Online Battery Impedance Identification Based on Small AC Signal Injection of Inverter

Battery impedance online identification is a non-destructive measurement technique for battery condition estimation and health diagnosis. However, conventional methods are costly, time-consuming, and can only be performed on offline small cells. Spurred by this, we concentrate on the development of a transparent and novel measurement method that, for the first time, uses a controllable sine-sweep perturbation generated by existing battery-connected inverter. In this paper, through inverter control a small AC excitation signal with controllable amplitude and target frequency can be transferred to battery due to power flow. This method eliminates the need to disconnect the battery from the operating system or add additional excitation circuit, which provides a low-cost, high-availability and real-time means for online battery impedance identification. Once it is developed as an embedded function in widely-used battery-inverter system and performed online during operation continuously or periodically, it could provide a monitoring system for the whole pack as well as a single cell. Simulations and experiments prove its effectiveness. The online Electrochemical Impedance Spectrum (EIS) tests, conducted on a 24-Ah battery module test bench, were completed within 300 seconds for a frequency range from 0.1Hz to 10kHz and within 50 seconds for a frequency range from 1Hz to 1kHz, respectively, much faster than electrochemical workstation. The accuracy and dynamic performance of the proposed method are validated by contrasting online EIS results with offline results with cell voltage being 3.4V, 3.5V, 3.6V. Moreover, a series of dynamic impedance spectra across varying discharge rates and State of charge (SoC) highlights its promising potential for battery state estimation.