Precise prediction of open circuit voltage of lithium ion batteries in a short time period

Measuring the open circuit voltage (OCV) of a battery is quite time-consuming due to the relaxation process after the battery enters the open-circuit state. In this study, without the need of the complete voltage relaxation information, an available on-the-fly computing approach is proposed, by which the OCV at each state of charge (SOC) can be obtained from a simple first-order RC circuit model in a short period. The time constant of the RC model is evaluated by the ratio between the first and second derivatives of the experimental relaxation voltage at the earliest moment as soon as a specified voltage criterion on the real relaxation process is satisfied. With this criterion, the estimated OCV error is found to be extremely small. Experiments under three different temperatures and three types of battery state of health (SOH) are performed to verify the accuracy and robustness of the proposed approach. It is found that comparing to the traditional method of the OCV prediction at each SOC, the proposed approach can reduce the time of measuring the OCV to a few minutes (in most cases less than 6 min) with a reliable accuracy (in most cases less than 3 mV).