On-board impedance diagnostics method of Li-ion traction batteries using pseudo-random binary sequences. Method evaluation and feasibility study of concept.

Abstract

This thesis deals with the on-board impedance measurements of Li-ion batteries on hybrid electric vehicles/electric vehicles by using pseudo-random binary sequences (PRBSs). The impedance of the battery can be related to its state of charge but the accurate impedance measurements are difficult to perform in the vehicles. By using an excitation signal like PRBS, it is possible to extract the impedance information of the battery packs. Both experiments and simulations are performed with different set-ups to verify the PRBS method. A non-parametric method is used to process the data and extract the impedance measurement. Experiments in the laboratory at different SOC levels and temperatures are made to validate the PRBS method. In the simulations, the noise sensitivity is analyzed. It is shown that the PRBS method can produce a valid electrochemical impedance spectrum in a limited frequency range, similar to the result from a high accuracy laboratory impedance analyzer. The method is stable at different SOC levels and temperatures. However, the battery impedance at high frequency is difficult to obtain with the PRBS method in the experiments. A simulation of the excitation signal in the vehicle is performed where the electric motor is used as the load. It shows that it is possible to some extent to use the drive line in a hybrid electric vehicle/electric vehicle to perform an on-board battery impedance measurement.