Abstract
This thesis is a collection of research articles exploring Modeling to Generate Alternative methods and energy system modeling, focusing on improving decision-making under uncertainty. Energy system models, an important tool for understanding the possibilities of future energy supplies, are subject to large uncertainties originating from, among others, an incomplete representation of reality and uncertain future developments of prices, weather, and demands. Addressing model imperfections and uncertainties is essential to improving the insights generated from modeling efforts. Studying alternative model solutions with costs deviating slightly from the cost-optimal can aid the decision-making process by providing solution ranges likely to contain desirable configurations.
In this project, a range of methods capable of exploring near-optimal solutions of energy system optimization models, building on the theory from Modeling to Generate Alternatives, have been developed. The methods have been applied to study several aspects of the transition of the European energy system toward carbon neutrality. In addition to the work on Modeling to Generate Alternative methods, a study investigating the implications of reduced natural gas imports from Russia on the transition of the European energy supply has been conducted.
Finding the optimal technology mix is fundamental in energy system planning. However, studying only the cost-optimal technology mix does not provide a full picture, as the mix might show considerable flexibility given only small increases in total system costs. All near-optimal technology mixes of a highly decarbonized European energy supply are studied using the developed Modeling All Alternatives method. An explicit definition of the near-optimal solution space is obtained using the convexity of the space, allowing fast sampling of all near-optimal technology mixes. Results reveal a large spread in the potential technology capacities exposing strong correlations between certain technologies.
Achieving EU emission targets requires actions by individual nations. Sharing the emission reduction effort fairly among EU nations is no easy task, as "fair" can be interpreted in many ways. 30.000 near-optimal effort-sharing configurations achieving a common EU emission target are explored using a sampling-based implementation of the Modeling All Alternatives method. Results reveal how some nations will experience high CO$_2$ abatement costs even at small emission reductions while others find it cost-optimal to reduce their electricity production. In addition, we find that large nations' emission reduction levels can severely impact neighboring nations' abatement costs.
Capturing the entire complexity of all near-optimal solutions of an energy system optimization model is no easy task. In the manuscript "Bounding the near-optimal solution space of energy system optimization models," I explore a bounding technique to improve the Modeling All Alternatives method. Testing the methodology on several benchmark cases and a full energy system optimization model reveals promising results.
In response to the events unfolding in Ukraine in the spring of 2022, a study was conducted investigating the medium- to long-term effects of a reduced natural gas import from Russia on the transition of the European energy supply. Using the full sector-coupled energy system model PyPSA-Eur-Sec, a scenario where natural gas availability from Russia was no longer available was investigated. Results show that the effects of the reduced natural gas supply are mitigated within a few years if Europe conforms to a climate target of 1.5C, as the transitioning to renewable energy alleviates the need for natural gas. The effects are, however, more severe under a 2C climate target, where the transition of the energy supply is forced to replace natural gas generators with alternatives such as renewable energy or coal power plants. The energy sector also largely transitions from using gas burners to heat pumps as a heat source.
In this project, a range of methods capable of exploring near-optimal solutions of energy system optimization models, building on the theory from Modeling to Generate Alternatives, have been developed. The methods have been applied to study several aspects of the transition of the European energy system toward carbon neutrality. In addition to the work on Modeling to Generate Alternative methods, a study investigating the implications of reduced natural gas imports from Russia on the transition of the European energy supply has been conducted.
Finding the optimal technology mix is fundamental in energy system planning. However, studying only the cost-optimal technology mix does not provide a full picture, as the mix might show considerable flexibility given only small increases in total system costs. All near-optimal technology mixes of a highly decarbonized European energy supply are studied using the developed Modeling All Alternatives method. An explicit definition of the near-optimal solution space is obtained using the convexity of the space, allowing fast sampling of all near-optimal technology mixes. Results reveal a large spread in the potential technology capacities exposing strong correlations between certain technologies.
Achieving EU emission targets requires actions by individual nations. Sharing the emission reduction effort fairly among EU nations is no easy task, as "fair" can be interpreted in many ways. 30.000 near-optimal effort-sharing configurations achieving a common EU emission target are explored using a sampling-based implementation of the Modeling All Alternatives method. Results reveal how some nations will experience high CO$_2$ abatement costs even at small emission reductions while others find it cost-optimal to reduce their electricity production. In addition, we find that large nations' emission reduction levels can severely impact neighboring nations' abatement costs.
Capturing the entire complexity of all near-optimal solutions of an energy system optimization model is no easy task. In the manuscript "Bounding the near-optimal solution space of energy system optimization models," I explore a bounding technique to improve the Modeling All Alternatives method. Testing the methodology on several benchmark cases and a full energy system optimization model reveals promising results.
In response to the events unfolding in Ukraine in the spring of 2022, a study was conducted investigating the medium- to long-term effects of a reduced natural gas import from Russia on the transition of the European energy supply. Using the full sector-coupled energy system model PyPSA-Eur-Sec, a scenario where natural gas availability from Russia was no longer available was investigated. Results show that the effects of the reduced natural gas supply are mitigated within a few years if Europe conforms to a climate target of 1.5C, as the transitioning to renewable energy alleviates the need for natural gas. The effects are, however, more severe under a 2C climate target, where the transition of the energy supply is forced to replace natural gas generators with alternatives such as renewable energy or coal power plants. The energy sector also largely transitions from using gas burners to heat pumps as a heat source.
Translated title of the contribution | Metoder til udforskning af det nær-optimale løsningsrum af makro energi systems modeller |
---|---|
Original language | English |
Publisher | Århus Universitet |
---|---|
Publication status | Published - Oct 2023 |