UC San Diego

In order to meet emerging requirements for better utilization of electrical energy, some novel modeling and control approaches for battery systems and electric grids are proposed and validated in this dissertation.

To better utilize a single battery system, the energy delivery capability and available energy stored in the battery needs to be understood first. A battery modeling approach is proposed to characterize power delivery dynamics, given charge and discharge demand as an input, and also estimate the state-of-charge of a battery, not only in normal operating range, but also in extreme cases, such as battery over-charging. The model is composed of separated voltage and current models. Several non-linear models, including Hammerstein model, open-circuit voltage characteristics, and Takacs hysteresis model, are combined in the voltage and the current model, respectively. The state-of-charge of the battery is estimated in a recursive optimization fashion. The parameterization and estimation methods of the model are described and validated on experimental data from a lithium iron phosphate cell.

Several individual battery systems are usually connected in parallel to expand the total capacity of a network. To coordinate the output of each battery system, three current scheduling strategies are proposed. Besides simultaneous and sequential discharge scheduling algorithms, a hybrid algorithm is formulated by solving a Quadratic Programming problem. The simulation results indicate the feasibility of the proposed scheduling algorithms and motivate the use of parallel connected battery modules despite changes in battery operating parameters. The simultaneous and sequential discharge scheduling algorithms are extended to power scheduling. A complete modular battery system for an experimental Electric Vehicle with the same topology is developed for future experimental validation and research.

Integrating inverters in battery systems or other DC sources is required when connecting to electric girds. To maintain the stability of the grid, disturbance rejection control aiming to mitigate fluctuations in AC power flow is studied based on an experimental setup created to mimic a local electric grid. Through demodulating real power oscillations, modeling of actuator and disturbance, and implementing controller designed by combining the internal model principle and optimal control, the feasibility of proposed control method is validated.