Battery health state estimation method based on data feature mining

The health status estimation of lithium-ion battery is a challenging through measurement. To establish connection between and data features, State Health (SOH) method based on feature mining proposed. Four features are extracted from the charging curve, grey correlation analysis used to determine high status. combines Backpropagation (BP) neural network with Genetic Algorithm (GA) for training learning, enabling SOH. feasibility proposed validated using NASA dataset. results show that SOH in this paper outperforms traditional BP achieving accurate estimation.