SOC prediction based on evolutionary network

Based on biological immune theory, a new immune algorithm is presented. Compared with the classical evolutionary programming and evolutionary algorithms with chaotic mutations, experimental results show that the proposed algorithm, parallel chaos immune evolutionary programming, is of high efficiency and can effectively prevent premature convergence. A three-layer feed-forward neural network is designed to predict the state of charge (SOC) of Ni-MH batteries. Initially, partial least square regression is used to select input variables. Then, five variables, battery terminal voltage, voltage derivative, voltage second derivative, discharge current and battery temperature, are selected as the inputs of NN. In order to overcome the weakness of BP algorithm, the proposed algorithm is adopted to train weights. Finally, under the state of dynamic power cycle, the estimated SOC from NN model and the measured SOC from experiments are compared, and the results conform that the proposed approach can provide an accurate estimation of the SOC.