State-of-health estimation of Lithium-ion battery based on back-propagation neural network with adaptive hidden layer

Abstract The reliability and safety of lithium-ion batteries (LIBs) are key issues in battery applications. Accurate prediction the state-of-health (SOH) LIBs can reduce or even avoid battery-related accidents. In this paper, a new back-propagation neural network (BPNN) is proposed to predict SOH LIBs. BPNN uses as input LIB voltage, current temperature, well charging time, since it strongly correlated with SOH. number hidden layer nodes adaptively set based on training data order improve generalization capability BPNN. effectiveness robustness scheme verified using four distinct datasets different data. Experimental results show that able accurately LIBs, revealing superiority when compared other alternatives.