No abstract

The practice of hybridizing energy storage systems is vital to high ramp rate power applications, in which energy storage systems are constrained by strict power and energy requirements. Hybrid energy storage is typically studied in the electrical and thermal domains separately, but due to the inherent link between electrical and thermal energy domains, it is necessary to examine hybrid energy storage in both domains simultaneously. In this paper, a combined electro-thermal energy storage system is modeled and simulated. Equivalent circuit and lumped-parameter models are used to facilitate control design. PI controllers are designed for both the electrical and thermal domains to demonstrate the ability to perform multi-domain energy management.

No abstract

Hybrid energy storage systems are a popular alternative to traditional electrical energy storage mechanisms for electric vehicles. Consisting of multiple heterogeneous storage elements, these systems require thoughtful design and control techniques to ensure adequate electrical performance and minimal added weight. In this work, a graph-based design optimization framework is extended to facilitate design and control optimization of a battery-ultracapacitor hybrid energy storage system. For a given high ramp rate load profile, a hybrid electrical energy storage system consisting of battery and ultracapacitor packs with proportional-integral controllers is considered. A multi-objective optimization problem is formulated to simultaneously optimize sizing and performance of the system by minimizing mass and deviations from ideal controller performance. This optimization is achieved by adjusting the size of the energy storage system and parameters of the feedback controller. A Pareto curve is provided, which exhibits the tradeoffs between sizing and performance of the hybrid energy storage system. Dynamic simulation results demonstrate optimized designs outperform initial designs in both sizing and electrical performance objectives. The design and control optimization approach is shown to outperform a similar sizing optimization approach.