e-Highway 2050: Methodology for 2050 scenario quantification

Challenge: How to quantify demand level, generation and storage installed capacities as well as exchanges at the level of each geographical cluster and for each of the considered qualitative scenario capturing “extreme but realistic” features of power system evolutions until the 2050 time horizon?

Background and assumptions 
The scenarios have been described following in  "eHighway2050: Challenging energy scenarios for the pan European transmission system by 2050". They are fully consistent with similar scenario-type approaches developed in related research activities, in particular with the scenarios considered by the EC Energy Roadmap 2050. As shown in Figure 1 the five scenarios can be characterized according to three dimensions: Level of energy demand; Share of RES in the energy mix;  Share of centralized/decentralized generation. 

Figure 1 : Overview of the five “extreme but realistic” e-Highway2050 scenarios

The objective of the scenario building approach is to go from a European description of the scenario, mostly qualitatively, to a quantification of demand, storage, exchange and generation at country and cluster level.
The calculation and location of the demand and of the installed capacities for each generation technology over a geographical zone is a complex problem.  The scenario development needs indeed:
- to consider a sufficient spatial level of details in order to take into account distributed phenomena such as the demand and the RES,
- to consider a sufficient temporal level of details in order to take into account inter-temporal correlation of renewable generation,
- to ensure adequacy at any time (in fact at each hour).
Regarding the political issues, a methodology and a process are defined to ensure consistency between the national policies and the scenarios at European level.

Description of the result: the methodology
A three-step top-down approach has been developed in that perspective. First, the generation capacities are defined on a “macro area” level (a macro area contains several countries (1)). Then, the installed capacities of the macro-areas are broken down to country level, where thirty-three countries are considered. Finally, the installed capacities are distributed across each of the ninety-six clusters into which Europe is split to.

Figure 2: Three steps for the quantification: from macro area level to country level and then to cluster level

Step 1
The whole process starts with the computation of yearly demand values and of energy targets per generation technology at European level. Next, the installed capacities are firstly defined for each macro-area based on average capacity factors and distribution keys (2) which combine information about the potential of generation capacities and demand in a given macro-area.
Then, a system adequacy simulation over Europe, without considering grid constraints, allows modifying the installed capacities of generation and storage to reach a sufficient level of adequacy. Finally, the generation/demand ratio of each macro area is assessed and some generation can be redistributed to ensure an acceptable level of independence between macro areas (this level depends on the scenario).
In Step 1, no country or national considerations were taken into account and only the perspectives of whole Europe are considered.  A similar approach is then repeated at lower levels (countries and clusters), where the National Renewable Energy Action Plans at 2020 (NREAP) and other local constraints are also taken into account.

Step 2
Priority is given to national policies and trends are considered while splitting the installed capacities of a macro area among its countries.
The installed capacities of each macro-area are broken down to country level, where thirty-three European countries are considered.   
The weighting distribution keys used to go from macro level to country level are the combination of information about potential of generation capacities, demand and policies, and trends of each country. A particular attention is given to the National Renewable Energy Action Plan for 2020, providing the RES target for each European country. These plans set the minimal values to be reached in each scenario for 2050.
Then, as in the previous step, the system simulation is performed for whole Europe, without considering grid constraints, in order to provide the installed capacities and storage for a sufficient level of adequacy, and to improve the imbalances of the countries according to the scenarios.

Step 3
Finally the installed capacity generation values are distributed across each country.  The weighting distribution keys used combine information about potential of generation capacities, demand and local constraints such as cities, mountains, natural areas. 

The diagram below details the three steps of the scenario quantification methodology. 

Figure 3: The e-Highway2050 three-step top-down approach for scenario quantification: the different phases

Assessment of the methodology use: case of the “Big and market” scenario
The implementation of the above methodology is illustrated for one of the five e-Highway2050 scenarios: the “Big and market” scenario.

