Analysis of a safe utilization algorithm for retired power batteries from new energy vehicles

Abstract

The graded utilization of waste batteries has gained research significance due to recent reports of new energy vehicle lithium-ion batteries exploding whilst awaiting recycling or in end-of-life storage. In this study, we innovatively selected battery performance parameters such as the internal resistance, charge and discharge rate, and current maximum available capacity to evaluate the safety of retired power batteries from the perspective of inducing thermal runaway. A fractional calculus theory was then introduced, and the fractional second-order resistance as well as a capacitance model and an adaptive genetic algorithm were established for the identification of the parameters. An improved dual-scale filtering algorithm was generated, which combined the extended Kalman filter algorithm and the unscented Kalman filter algorithm to improve the accuracy of the parameter estimation. The final test outcomes indicated that the equivalent circuit model optimized by incorporating multiple filtering algorithms had error rates of 1.87 %, 1.65 %, and 1.27 % for the state of charge of the battery in three different operating condition testbeds, with average errors of 0.62 %, 0.69 %, and 0.59 %, respectively. When an initial experimental platform was constructed for the detection of the parameters, the voltage error quickly stabilized to within 0.03 V. It also displayed many advantages of data detection and calculation, such as faster convergence, faster tracking, and the highest result accuracy when compared with the battery model using other algorithms. This experiment highlighted that a fractional second-order resistive–capacitive circuit equivalent battery state detection model incorporating various filtering algorithms has practicality and feasibility.