A comparison of scenario generation methods for the participation of electric vehicles in electricity markets
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
A comparison of scenario generation methods for the participation of electric vehicles in electricity markets. / Jensen, Ida Græsted; Møller, Niels Framroze; Pantuso, Giovanni; Juul, Nina.
In: International Transactions on Electrical Energy Systems, Vol. 29, No. 4, e2782, 2019.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - A comparison of scenario generation methods for the participation of electric vehicles in electricity markets
AU - Jensen, Ida Græsted
AU - Møller, Niels Framroze
AU - Pantuso, Giovanni
AU - Juul, Nina
PY - 2019
Y1 - 2019
N2 - We consider the procurement of electricity for a large fleet of electric vehicles operating in electricity markets. Due to uncertain regulating prices, this problem has typically been modelled as a stochastic program. In this study, we address the issue of generating scenario trees, ie, simplified representations of the uncertainty necessary to solve the corresponding stochastic programs. A trade-off between accurate descriptions of the uncertainty and tractability of the stochastic program is sought. Based on data describing electric mobility and electricity prices in Denmark, general-purpose scenario generation strategies are tested and compared. Such strategies include state-of-the-art property matching methods and time-series analysis. The results show that the co-dependence between the regulating prices at different hours of the day plays a crucial role when generating scenario trees for these problems, making copulas an important property to consider. This information can help decision makers to achieve better (cheaper) electricity procurement by accurately preprocessing the uncertainty in the regulating prices.
AB - We consider the procurement of electricity for a large fleet of electric vehicles operating in electricity markets. Due to uncertain regulating prices, this problem has typically been modelled as a stochastic program. In this study, we address the issue of generating scenario trees, ie, simplified representations of the uncertainty necessary to solve the corresponding stochastic programs. A trade-off between accurate descriptions of the uncertainty and tractability of the stochastic program is sought. Based on data describing electric mobility and electricity prices in Denmark, general-purpose scenario generation strategies are tested and compared. Such strategies include state-of-the-art property matching methods and time-series analysis. The results show that the co-dependence between the regulating prices at different hours of the day plays a crucial role when generating scenario trees for these problems, making copulas an important property to consider. This information can help decision makers to achieve better (cheaper) electricity procurement by accurately preprocessing the uncertainty in the regulating prices.
KW - electric vehicles
KW - regulating market
KW - scenario generation
KW - stochastic programming
UR - http://www.scopus.com/inward/record.url?scp=85057719378&partnerID=8YFLogxK
U2 - 10.1002/etep.2782
DO - 10.1002/etep.2782
M3 - Journal article
AN - SCOPUS:85057719378
VL - 29
JO - International Transactions on Electrical Energy Systems
JF - International Transactions on Electrical Energy Systems
SN - 1430-144X
IS - 4
M1 - e2782
ER -
ID: 210195439