Search Results

Cell balancing is a key function of battery management system (BMS) that is implemented to maximize the battery's available capacity and service life. The increasing demand of larger and better performance pack has raised the need to investigate the various cell balancing techniques so that the energy of the battery can be fully realized. In this work we develop a phenomenological model in order to quantify the benefits of passive balancing and active balancing. The electrical response of a model pack consisting of serially connected lithium ion cells is simulated with Matlab. The effects of the variance of cell capacity, internal resistance, self-discharge rates, pack configuration and size are studied. The possible optimization rooms for implementing passive and active balancing are suggested.

Start-stop, aka engine-stop or idle-stop, technologies are increasingly being applied to automotive vehicles to increase fuel economy. Start-stop vehicles turn off the engine during periods of zero speed and/or during prolonged coast down. During engine-stop, the vehicle electronics are powered solely by the battery. To replenish the battery, the battery needs to be recharged. In typical ICE vehicles, the battery is continuously charged. However, fuel economies can be improved if strategic charging of the battery can be achieved through selective charging through the alternator or through regenerative braking. To optimize fuel economy, an accurate estimation of the battery state of charge (SOC) during vehicle operation is required. Although state of charge estimation has mainly focused on Li-ion batteries, lead-acid batteries may be used successfully in start-stop applications.

Traditional testing approaches for fundamental battery materials focus on highly artificial test profiles, for example constant current (CC) or constant voltage (CV) testing. Additionally, the currents used for capacity and cycle tests are often very low. These profiles are not indicative of the types of current/voltage profiles that the battery will experience during actual vehicle operation. As a result, these simple tests may fail to sufficiently elicit the reduction in performance and failure modes that occur during more dynamic cycling. In this paper, we outline an approach in which vehicle-level modeling is applied to regulatory drive cycles in order to derive power vs. time requirements for an energy storage system. These requirements are used to identify segments of the regulatory drive cycles that present significant challenges to the battery. Finally, the most stressing portions of the drive cycle are used to determine limiting physical characteristics of batteries.