Large scale renewable energy projects, as well as small installments on houses and businesses, have the ability to transform the composition of energy sources that fuel society. While such sources are beneficial in reducing greenhouse gas emissions that result from power production, they provide new obstacles for the use and storage of energy. As renewable energy sources continue to produce larger quantities of energy, the electricity grid will have to adapt to this decentralized and unpredictable production. Ridder et al. (2013) examined the feasibility of charging electric vehicles (EVs) in a smart grid scenario and their ability to compensate for shortages and surpluses that occur in the energy market due to unpredictable energy supply. Utilizing behavioral information about the EV users, EV capacity, and the capacities of charging stations, the authors created a model to simulate EV potential in Flanders, Belgium. The coordination algorithm that was determined to fit the population resulted in the proper distribution of EVs amongst available charging stations so that capacity was maximized. In the second scenario, the authors demonstrated the ability of EV users to take advantage of varying prices of energy by strategically charging when excess day-ahead energy was available and selling energy back to the grid when demand exhausted supply and only day-of, or imbalance market energy was available. Ultimately, the study demonstrated that as imbalance market prices increased compared to day-ahead prices, flexible chargers were able to save money and charger/dischargers were able to make money by discharging energy at high market prices.—Stephanie Oehler
Ridder, F., D’Hulst, R., Knapen, L., Janssens, D., 2013. Applying an activity based model to explore the potential of electrical vehicles in the smart grid. Procedia Computer Science,847—853
In order to determine the role of EVs in conjunction with variable tariffs in matching energy consumption to supply, Fjo D. Ridder and colleagues utilized algorithms and activity based models to simulate the behavior of the population of Flanders, Belgium. The scheduling algorithm utilized in this study takes into consideration EV owner behavior, capacity and characteristics of current EV batteries, and availability of smart grid technologies. The FEATHERS system, an activity based model specific to Flanders, created a daily schedule for each subject within the simulation. The system created data for the 6 million people in the artificial population that reflected the census data for the area, incorporated land-use data and applied it to 2368 traffic analysis zones (TAZ), combinations of times and distances for different routes based on type of transport, and decision-making tendencies of individuals based on their corresponding socio-economic data. The FEATHERS model was able to mirror the decision-making processes of the population, and thus could be used to create day-ahead demand schedules. Using data that represented the city of Flanders, the authors constructed a simulation under the assumptions that there would be 200 EVs spread out over 56 parking areas, equal electricity tariffs determined by the Belgian Power Exchange, and equal maximum charging powers, battery capacities, initial charges, average consumption while driving, and driving speeds for all EVs. Finally, a utility function was attributed to each EV to calculate the charging cost and adjust for lost power. Many of the same assumptions used in the activity-based model were applied to the utility functions.
When the first application was performed, multiple iterations resulted in a balancing of the distribution of EVs charging at each location based on the total amount of power available daily. This demonstrated that EV owners would quickly adjust to the supply of power at charging stations and adjust their locations accordingly. The second application sought to minimize costs for electricity retailers. In this simulation, retailers purchased power using day-ahead models to predict the demand. Since electricity prices that consumers paid fluctuated throughout the day based on demand, the prices would be lowest when there was less actual demand than the retailer had anticipated. Another scenario could occur if the retailer underestimated the electricity demand and some had to be purchased that day. If there was a plethora of power available, these prices would be more competitive than day-ahead prices. The opposite scenarios were also possible and would result in higher prices.
The authors simulate three consumption situations in the second application; the no-strategy benchmark users, charging only users, and users with the ability to charge and discharge. The results demonstrated that EV users were sensitive to price and would adjust their usage in order to minimize costs. In the second case, users determined when they charged their vehicles and they saved 60%, on average. In the final case, users were able to charge when the price was the lowest as well as discharge when the price was the highest. On average, this scenario resulted in profits of 128 Euro for each user, each year.
While there are not enough EVs in use currently to play a large role in the electricity grid, they do have the potential to bridge the gap between the shortages of supply or demand of electricity which are often brought about suddenly by renewable energy sources. Charging vehicles when surplus energy is available prevents it from going to waste by providing a form of storage, and discharging power during shortages reduces the need to suddenly generate more electricity. And by giving the user control over the demand schedule, EVs do not create the privacy issues that other demand side management methods do by relying on an information processing controller. By providing a consumer-driven method for storing electricity, EVs have the potential to flatten the price curve for electricity throughout the day and more evenly distribute demand in conjunction with supply.