Health state estimation of lithium ion batteries based on electrochemical impedance spectroscopy

Abstract The estimation method based on the electrochemical impedance spectrum can comprehensively describe reaction process of Li-ion battery without causing damage to battery, and is suitable for estimating state health battery. In this paper, equivalent circuit model established by analyzing ac internal corresponding each component composite link with curve segments in spectrum, parameters are identified offline help Nyquist diagram spectrum. It found that charge transfer resistance low-frequency amplitude increase monotonically aging Then, temperature compensation coefficient was introduced solve influence factors SOH lithium batteries, And root mean squared error before after reduced from 31.7 3.0, sum squares Error 13054.7 19.8, which significantly improved batteries different temperatures; a BP neural network constructed, correlation prediction 0.994. Solving production differences advantages disadvantages two methods summarized.