Estimation of Critical Battery States via Strain and Stress Measurement

点击翻译按钮获取中文摘要

Energy storage utilizing lithium-ion batteries is ubiquitous from cell phones and satellites to vehicles and the electric grid. All these applications require reliable estimates of the available power and energy. Consumer concerns, such as range anxiety for Electric Vehicles (EVs), hamper market penetration. Such concerns can be alleviated by improving estimates of critical battery states such as capacity and internal resistance that change over time. This research seeks to address those concerns by improving battery health diagnostics. This advance will benefit all battery-powered devices and could be especially crucial for the second-life use of automotive batteries for energy storage on the electric grid which has the potential to enable more substantial penetration of renewable resources through expanded utilization of existing battery technology. Experimentally validated models of the battery mechanical response and its connection to battery health would fill a critical gap in the availability of public-domain data and models. The state of the art battery estimation methods relies on the cell terminal voltage measurements, and the plan is to improve the battery health diagnostics in this project by measuring and interpreting the battery cell expansion which happens when the electrode layers fill and empty with lithium-ions during charging and discharging. This research aims to create a multi-physics model and estimation techniques to harness the information in the deformation of the electrodes and to analyze the measured electrical and mechanical signals to enhance battery health estimation. The graphite electrode exhibits distinct expansion patterns as it fills with lithium-ions, which can enable diagnosis of electrode-specific degradation. Several fundamental gaps in the modeling of multi-scale inter-dependent thermal, electrochemical and mechanical responses of the battery need to be addressed before the benefits of this approach can be realized in relevant usage cycles involving dynamic charging and discharging. Moreover, the accuracy of joint state and parameter estimation depends on the operating region and the current excitation or usage profile. Systematic techniques to assess the identifiability would benefit from physics-based models that can be used to investigate the influence of various degradation mechanisms, cross-sensitivity with thermal swelling, sensor location, and commercially relevant packaging.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.