In this scenario, a global agreement for climate mitigation is achieved. Thus, CO2 prices are high due to the existence of a global carbon market. Europe is fully committed to meet its 80-95% GHG reduction orientation by 2050; but it relies mainly on a market based strategy. Moreover, there is a special interest for large scale centralized solutions, especially for RES deployment and storage. Public attitude towards deployment of RES technologies is indifferent in the EU, while acceptance  of nuclear and shale gas, as energy sources, is positive since being preferred to decentralized local solutions. The CCS (Carbon Capture and Storage) technology is also assumed to be mature in this scenario. Electrification of transport, heating and industry is considered to occur mainly at centralized (large scale) level. Only a minor shift towards ‘greener’ behaviours is experienced in this scenario compared to present practices. Therefore, the efficiency level is low. In general, the public is somehow passive, and the players are active in a market-driven energy system.

The annual electricity demand is 4387 TWh in this scenario, which, compared to other scenarios, is at an intermediate level (due to low level of new uses combined to low efficiency level). The generation mix is diversified: it includes nuclear, fossil with and without CCS and 60% RES. Among renewables, wind has the most important share.

Medium level of international exchanges and rather balanced macro-areas (± 30%) are expected in this scenario. Moreover, the share of committable units is expected to be high enough (about 50%) to address security of supply issues.


Figure 4: Volume of annual demand for the five e-Highway2050 scenarios


Figure 5: Targeted energy mix of the “Big and market” scenario

These input data are used to deploy the three successive steps for quantifying such scenario.

Step 1: Computation of installed generation capacities in each macro-area
Installed capacities are first distributed between the macro areas using distribution keys and average capacity factors so that the annual generation targets of the scenario are respected. This first set of installed capacities leads to an average of 2700 hours of loss of load per year when running hourly simulations. To limit the loss of load duration (3) to an acceptable limit (4), it is necessary to increase generation capacities, especially fossil thermal capacity which is more than doubled. Moreover, exports in North Europe are too high whereas Central Europe is importing too much compared to the maximal imbalance accepted in this scenario (± 30%). Thus, some generation is moved from North to Central Europe.

The two graphs below illustrate the intermediate results coming from step 1: energy mixes with loads and imbalances at macro-area level for the “Big and market” scenario (Figure 6) and installed capacities per technology at macro-area level for the same scenario, today (2012), NREAP 2020 and the potential at 2050 (Figure 7).

Figure 6: Energy mixes and imbalances per macro-area (interim results), “Big and market” scenario

Figure 7: Installed capacities per technology at macro-area level (interim results), “Big and market” scenario

At macro-area level, the results for the “Big and market” scenario respect the European energy mix (except for minor deviations). The analysis shows that Central Europe, Southern Europe and South Western Europe are the macro-areas with the highest imbalances (highlighted in orange and red in Figure 6) while Nordic Europe is a net electricity exporter (highlighted in green in Figure 6).
In terms of installed capacities per technology (Figure 7), fossil thermal capacity is decreased by 40% compared to today, while nuclear capacity should remain unchanged. PV and wind capacity must be more than doubled compared to NREAP 2020 targets.

Step 2: Distribution of installed capacities at country level
At this step, the installed capacities are split between the countries of each macro area. In this scenario, an issue appears for PV in Germany: the use of distribution keys leads to an installed capacity of 21 GW which is less than the NREAP target for 2020 (51 GW). Installed capacity of PV in Germany is thus increased up to 51 GW to follow national policies. No additional adequacy issues are identified when running hourly simulations. However, some imbalances are too extreme for this scenario and thus some installed capacities have to be redistributed within the countries of some macro areas.
Some further adjustments are also made thanks to the feedbacks from TSOs and stakeholders.

The two graphs below illustrate the final results of step 2: energy mixes with loads and imbalances at country level for scenario “Big and market” (Figure 8) and installed capacities per technology at country level for the same scenario (Figure 9).

