Market based intelligent charging system for electric vehicles

Abstract

The existing electrical infrastructure is very unlikely to expand overnight. Therefore, a smart solution is certainly needed to integrate the additional load which electric vehicles (EV) bring to the network. The aim of the thesis is to study the electricity market, different intelligences related to electric vehicle charging and establish an algorithm that produces an optimized charging schedule for electric vehicles. The algorithm ensures a cost profit for user and takes part in demand response by shifting the timing of charging loads based on energy prices. The core intelligences integrated to the EV charging system in the thesis are cost optimization, peak shaving and load shifting. The algorithm follows the hourly unit cost related to the energy consumption and distribution fee in order to find the cheapest time slot for charging operation. It allocates as high charging power as possible to the cheapest time slots and then start selecting the expensive time slots until the battery is charged to desired state of charge. Along this process, the algorithm continuously calculates the maximum charging power available after other house-hold usage. The Elspot area price of Finland for 2018 added with 0.3 cents/kWh margin and 24% VAT are used as energy prices. Distribution unit prices include time-of-use pricing for day and nighttime energy use in addition to the fixed fuse-based fee. By following these unit prices, the algorithm shifts the load from high demand to low demand hours in order to minimize the total costs. The algorithm is implemented in MATLAB and tested through a case study on different type of Finnish detached houses. Detached houses with different load profile data are used as input for charging a 75 kWh EV with a 10 kW and 7.5 kW charger in different cases, where the other inputs remain same for all the test cases. The Elspot area price of Finland for 2018 added with 0.3 cents/kWh margin and 24% VAT are used as energy prices. Different day and night-time distribution prices are applied depending on the consumption. The simulation results are compared to regular EV charging, where the charging operation starts right after the EV is plugged in and finishes charging within shortest time. The results from the simulation are investigated from user’s and grid’s point of view. From user’s perspective, all the charging events with intelligent charging have costs savings over regular charging. The monetary profit is higher for higher charger rating (10 kW). In cases where the household usage is low, the proportional profit is high. From grid point of view, over 99% of the load gets shifted to night-time for 10 kW charger cases. For the 7.5kW charger, the amount of shifted load is over 97%, which is a little lower than 10 kW charger cases because of longer charging time. The findings of the case study validate the use of smart charging algorithm in order to ensure cost savings for the user.