Modelling and Identification of Lithium-Ion Battery Using Relay Feedback Response

In this paper, attempts have been made to contribute a continuous-time identification approach for modelling and estimation of lithium-ion battery parameters. Considering a first-order RC equivalent circuit model (ECM) with a signal propagation delay for battery dynamics representation, a closed loop relay feedback test is to be an optimal choice for ascertaining the stable responses. Using a frequency domain analysis, an explicit set of analytical expressions for each of the unknowns in the considered system model is derived by using the important output measurements and relay settings. As the practical output is quite sensitive to measurement noise owing to the presence of erroneous current and voltage sensors, a noise-free waveform of the output is obtained by using a curve fitting method based on a Fourier series, and a point of relay switching instant is varied by an appropriate choice of hysteresis width. The proposed identification algorithm is validated by considering a benchmark example from literature, and its efficacy is demonstrated through an integral of the absolute error criterion and the step response plots.