<div class="csl-bib-body">
<div class="csl-entry">Theis, F. (2026). <i>Quantifying the effect of forecast uncertainty on the techno-economic performance of large-scale prosumers</i> [Diploma Thesis, Technische Universität Wien]. reposiTUm. https://doi.org/10.34726/hss.2026.76862</div>
</div>
-
dc.identifier.uri
https://doi.org/10.34726/hss.2026.76862
-
dc.identifier.uri
http://hdl.handle.net/20.500.12708/232132
-
dc.description.abstract
The increasing penetration of distributed, weather-dependent generation is transforming prosumers into active participants in electricity markets, yet the operational scheduling of these portfolios must contend with short-term uncertainty in renewable generation, demand, and market prices. Despite this, many influential energy system studies continue to rely on the simplifying assumption of perfect foresight, which risks systematically overestimating achievable technical and economic performance. This thesis quantifies the magnitude of this overestimation by comparing three operational strategies - perfect foresight, imperfect foresight with rule-based control, and imperfect foresight with an optimized operation strategy - applied to a real-world, large-scale prosumer asset in Lower Austria: an electric bus fleet operator with a ground-mounted photovoltaic system and a battery electric storage system participating in the day-ahead market. Using the open-source optimization framework IESopt, each method is evaluated over a full year of operation at 15-minute resolution, explicitly accounting for time-varying day-ahead and imbalance settlement prices alongside grid tariff structures. The results show that perfect foresight substantially overstates performance: the benchmark achieves a total financial result more than ten times higher than rule-based control and roughly twice that of the optimized strategy, while grid consumption peaks and battery cycling are likewise underestimated by 50–75% and around 12%, respectively. A sensitivity analysis further reveals that forecast accuracy and assumptions about initial grid consumption peaks affect the two imperfect-foresight strategies very differently, with the optimization-based approach consistently extracting more value from improved information than the rule-based controller. These findings indicate that perfect foresight assumptions are unsuitable for investment appraisal and infrastructure planning involving large-scale prosumer flexibility, and that the choice of dispatch method - not just forecast quality - materially determines the value and grid impact realized in practice. The thesis provides a standardized, open-source framework for assessing prosumer performance under realistic uncertainty, applicable across market and portfolio configurations.
en
dc.language
English
-
dc.language.iso
en
-
dc.rights.uri
http://rightsstatements.org/vocab/InC/1.0/
-
dc.subject
Prosumer
en
dc.subject
Uncertainty
en
dc.subject
Optimal Scheduling
en
dc.subject
Imperfect Foresight
en
dc.subject
Flexibility
en
dc.title
Quantifying the effect of forecast uncertainty on the techno-economic performance of large-scale prosumers
en
dc.type
Thesis
en
dc.type
Hochschulschrift
de
dc.rights.license
In Copyright
en
dc.rights.license
Urheberrechtsschutz
de
dc.identifier.doi
10.34726/hss.2026.76862
-
dc.contributor.affiliation
TU Wien, Österreich
-
dc.rights.holder
Franziska Theis
-
dc.publisher.place
Wien
-
tuw.version
vor
-
tuw.thesisinformation
Technische Universität Wien
-
dc.contributor.assistant
Zwickl-Bernhard, Sebastian
-
tuw.publication.orgunit
E370 - Institut für Energiesysteme und Elektrische Antriebe