ISCHESS – Integration of stochastic renewables in the Swiss electricity supply system
Abstract
The project integration of stochastic renewables in the Swiss electricity supply system (ISCHESS) addresses the problem to be dealt with by network operators due to the integration of large scale stochastic renewable energy sources (RES) in the Swiss electricity systems during the coming decades. Located at the center of the ENTSO-E transmission grid, the Swiss electricity network also forms a benchmark case for the European energy transition. The contributions required for the planning of the future electricity supply system are highly interdisciplinary. This project brings together expertise from economics, technology assessment, life cycle analysis, network security and optimisation. It follows a bottom up approach, βirst studying the short-term distribution grid aspects; and extending the methods both in temporal and spatial scale to a comprehensive study at the national level. The βirst part of the analysis concerns the distribution grid aspects of the future electricity supply. To facilitate the technical solution approaches, a broad review of different strategies for the integration of intermittent RES is performed with respect to their environmental and cost aspects. New inventory data have been established for several battery as well as hydrogen generation technologies. Environmental burdens and potential impacts are quantified using Life Cycle Assessment (LCA) methodology. A time trajectory generator of grid components and a software tool for the operation and planning of electricity grids have been developed and are used through- out the project. The benchmark case is the integration of solar photovoltaic (PV) power in a distribution grid from central Switzerland using grid extension, curtailment, reactive power control, storage units and demand side management. The following βindings and conclusion have been made in the first project part:
1. Storage is only economical for higher electricity prices or lower storage costs than today. For low demand scenarios, storage units are only economical for extremely low battery cost. For high demand scenarios with little or no demand side management, storage units can become an economic approach that reduce the curtailment of the solar PV supply.
2. Grid upgrade evaluation depends on the calculation costs of the grid operator. The tool determines the potential gain from the reduced operating costs after a system upgrade.
3. Economically, curtailment is in almost all scenarios reasonable to some extent. The controllability of the PV components models the curtailment of available PV injections that would otherwise overload parts of the network. Costs result from the opportunity costs of not injecting the available PV energy.
4. A strong reduction of the system operating costs can be reached if parts of the nominal load demands can be shifted in time by the distribution system operator (DSO), even if just using 10 % of the daily energy demand.
5. Uncertainties related to some of the available LCA, in particular network expansion and some battery technologies are high. The LCA could only be performed on the electricity grid/supply component level, but not on an integrated system level due to time constraints.
The second modelling framework used in this study was the Swiss TIMES energy system model (STEM), which covers Swiss energy system from resource supply to end uses over a long-term horizon. A distinguished feature of the model is its hourly intra-annual time resolution for three typical days (working day, Saturday and Sunday) in four (summer, autumn, winter and spring) seasons. Most importantly, within this project, the electricity sector in STEM has been enhanced by
1. representation of different grid voltage levels with a spatial representation of 15 aggregated
nodes;
2. inclusion of new/emerging electricity storage technologies of various sizes for the different
grid levels;
3. introduction of variability in wind and solar PV based electricity supply; and
4. representation of secondary (and primary) control reserve provision markets.
In this context, this study contributes with its methodological advancements to the introduction of RES variability as well as grid topology in long term energy systems model. Incorporating grid infrastructure in energy system models provides significant benefits because RES integration can be modelled more realistically, including grid congestion and price effects. A range of ’what-if’ type scenarios was assessed along three main dimensions (namely 1: future energy policy and demand, 2: location of new gas power plants, and 3: electricity network expansion and availability of batteries) to evaluate strategies for integrating stochastic RES in the Swiss electricity and heating sectors. Across the selected scenarios, electricity demands continue to increase and reach over 70 TWh by 2050. At the supply side, up to 3 GW of gas power plants are installed by 2050 to replace the existing nuclear power plants. At the same time, supply from variable renewables sources increases and contributes up to 24 TWh by 2050, under stringent climate policy. This high uptake of variable RES requires pumped hydro storage of about 5.6 TWh (3.3 GW) and batteries of 3.5 TWh (5.3 GW) by 2050. The need for electricity storage increases almost linearly with the deployment of wind and solar PV up to about 14 TWh. However, beyond this threshold of wind and solar PV based electricity generation, an accelerated deployment of storage is inevitable. In this context, batteries offer distributed (localized) balancing solutions, with their deployment potential depending on the grid level to which they are connected. The up- take of battery storage is driven by solar PV (at low voltage levels), and wind and CHP (at medium voltage levels). At the same time, large-scale batteries complement pump-hydro at high voltage levels when the latter is only partially available due to water resource restrictions or participate in other markets (e.g. balancing markets or international trade). In 2050, up to about 13% of the summer electricity production from wind and solar PV is stored for consumption in autumn and winter seasons. In conditions where stringent climate policy and restriction of grid enforcement are applied, power-to-gas technologies represent one option for seasonal energy storage driven by differences in seasonal electricity production costs. On the demand side, dispatchable loads such as water heaters and heat pumps, contribute in easing the electricity peak by shifting 10-25% of the electricity demand on a daily basis.
The analysis indicates that if there were to be no further grid expansion other than planned for 2025, grid congestions, the operation of certain network elements close to the loading limit, would become a major bottleneck to cope with increasing electricity demands. For example, under stringent climate change mitigation policy congestion could occur up to 7000 hours in the year 2050. Importantly, congestion affects both the electricity supply and demands. At the sup- ply side, it could lead to the deployment of non-cost effective options in some grid nodes, such as geothermal for base load electricity, if full dispatchability of some more cost effective options in other grid nodes, such as large gas power plants, cannot be achieved due to congestion. It can also hinder the penetration of renewable electricity. On the demand side, grid congestion limits the electrification and retains fossil-based heating supply, compared to a reversed trend when the grid is expanded.
In fact, when grid infrastructure is to be reinforced, the overall net economic benefits for the Swiss electricity and heat system outweigh the costs of expansion. In this case, congestion levels reduce to less than 3000 hours (43% lower than the no grid expansion scenario). The savings in the whole electricity and heat supply system of Switzerland are in the range of 0.5 – 3 billion CHF per year over the period of 2020 – 2050 depending on the scenario. These cost savings result from changes in the electricity supply side (35% of the total cost savings on average), reduced imported electricity and fossil fuels (38%) and structural changes in the heat supply (27%).
Both electricity storage and grid expansion are necessary to realise the full potential of variable renewable energy sources. When both storage and grid expansion are excluded, then 20-50% less solar PV and wind are deployed, but the contribution from gas based generation in- creases up to 45%. The latter results in higher CO2 emissions and consequently, incurs additional climate change mitigation costs up to 6 billion CHF per year (or +17%) on average over the period 2020 – 2050, compared to the case where both options are enabled. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000395791Publication status
publishedPublisher
ETH ZurichSubject
Distribution Grids; Integration of Photovoltaic Energy; Multi-period AC-OPF; Nonlinear Model Predictive ControlOrganisational unit
02279 - Forschungsstelle Energienetze-ETH Zürich / Research Center for Energy Networks
Notes
Final project reportMore
Show all metadata
ETH Bibliography
yes
Altmetrics