| An ever increasing acceptance of electric vehicles as passenger cars relies on better operation and control of large battery packs. The individual cells in large battery packs do not have identical characteristics and may degrade differently due to their manufacturing variability and other factors. It is beneficial to evaluate the performance gain by replacing certain battery modules/cells during actual driving. The following are the objectives of our research. We will develop an on-line battery module degradation diagnostic scheme using the intrinsic signals of a battery pack equalization circuit. Therefore, a battery "health map" can be constructed and updated in real time. Next based on the derived battery health map, the performance of the battery pack will be evaluated a user specified trip so as to evaluate the "worthiness of replacing" certain modules/cells. Different electric vehicles have different performance for the same driving cycle. These variations are due to variation in driving patterns, traffic, different light patterns, random behavior of the drivers etc. To account for this random behavior of the electric vehicle performance we generate 100 random trip cycles. We aim to model the behavior of the driving cycle and battery behavior. Finally, the thesis also explores the possibility of energy exchange between the battery packs and the smart grid. In the smart grid scenario where we have the knowledge of the electricity price and the load patterns on the grid, it is beneficial for the user to schedule charging and discharging patterns for electric vehicles. Our research will define charging and discharging patterns throughout the life of the battery. We will optimize the charging and discharging times and define the opportunity cost for each day during summer and winter months. The objective is to maximize the profit earned by selling excess energy in the battery to the grid and minimize the charging cost for the electric vehicle. |