Figure 8: Energy mix and imbalances per country, “Big and market” scenario

Figure 9: Installed capacities at country level, “Big and market” scenario (GW)

As shown in  Figure 8, countries with a high wind potential and/or no nuclear phase-out tend to be energy exporters, namely Ireland, United Kingdom, France, Czech Republic, Slovenia, Romania, Bulgaria, Lithuania and Sweden. While Italy, Germany, Belgium, the Netherlands and Switzerland are the countries with the most critical imbalances in the “Big and market scenario”(5)

In terms of installed capacities (Figure 9), RES generation predominates in the Iberian Peninsula (mainly wind and PV), while Western European countries (France, United Kingdom, Ireland, Belgium and the Netherlands) mostly include wind and nuclear power, the availability of nuclear power being the main reason for the energy surplus in this macro-area. Wind, PV and thermal power are prevailing in Central European countries (Czech Republic, Slovenia, Austria, Switzerland, Germany and Denmark), as well as Eastern European countries; while installed capacities in Northern European countries mainly consist of hydro and wind generation.

Step 3: Distribution of installed capacities at cluster level
Finally, countries installed capacities are distributed among clusters taking into account local potentials and feedbacks from TSOs and stakeholders. The figure below illustrates the final results of step 3 for Germany.


Figure 10: Installed capacities at cluster level: example of Germany, “Big and market” scenario (GW)

A final sanity check is then performed to ensure consistency with European and National policies
In the “Big and market” scenario, CCS-technology plays an important role to meet the CO2-emission targets. The annual emissions of CO2 resulting from hourly simulations is 43 Mt which is slightly below the European Commission target (around 70Mt) but 270 Mt of CO2 need to be stored annually. In this scenario, apart from 297 GW of thermal plants, 113 GW are gas with CCS, 25 GW hard coal with CCS and 13 GW lignite with CCS.

NREAPs 2020 are not enough to meet the high RES shares. Wind installed capacity is high, with 500 GW including 76 GW in North Sea. It requires a lot of investment compared to 2020 targets (installed capacity should be more than doubled). Although PV generation plays a less significant role (275 GW), the need for investment is also high: the capacity is more than twice the 2020 targets.

In this scenario, the European installed capacity of nuclear is similar to the current level. However, installed capacities have to be increased in the countries which tolerate nuclear to mitigate the phase out in Germany, Belgium and Switzerland.

References
[1]     G. Sanchis, RTE et alia, “A methodology for the development of the pan-European Electricity Highways System for 2050”, CIGRE Paris, August 2014  
[2]     B. H. Bakken, M. Paun, R. Pestana, G. Sanchis, “e-Highway2050: A Modular Development Plan on Pan- European Electricity Highways System for 2050”, Cigre Lisbon, April 2013
[3]    e-highway project 
[4]     Thomas Anderski, Amprion et alia, “ Consultation Package: Methodology for Scenario Quantification”, 22 April 2014
[5]     Thomas Anderski, Amprion et alia, “Consultation : results of Scenario Quantification” , 22 April 2014
[6]     G. Sanchis, B. Betraoui, T. Anderski, “Electricity Highways Stakeholder Platform” e-Highways 2050 presentation, 16 April 2014
[7]     G. Sanchis, B. Betraoui, N. Grisey, T. Anderski, Y. Surmann, M. Paun, A. Vaféas, T. Pagano, A Top-Down Scenario Quantification Methodology For Electricity Highways at 2050, 14th  IAEE European Energy Conference on “Sustainable Energy Policy and Strategies for Europe”, Rome, 28-31 October 2014

Notes
(1) Six macro-areas are considered. For instance, the North Europe macro-area consists of Norway, Sweden and Finland. Additionally two areas for North-Sea and North Africa are defined to consider centralized production of RES.
(2) The weights of the distribution keys are scenario dependent. For example, in the scenario S1 (large scale RES), the weight for the Wind potential is more important than for the demand. In the scenario S5 (small and local), it is the opposite.
(3) The number of hours during the year in which the demand can’t be satisfied completely. This means that at least 1 MW of load is lost
(4) 3 hours of LOLD were assumed as the threshold, for which the systems reliability was considered as sufficient
(5) for countries bordering North Sea, wind offshore generation is attached to the North Sea cluster