基于遗传算法的Elman神经网络在镍氢电池容量预测中的应用

Elman neural networks based on genetic algorithms and its application in prediction of MH-Ni battery capacity

为了准确地对镍氢电池荷电状态进行预测,在分析影响镍氢电池剩余容量的多种因素的基础上,综合国内外常用的预测镍氢电池的几种方法,采用ELman网络对镍氢电池容量预测建立模型并用遗传算法对其初始权值和阈值进行优化。仿真结果表明:该网络不仅局部泛化能力好,而且收敛速度快。证明该网络对MH-Ni电池剩余容量的预测是有效的。

For predicting the state of charge （SOC） of MH - Ni battery precisely, this paper adopts Elman neural network to predict the state of charge （SOC） of MH - Ni battery, to create model and to utilize GA to optimize its original weights and bias, analyzing several methods that people often use to predict the state of charge （SOC） of MH - Ni battery at present on the basis of analyzing many factors that affecting the MH - Ni battery residual capacity. The simulation results show that not only the local predicting capability is better but also is convergence rate quick, demonstrating the prediction the network has made to the battery residual capacity is effective.