Electric vehicle charge planning using Economic Model Predictive Control

Rasmus Halvgaard, Niels K. Poulsen, Henrik Madsen, John Bagterp Jørgensen, Francesco Marra, Daniel Esteban Morales Bondy

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


Economic Model Predictive Control (MPC) is very well suited for controlling smart energy systems since electricity price and demand forecasts are easily integrated in the controller. Electric vehicles (EVs) are expected to play a large role in the future Smart Grid. They are expected to provide grid services, both for peak reduction and for ancillary services, by absorbing short term variations in the electricity production. In this paper the Economic MPC minimizes the cost of electricity consumption for a single EV. Simulations show savings of 50–60% of the electricity costs compared to uncontrolled charging from load shifting based on driving pattern predictions. The future energy system in Denmark will most likely be based on renewable energy sources e.g. wind and solar power. These green energy sources introduce stochastic fluctuations in the electricity production. Therefore, energy should be consumed as soon as it is produced to avoid the need for energy storage as this is expensive, limited and introduces efficiency losses. The Economic MPC for EVs described in this paper may contribute to facilitating transition to a fossil free energy system.
Original languageEnglish
Title of host publication2012 IEEE International Electric Vehicle Conference (IEVC)
Number of pages6
Publication date2012
ISBN (Print)9781467315623
Publication statusPublished - 2012
Event2012 IEEE International Electric Vehicle Conference (IEVC) - TD Convention Center, Greenville, SC, United States
Duration: 4 Mar 20128 Mar 2012


Conference2012 IEEE International Electric Vehicle Conference (IEVC)
LocationTD Convention Center
CountryUnited States
CityGreenville, SC


Dive into the research topics of 'Electric vehicle charge planning using Economic Model Predictive Control'. Together they form a unique fingerprint.

Cite this