针对传统BP神经网络估算电池SOC过程中,存在初始权值和阈值对预测精度影响较大的问题,引入Tent混沌映射和自适应收敛因子对灰狼算法(GWO)进行改进,改善灰狼算法易陷入局部最优、后期迭代效率不高的缺点.将改进灰狼算法(improved grey Wolf algorithm,IGWO)与BP神经网络模型结合,得到BP神经网络最优初始权值和阈值,提高预测精度和收敛速度.对锂电池充放电实验数据预处理,得到样本数据.利用MATLAB进行仿真验证,结果表明,IGWO-BP神经网络算法的预测精度相较于传统BP神经网络算法、GWO-BP神经网络算法更优,基于改进灰狼优化BP神经网络估算电池SOC的方法的绝对误差能控制在1.53％以内,有效提高了预测精度和收敛速度.
SOC Estimation of Li Ion Battery Based on IGWO BP Neural Network
Aiming at the problem that the initial weight and threshold have a great influence on the predic-tion accuracy in the process of estimating battery SOC by traditional BP neural network,tent chaotic map and adaptive convergence factor are introduced to improve the gray wolf algorithm(GWO).It is easy to improve the gray wolf algorithm,the shortcomings of falling into local optimum and low efficiency of later iteration.Combining the improved grey wolf algorithm(IGWO)with the BP neural network model,the optimal initial weights and thresholds of the BP neural network are obtained,which improves the predic-tion accuracy and convergence speed.Preprocess the experimental data of lithium battery charge and dis-charge to obtain sample data.Using MATLAB for simulation verification,the results show that the pre-diction accuracy of the IGWO BP neural network algorithm is better than that of the traditional BP neural network algorithm and the GWO BP neural network algorithm.The absolute error of the method based on IGWO BP can be controlled within 1.53％,which effectively improves the prediction accuracy and conver-gence speed.
Li Ion batterySOCBP neural networkadaptive strategyimproved grey Wolf algorithm
万方数据
维普
