGALDE Cardiff University, Wales, UK 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

BEST PATHS Project: Real-Time Demonstrator for the Integration of Offshore Wind Farms using Multi- Terminal HVDC Grids

BEST PATHS Project: Real-Time Demonstrator for the Integration of Wind Farms using Multi- Terminal HVDC Grids Carlos UGALDE (Cardiff University, Wales) Salvatore D ARCO (SINTEF, Norway); Daniel ADEUYI,