Smart Grid Gotland - Wind Power Integration
1 like747 views
1) The document discusses a research project analyzing how distributed demand response (DR) resources on the island of Gotland could help integrate additional wind power capacity into the local distribution network. 2) The research used real wind power and load data from 2012 to model different DR clusters and scenarios, including detached houses, industries and an energy storage system. 3) The results found that a minimum of 1300-1700 DR participants could help balance up to 5 MW of additional wind power across different seasons and scenarios while maintaining customer comfort levels. Participation from large industries further reduced the required cluster size. 4) While the study had some limitations, it demonstrated the potential benefits of DR for the Smart Grid Gotland project's objectives
![Required cluster
size (w/o Cementa)
x 1900
x 1600 (LT)
x 300 (ST)
Research > Results & Findings
Total power to balance per day and scenario
0,00
5,00
10,00
15,00
20,00
25,00
30,00
35,00
40,00
Scenario 1: Winter days Scenario 2:Spring days Scenario 3: Summer days Scenario 4: Autumn days
Scenarios
Power[MW]
day 1
day 2
day 3
Total number of hourly export problems per day and scenario
6
7
8
Scenario 1: Winter
Scenario 2: Spring
Scenario 3: Summer
Scenario 4: Autumn
17%
20%
23%
16
The minimum number of par4cipants required to
solve all export problems for all scenarios
1300
1500
1700
1100
1000
900
900
1000
1500
1000
900
1700
Minimum
cluster size/day
h](https://image.slidesharecdn.com/danielbrodensggwpipresentation050314-160421084835/85/Smart-Grid-Gotland-Wind-Power-Integration-16-320.jpg)
![0 12 24 36 48 60 72
16
Time [hours]
Indoor temperature
Less than +/-‐ 1°C
varia4on for all
seasonal scenarios.
Comfort level is kept!
Research > Results & Findings
17%
20%
23%
17
0 12 24 36 48 60 72
16
17
18
19
20
21
22
23
24
25
Time [hours]
Temperature[degeesC]
Indoor temperature change for a LT household participant
Winter
Spring
Summer
Autumn](https://image.slidesharecdn.com/danielbrodensggwpipresentation050314-160421084835/85/Smart-Grid-Gotland-Wind-Power-Integration-17-320.jpg)
![0 12 24 36 48 60 72
16
Time [hours]
Tank temperature
Varia4ons are
within the
boundaries
for all scenarios.
Comfort level is kept!
Research > Results & Findings
18
Autumn
Summer
Spring
Winter
Tank temperature change for LT household participant
Tanktemperature[degeesC]
Time [hours]
0 12 24 36 48 60 72
40
50
60
70
80
90
100
110
120
130
Figure 12:
17%
20%
23%](https://image.slidesharecdn.com/danielbrodensggwpipresentation050314-160421084835/85/Smart-Grid-Gotland-Wind-Power-Integration-18-320.jpg)
![BESS opera4on
(winter scenario)
No wind curtailment
needed in this
scenario
Research > Results & Findings
17%
20%
23%
19
Max BESS capacity
Wind curtailment
BESS level
Power[kW]
Time [hours]
0 12 24 36 48 60 72
0
50
100
150
200
250
300
h
BESS charges to account for prognosis errors not
accounted by the DR par4cipants](https://image.slidesharecdn.com/danielbrodensggwpipresentation050314-160421084835/85/Smart-Grid-Gotland-Wind-Power-Integration-19-320.jpg)
Smart Grid Gotland - Wind Power Integration
- 1. Smart Grid Gotland Smart Grid Gotland -‐ Wind Power Integra4on Reference Group Mee4ng, Visby, 5th March 2014, Daniel A. Brodén
- 2. SGG > Wind Power Integra4on Power quality with distributed generation Smart SCADA Smart Meters Energy storage Market installations Information and Communication Technology (ICT) Wind power integration Market test Smart substations and rural grid 2
- 3. Situa4on on Gotland -‐ Wind power capacity ~170 MW -‐ Max grid capacity 195 MW -‐ HVDC capacity 2x130 MW -‐ ~21,000 detached houses -‐ 3 major industries HVDC cables 3 ~170 MW wind power
- 4. Situa4on on Gotland 2012 Produc4on & Consump4on HVDC cables 170 MW installed capacity195 MW installed capacity 70 MW export 90 MW export 195 MW wind power 4
- 5. Situa4on on Gotland HVDC cables +5 MW 2012 Produc4on & Consump4on 200 MW installed capacity 95 MW export However, the risk s4ll exists! Max prod -‐ Min cons = 200 -‐ 65 = 135 135 MW > 130 MW! 2012 was risk free! 5 200 MW wind power
- 6. Situa4on on Gotland Demand-‐Response HVDC cables +5 MW Demand-‐Response Demand-‐Response can help increase the hos4ng capacity of wind power on Gotland and solve conges4on problems in the network Demand-‐Response Management System (DRMS) 200 MW wind power 6
- 7. Gotland Challenges Produc4on Prognosis on Short-‐Term (hour-‐ahead) Produc4on Prognosis on Long-‐Term (day-‐ahead) Actual Prognosis Uncertainty in produc4on prognosis makes it difficult to rely on Demand-‐Response for conges4on management 7
