Scenario analysis of data centres in the Irish energy system
1. Data centres in Ireland
SCENARIO ANALYSIS FOR ETSAP WORKSHOP
Leonardo Collina, MSc student
Università di Bologna / University College Cork
2. Data centres overview
Physical facilities that companies use to host their critical applications and data.
Demand increased rapidly in recent years and will further increase stimulated by new data-intensive
technologies such as artificial intelligence, machine learning.
Why is it relevant for
the energy system?
Electricity consumption for:
• IT equipment (servers, storage,
transmission)
• Cooling system
• Others
Hyperscale DCs can have a demand
capacity >100 MW
But… waste heat recovery is possible!
Why is it relevant for Ireland?
Factors that make it perfect for hosting DCs:
• Weather conditions
• Good fibre connections with US
• Attractive corporation tax
Factors that make DCs hard to handle:
• Big growth expected in the next decade
• Concentration in Dublin area
• Isolated energy system (lack of grid
connections with other countries)
• Decarbonization targets
ETSAP workshop, 30.11.’21
3. Ireland decarbonization pathways
A33E61:
• 33% reductions from agriculture,
• 61% reductions from energy & industry
by 2030
ETSAP workshop, 30.11.’21
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
kton
CO2
“Climate Action & Low Carbon Development (Amendment) Act 2021” targets to reduce greenhouse-gases [GHGs] by:
• 51% by 2030,
• net-zero by 2050.
https://tim-carbon-budgets-2021.netlify.app/scenarios
https://zenodo.org/record/5517363#.YZ-uVtDP02w
More mitigation scenarios developed by
the MaREI Energy Policy and Modelling Group
available:
4. DCs in TIM
ETSAP workshop, 30.11.’21
Season Weekly DayNite
Wi NW A (0-7)
Sp WD B (7-9)
Su C (9-17)
Au D (17-19)
E (19-0)
40 time-slices
5. DCs in TIM
Cooling system electricity and waste heat are expressed as share of the total consumption of
the process.
Shares are taken from a thermodynamic model on hourly level, then aggregated on time-slices
level in TIM.
ETSAP workshop, 30.11.’21
11. New capacity installed
• Investments in gas power plants in 2024 due to coal phase out in 2025
• Each year a considerable level of new wind off-shore capacity
• Some investment in CCS technologies in 2030
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
E61-1_Flat
E61-2_LF
E61-3_Cap
E61-4_LFCap
2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
GW
PWR-WIN-ON
PWR-WIN-OF
PWR-SOL
PWR-HYD
PWR-GAS-CCS
PWR-GAS
PWR-BIO-CCS
12. New capacity installed (Up to 2030)
ETSAP workshop, 30.11.’21
• In the next 8 years the system will need from 16 to 17.8 GW of new capacity to meet decarbonization targets
• This 1.8 GW difference is covered mostly from wind off-shore
• But… in the case of very high demand the further difference is covered by new gas power plants
1.85 1.97 2.06 2.37
3.02 3.02 3.02 3.05
6.33
7.49 7.56 7.56
4.01
4.01 4.01 4.01
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
GW
Total new capacity up to 2030
PWR-WIN-ON
PWR-WIN-OF
PWR-SOL
PWR-HYD
PWR-GAS-CCS
PWR-GAS
PWR-BIO-CCS
-0.20
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
GW
Demand projections comparison
PWR-WIN-ON
PWR-WIN-OF
PWR-SOL
PWR-HYD
PWR-GAS-CCS
PWR-GAS
PWR-BIO-CCS
13. Waste heat recovery
ETSAP workshop, 30.11.’21
0
10
20
30
40
50
60
2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
PJ
Heat recovery potential
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
0%
5%
10%
15%
20%
25%
30%
0
10
20
30
40
50
60
2022 2023 2024 2025 2026 2027 2028 2029 2030
PJ
Heat to DH Network
E61-4_LFCap E61-3_Cap E61-2_LF E61-1_Flat
% % % %
From 2027, in 3° and 4° demand senarios, new investments in DH networks are no longer cost optimal
While the potential grows the share of heat actually used decrease
14. What if no heat recovery?
0
5
10
15
20
25
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
PJ
Heat usage by sector (2022-2030)
FT-RSDHET FT-SRVHET
1.79
10.72
16.25
18.09
-30.00
-20.00
-10.00
0.00
10.00
20.00
30.00
40.00
E61-1_Flat-NoHeatRec E61-2_LF-NoHeatRec E61-3_Cap-NoHeatRec E61-4_LFCap-NoHeatRec
PJ
Service fuel demand for SC, SH, WH (2022-2030):
Difference between heat recovery/no heat recovery
SRVAHT SRVAHT2 SRVBGS SRVBIO SRVELC SRVGAS SRVHET SRVOIL SRVSOL
SRV sector
• The heat recovered from DCs is used for space and water
heating in service and residential sectors.
• Heat is replaced mostly by gas and to a small extent by oil.
In the 4th case there’s also a big contribution of biomass.
• The amount of gas avoided by recovering the waste heat is
lower than the amount of additional gas required in the power
sector due to DCs growth (18.09 PJ vs 53.90 PJ in the 4th case)
15. Emissions by sector
ETSAP workshop, 30.11.’21
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
2019 2029 2030
CO2
ton
ENV-CO2_AGR ENV-CO2_IND ENV-CO2_PWR ENV-CO2_RSD ENV-CO2_SRV ENV-CO2_TRA
• What happens in 2029 to the industrial
sector?
• What happens in 2030 to the residential
sector?
Emissions from the power sector increase
as the demand from DCs grows
more pressure on the other sectors to
speed up the transition
New investments in
greener technologies?
16. CO2 Price
ETSAP workshop, 30.11.’21
0
0.5
1
1.5
2
2.5
2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
kEur/ton
CO2 price
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
CO2_price = 2 kEur/ton
No cheaper options
to mitigate emissions than
Direct Air Capture
«backstop» technology used to
make the scenario feasible
0
500
1000
1500
2000
2500
3000
3500
4000
2028 2029 2028 2029 2028 2029 2030 2027 2028 2029 2030
E61-1_Flat E61-2_LF E61-3_Cap E61-4_LFCap
CO2
tons
Backstop emissions
AGRCO2N INDCO2N INDCO2P RSDCO2
17. Discussions
ETSAP workshop, 30.11.’21
Assumptions / input data:
• Cooling system
• Waste heat
• Load factor increase
Analysis:
• Impact to other sectors
• Comparison with less challenging scenarios
Lack of reliable data:
Literature review and
sensititvity analysis
New ideas are welcome
18. THANK YOU FOR LISTENING
leonardo.collina4@studio.unibo.it