基于改进鹭鹰优化算法的锂电池RUL预测研究OA
Research on lithium battery Remaining Useful Life prediction based on an improved secretary bird optimization algorithm
针对锂电池容量退化固有的非线性与不稳定性,导致锂电池RUL(Remaining Useful Life,RUL)预测精度难以提升的问题,本文提出一种基于ISBOA-LSTM的锂电池RUL预测模型.首先,通过均值插补的方法对数据中的异常值进行了调整.然后,引用Fuch映射和柯西突变策略优化鹭鹰算法(Secretary Bird Optimization Algorithm,SBOA),利用改进后的ISBOA(Improved Secretary Bird Optimization Algorithm)优化LSTM网络的参数配置.最后基于NASA公开数据集,对ISBOA-LSTM模型进行对比实验,结果表明,所提模型在锂电池RUL预测方面表现出更高的预测精度,模型的均方根误差以及平均绝对误差最低可达0.0085 和 0.0064.
Aiming at the inherent nonlinearity and instability of lithium battery capacity degradation,which leads to the problem of difficult to improve the prediction accuracy of lithium battery RUL(Remaining Useful Life,RUL).In this paper,a lithium battery RUL prediction model based on ISBOA-LSTM is proposed.First,the outliers in the data are adjusted by the method of mean interpolation.Then,the Fuch mapping and Cauchy mutation strategies are invoked to optimize the Secretary Bird Optimization Algorithm(SBOA),and the improved Secretary Bird Optimization Algorithm(ISBOA)is utilized to optimize the parameter configuration of the LSTM network.Finally,the ISBOA-LSTM model is validated based on the NASA public dataset,and the results show that the proposed model exhibits higher prediction accuracy in lithium battery RUL prediction,and the root-mean-square error as well as the mean absolute error of the model can be as low as 0.0085 and 0.0064.
宁佳伟;曾进辉;曾冠林;黄梓涵
湖南工业大学交通与电气工程学院,湖南 株洲 412007湖南工业大学交通与电气工程学院,湖南 株洲 412007湖南工业大学交通与电气工程学院,湖南 株洲 412007湖南工业大学交通与电气工程学院,湖南 株洲 412007
信息技术与安全科学
鹭鹰优化算法剩余使用寿命锂电池
Secretary bird optimization algorithmremaining useful lifelithium-ion battery
《船电技术》 2026 (1)
8-13,18,7
国家自然科学基金项目,"多微电网互联电能变换器及其控制方法研究(52377185)".
评论