- 8. Research Is it technically feasible to balance 5 MW addi4onal wind power capacity in the exis4ng distribu4on network with an Ancillary Service Toolbox? 8
- 9. Data Inputs (2012) Research > AS Toolbox Wind data Load data Network Simulator Flexibility tools Long-‐Term DR Short-‐Term DR Bacery Wind Curtail. 9
- 10. Research > Forming Clusters HVDC cables The cluster op4miza4on is executed sequen4ally where the ST cluster minimizes the prognosis errors from the LT cluster Long-‐Term Cluster Op4mized consump4on schedule set 24 hours ahead Op4mized consump4on schedule set hourly +5 MW Short-‐Term Cluster 10 Prod Cons Peak hours Load shift Prod Cons Load shift LT prognosis Peak hours ST+LT prognosis 200 MW wind power
- 11. Research > Forming Clusters HVDC cables Long-‐Term Cluster Short-‐Term Cluster +5 MW Bacery Energy Storage System Absorbs the prognosis errors from the ST cluster. 11 200 MW wind power
- 12. Research > Detached House Model 12 Mathema4cal Modeling Domes4c hot waterSpace hea4ng 75 % of all electricity in a detached house is consumed by space hea4ng and domes4c hot water Consump4on es4mates of detached houses
- 13. Research > Detached House Model Consump4on Model based on “Forecasting household consumer electricity load profiles with a combined physical and behavioral approach” by Claes Sandels, ICS, KTH Model validated with the consump4on of 41 Swedish residents living in detached houses 13
- 14. Research > Industry Model 4me (h)12 24 MWh 2.8 14 Industrial consump4on shares Cementa (86%), Nordkalk (6% ) Arla (5%), Others (3%) A poten4al DR ac4vity for Cementa modeled Addi4onal produc4on ac4vity during weekdays & dayshijs
- 15. Research > Simula4on Setup 17% 20% 23% -‐ 3 day periods simulated -‐ Considering seasonal varia4on -‐ 2012 data in hourly granularity -‐ Data adjusted to provoke export problem 15
- 16. Required cluster size (w/o Cementa) x 1900 x 1600 (LT) x 300 (ST) Research > Results & Findings Total power to balance per day and scenario 0,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 40,00 Scenario 1: Winter days Scenario 2:Spring days Scenario 3: Summer days Scenario 4: Autumn days Scenarios Power[MW] day 1 day 2 day 3 Total number of hourly export problems per day and scenario 6 7 8 Scenario 1: Winter Scenario 2: Spring Scenario 3: Summer Scenario 4: Autumn 17% 20% 23% 16 The minimum number of par4cipants required to solve all export problems for all scenarios 1300 1500 1700 1100 1000 900 900 1000 1500 1000 900 1700 Minimum cluster size/day h
- 17. 0 12 24 36 48 60 72 16 Time [hours] Indoor temperature Less than +/-‐ 1°C varia4on for all seasonal scenarios. Comfort level is kept! Research > Results & Findings 17% 20% 23% 17 0 12 24 36 48 60 72 16 17 18 19 20 21 22 23 24 25 Time [hours] Temperature[degeesC] Indoor temperature change for a LT household participant Winter Spring Summer Autumn
- 18. 0 12 24 36 48 60 72 16 Time [hours] Tank temperature Varia4ons are within the boundaries for all scenarios. Comfort level is kept! Research > Results & Findings 18 Autumn Summer Spring Winter Tank temperature change for LT household participant Tanktemperature[degeesC] Time [hours] 0 12 24 36 48 60 72 40 50 60 70 80 90 100 110 120 130 Figure 12: 17% 20% 23%
- 19. BESS opera4on (winter scenario) No wind curtailment needed in this scenario Research > Results & Findings 17% 20% 23% 19 Max BESS capacity Wind curtailment BESS level Power[kW] Time [hours] 0 12 24 36 48 60 72 0 50 100 150 200 250 300 h BESS charges to account for prognosis errors not accounted by the DR par4cipants
- 20. Industry par4cipa4on The modeled DR ac4vity for Cementa significantly reduced cluster size! Research > Results & Findings 17% 20% 23% -‐700 +100 20 -‐700 DR dynamics changes when cluster size is reduced. This explains the increase on Saturday when Cementa is no longer par4cipa4ng
- 21. Research > Validity & Reliability? 21 -‐ Worst case condi4ons reflected -‐ Uniform household consump4on model -‐ DR par4cipant can not override the consump4on -‐ Network simulator not included in study -‐ Implementa4on difficul4es not considered -‐ Economical constraints not considered17% 20% 23%
- 22. Research > Benefits for SGG project 22 17% 20% 23% Opera4on Strategies Simula4on results Household modeling
- 23. Research > From Model to Reality? 23 17% 20% 23% Ongoing collabora4on with VENTYX on development & implementa4on
- 24. Smart Grid Gotland Thank you for your acen4on Daniel A. Brodén, +46 762185980 daniel.broden1@vacenfall.com 24