Beta 1


Title Optimal system architecture for implementation of Electrical Vehicles on Bornholm
Author Münster-Swendsen, Janus
Supervisor Elmegaard, Brian (Thermal Energy Systems, Department of Mechanical Engineering, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Xu, Zhao (Electric Energy Systems, Department of Electrical Engineering, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Institution Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
Thesis level Master's thesis
Year 2010
Abstract Purpose This report is a study of the integration of Electric Vehicles (EV's) in the power system of the island Bornholm. Three different infrastructures has been investigated: distributed, fast charging and battery swapping, all with a 15% share of EV's. Approach The characteristics of the three infrastructures are found by constructing programs in matlab and GAMS to optimize the charge pattern according to price variations and then simulating different scenarios. The EV fleet is regarded as one aggregated battery. The infrastructures are evaluated on their ability to provide peak power and system services, the charging costs and the overall capability of facilitating a higher share of wind power in the power system. Sensitivity analyses are performed to evaluate impact of assumptions. Results The distributed infrastructure can reduce peak loads with 15% with the assumptions used, and about 90% of the total demand for regulating power can be covered by the EV's. The ability to limit power overflow is a factor 10 better than the battery swapping infrastructure and a factor 30 better than the fast charging infrastructure. The fast charging infrastructure causes an increase in peak demand. Given that a battery is installed at the station, about 5% of the regulating power can be supplied. The battery swapping infrastructure has very limited peak having capacity and about 32% of regulating power can be supplied. Conclusion The simulations show that the distributed infrastructure is superior to the other two infrastructures in all aspects. The charging costs are lower, more peak power can be provided, more system services can be provided and it is better suited for facilitating more wind power in the system. The assumption of an aggregated EV fleet is compared to individual simulations and proves to be acceptable.
Abstract Formål Denne rapport er en undersgelse af integrationen af elbiler i elsystemet på Bornholm. Tre forskellige infrastrukturer er blevet undersgt: distribueret, hurtigladende og batteri ombyttende, alle med en andel på 15% elbiler. Fremgangsmåde Kendetegnene for de tre infrastrukturer er fundet ved at opbygge programmer i matlab og GAMS for at optimere ladningsmønstret efter prisvariationer og derefter simulere forskellige scenarier. Elbilerne er anset som et sammen- lagt batteri. Infrastrukturerne evalueres på deres evne til at levere spidslast og systemydelser, ladningsomkostningerne og den totale evne til at muliggøre højere andele af vindenergi i elsystemet. Der er udført sensitivitets analyser for at evaluere antagelsernes indflydelse. Resultater Den distribuerede infrastruktur kan reducere spidsbelastningerne med 15% med de gjorte antagelser, og omkring 90% af den totale efterspørgsel af regulerkraft kan dækkes af elbilerne. Evnen til at begrænse eloverløb er en faktor 10 bedre end for batteriombytningsinfrastrukturen og en faktor 30 bedre end den hurtigladende infrastruktur. Den hurtigtladende infrastruktur forårsager en forøgelse af spidsbelastningen. Hvis der er installeret et batteri ved stationen, kan omkring 5% af regulerkraften leveres. Den batteriombyttende infrastruktur har begrænset evne til at nedsætte spidslasten og omkring 32% af regulerkraften kan leveres. Konklusion Simuleringerne viser at den distribuerede infrastruktur er de to andre overlegne i alle aspekter. Ladningsomkostningerne er lavere, der kan leveres mere el i spidsbelastninger, der kan leveres flere systemydelser og den er bedre rustet til at muliggøre højere andele af vindkraft. Antagelsen om de sammenlagte elbiler er sammenlignet med individuelle simuleringer og fundet til at være acceptabel.
Imprint Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Pages 140
Fulltext
Original PDF MASTER.pdf (5.36 MB)
Admin Creation date: 2010-02-25    Update date: 2010-10-28    Source: dtu    ID: 258774    Original MXD