This title appears in the Scientific Report :
2022
Please use the identifier:
http://dx.doi.org/10.1007/s11590-021-01826-w in citations.
Please use the identifier: http://hdl.handle.net/2128/31169 in citations.
Budget-cut: introduction to a budget based cutting-plane algorithm for capacity expansion models
Budget-cut: introduction to a budget based cutting-plane algorithm for capacity expansion models
We present an algorithm to solve capacity extension problems that frequently occur in energy system optimization models. Such models describe a system where certain components can be installed to reduce future costs and achieve carbon reduction goals; however, the choice of these components requires...
Saved in:
Personal Name(s): | Singh, Bismark (Corresponding author) |
---|---|
Rehberg, Oliver / Groß, Theresa / Hoffmann, Maximilian / Kotzur, Leander / Stolten, Detlef | |
Contributing Institute: |
Technoökonomische Systemanalyse; IEK-3 |
Published in: | Optimization letters, 16 (2022) S. 1373–1391 |
Imprint: |
Heidelberg
Springer
2022
|
DOI: |
10.1007/s11590-021-01826-w |
Document Type: |
Journal Article |
Research Program: |
Societally Feasible Transformation Pathways Effective System Transformation Pathways |
Link: |
OpenAccess |
Publikationsportal JuSER |
Please use the identifier: http://hdl.handle.net/2128/31169 in citations.
We present an algorithm to solve capacity extension problems that frequently occur in energy system optimization models. Such models describe a system where certain components can be installed to reduce future costs and achieve carbon reduction goals; however, the choice of these components requires the solution of a computationally expensive combinatorial problem. In our proposed algorithm, we solve a sequence of linear programs that serve to tighten a budget—the maximum amount we are willing to spend towards reducing overall costs. Our proposal finds application in the general setting where optional investment decisions provide an enhanced portfolio over the original setting that maintains feasibility. We present computational results on two model classes, and demonstrate computational savings up to 96% on certain instances. |