Intelligent Charging Algorithm for Electric Vehicles
Abstract: Electric vehicles play an important role in creating a fossil free transport sector. Making the vehicles efficient involves many new areas outside the manufacturing process, such as chargers, power grids and electricity markets. This thesis models the charging of electric vehicles using a Markov Decision Process and uses Reinforcement Learning solution models to derive an intelligent charging algorithm. This algorithm can take concepts such as electricity price, battery degradation and electrical losses into account in order to minimise the overall operational costs, and add more value to the use of electric vehicles. Models of how voltage varies in a battery is used and data on causes of battery degradation are derived from modern papers within battery technology. The intelligent charging algorithm is compared to baseline charging algorithms, one of which correspond to how charging is regularly performed today. Vehicle-to Grid is a promising future technology where electric vehicles can discharge some of their energy back to the grid in order to alleviate the stress of a power grid constrained by increasing demand as well an increasing penetration of intermittent sustainable sources of electricity such as wind and solar. Simulations are performed over scenarios with different electricity prices and the implications of being able to utilise Vehicle-to-Grid is studied. Results from simulations show that the intelligent charging algorithm effectively can reduce costs by approximately 30% on average compared to regular charging when the charging sessions last for 7 hours. Vehicle-to-Grid was seen to only be able to reduce costs in simulations with inexpensive batteries on days when there was a large difference in electricity price. The intelligent charging was able to save as much as 500 SEK for long charging sessions with expensive batteries, and powerful chargers. Results show a promising future for an intelligent charging algorithm to be used in order to improve the efficiency of electric vehicle charging.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)