Joint Estimation of SOH and RUL for Lithium-ion Batteries using CNN-GRU Based on Artificial Lemming Algorithm

This article innovatively proposes an innovative method for assessing the health status and predicting the lifespan of lithium batteries. This method deeply combines the advantages of convolutional neural networks (CNN) and gated recurrent units (GRU), fully utilizing CNN's outstanding ability in complex data feature extraction and GRU's unique advantage in deep analysis of time series data. In order to further improve the prediction accuracy of the model, the article also introduced the artificial lemming algorithm (ALA) to finely optimize the model parameters. In the data processing stage, we first added an appropriate amount of noise to the battery data to enhance the robustness of the model, and adopted advanced modal decomposition techniques to comprehensively and meticulously process the battery data. Subsequently, through repeated experimentation and careful adjustment, we successfully optimized the key parameters of the CNN-GRU model using ALA. A series of experimental results fully demonstrate that the ALA-CNN-GRU method proposed in this paper exhibits significant superior performance in predicting the state of health (SOH) and remaining useful life (RUL) of batteries when compared and tested with various industry-leading algorithms.