Modeling and Management of Energy Storage Systems

As society transitions to relying more heavily on renewable energy, the need for energy storage rises considerably, as storage facilitates power regulation between these sources and the grid. Battery packs are a popular choice for this, with lithium-ion batteries leading the market. These packs typically consist of multiple individual cells connected in series and parallel to meet the energy storage requirements. However, due to manufacturing tolerances and environmental conditions, variations in each cell’s parameters are inherently present in these packs, which can propagate state imbalances such as state of charge (SOC) and temperature. This is detrimental to pack operation, as cell-to-cell variations can limit total useable energy and increase degradation of the pack. An even further concern is safety, due to an increased chance of problems like thermal runaway when temperature differences are left unattended. Current work in the literature studies these parameter discrepancies by analyzing their effects or estimating the imbalances, but there are scarce efforts towards combining these tenets of addressing parameter mismatch. There are also numerous studies which propose charge balancing methods, though limited works explore analytical relationships between parameter and state heterogeneity during balancing. The challenges persist in temperature balancing, as current battery thermal management systems (BTMS) are typically limited in architecture and restricted in ability to attend to temperature gradients. Furthermore, in control scenarios, studies do not analyze how well conditioned battery assemblies are for control. This work focuses on understanding, estimating, and mitigating the state imbalances propagated by cell-to-cell parameter variations, as well as the impact they have on cell controllability. This is first done by (a) presenting a modeling framework conducive to both analysis and estimation which considers all types of battery state and parameter heterogeneity. This allows for (b) investigation of battery dynamics due to unequal parameters through analytical expressions, as well as the development of an online state estimator with over a 60% reduction in computational cost. After parameterization of 66 lithium-ion cells, the framework is used to determine contributions of multiple types of parameter heterogeneity on output imbalances. The proposed estimator is then validated experimentally, showing how the fewer required calculations benefits estimation runtime. The heterogeneity modeling framework is further used to (c) develop a modified SOC balancing strategy to equalize cells with different capacities. By studying the analytical stability and state imbalance, simulation results indicate that capacity heterogeneity propagates SOC imbalance while the pack is operating with a nonzero average current, something which is offset by the proposed controller. Next, (d) a control strategy for removing temperature differences is developed and enabled through a novel BTMS design which permits individual control of cell cooling. Analysis and simulation confirm that cell-level temperature control proves to be better equipped for reducing temperature differences and extending pack life as compared to pack-level control, even more so as total input power increases. Finally, (e) a nonlinear controllability analysis is performed for the experimentally parameterized cells. The condition number of each cell’s model is determined both with new and aged parameters, and a sensitivity analysis shows that the cells’ conditioning is equally impacted by all parameters. The results of this analysis are then used to determine the best conditioned assemblies for a batch of cells with a mix of new and second-life parameters. The insights gained from these contributions offer a deeper understanding of how batteries interact in a pack, as well as options to ensure these interactions are beneficial to overall operation.