A Cauchy perturbation cuckoo search particle filtering algorithm for remaining useful life prediction of lithium-ion battery considering capacity regeneration

The remaining useful life prediction of the lithium-ion battery can evaluate the battery's reliability, identify the occurrence of faults, and reduce battery risks. In this study, the particle filtering algorithm is employed along with the introduction of the Cauchy perturbation factor to enhance the local optimization capability of the cuckoo search algorithm. The improved cuckoo search optimization algorithm is utilized to transfer particles from prior distribution areas to the maximum likelihood area, resulting in the development of the Cauchy perturbation cuckoo search particle filtering algorithm. To achieve high-precision prediction of the remaining useful life, an attenuation model is established, taking into account the complex capacity regeneration phenomenon that occurs after a long period of resting of the lithium-ion battery storage. This model includes three types of capacity: underlying, recovery, and surface capacity. The experimental results, based on an aging dataset from the University of Maryland, demonstrate that the suggested algorithm provides clear advantages over widely used particle filtering and unscented particle filtering algorithms in terms of estimation accuracy. Additionally, it exhibits a fast convergence rate and low resampling rate, further enhancing its benefits.