Analysis of Large Scale Integration of Electric Vehicles in Nord-Trøndelag

Abstract

In recent years, the shift in attitude towards climate and CO2 emissions has accelerated the sale of electric and hybrid electric vehicles in Norway. Predictions indicate that Norway may surpass 200,000 chargeable vehicles by 2020, which corresponds to seven percent of the total vehicle fleet. This number includes both electric vehicles and hybrid electric vehicles. To explore the impact a large scale electric vehicle adoption will have on the power grid, simulations of an existing low voltage power system have been conducted. The load flow simulation tool Simpow was used for this purpose, and Nord-Trøndelag Elektrisitetsverk provided information on the grid structure and consumer consumption data. From the supplied data, February 2 was chosen for the 24-hour simulation period. This day has the highest energy consumption, and therefore represents the ?worst case? scenario. A hypothetically built wind turbine close to the residential areas was integrated in the system, using wind measurement data from a wind farm in Nord- Trøndelag. Different scenarios were explored, investigating how sensitive the grid is to additional load under different assumptions, and how the wind generation can contribute to a more self-supporting power system. Symmetrical and asymmetrical distribution of electric vehicle charging loads in relation to physical locations have been compared, and the results suggest that one cannot give an exact number of vehicles that the system can handle. The system capacity when operating with dumb charging strategies is varying depending on where the vehicles are situated physically. With many electric vehicles located close together, the given voltage level constraints of the model were violated with a seven percent electric vehicle penetration share. However, assuming that vehicles are more spread out physically, the system restrictions were not violated for a electric vehicle share of 20 percent. In other words, the placements of the additional loads are equally decisive for the system voltage variations as the number of loads. By applying smart charging strategies, the voltage fluctuations in the system during a day are mitigated. For the 20 percent EV penetration scenarios, given the assumptions presented in this thesis, the added load does not seem to put more stress on the system than it can handle. However, for the 50 percent EV penetration scenarios, the charging load might present the system with too much stress, even with smart charging strategies. Other measures will have to be taken if the power system ever experiences an EV share that high. A long term simulation was performed to verify the results obtained from the 24-hour simulations. It verified that February 2 can be assumed to be the ?worst case? scenario, that is, the lowest voltage levels throughout the year was observed on that day. It also gave an indication on how well the wind turbine is suited to relieve the system of increased consumption due to electric vehicle charging. If the wind generation is assumed to cover the additional load created by the electric vehicles, the need for imported power in the system will not increase. Wind generation during February 2 is higher than the electric vehicles consumption if 20 percent share is assumed. This relation, however, is not representative for the generation throughout the year. Wind generation is unpredictable, and generally higher during winter. Installing an energy storage system makes the wind energy more controllable. Still, days with little or no wind generation will inflict the need of a huge capacity storage system to cover the charging loads at all times. Assuming a lower electric vehicle adoption share, and not requiring the wind generation to cover charging loads at all times, the needed storage system capacity could be realizable.