A novel Fireworks Factor and Improved Elite Strategy based on Back Propagation Neural Networks for state-of-charge estimation of lithium-ion batteries

The state of charge (SOC) Lithium-ion battery is one the key parameters management system. In SOC estimation algorithm, Back Propagation (BP) neural network algorithm easy to converge local optimal solution, which leads problem low accuracy based on BP network. It proposed that Fireworks Elite Genetic Algorithm (FEG-BP) used optimize network, can not only solve traditional fall into maximum solution but also limitation algorithm. searchability improved has been significantly enhanced, and error become smaller propagation speed faster. Combining experimental data charging discharging, FEG-BP compared with genetic (GA-BP), results are analyzed. show standard predicts within 7%, while reduces 3%.