摘要
为解决锂离子电池荷电状态(state of charge,SOC)难以精确计算的难题,提出一种增强的混合蛙跳算法(mutation opposition shuffled frog-leaping algorithm,MOSFLA)优化快速学习网(fast learning network,FLN)的SOC预测模型。在混合蛙跳算法中引入几何中心变异策略和反学习策略增强算法的全局优化性能;为改善FLN的预测性能,采用MOSFLA优化FLN模型参数并建立MOSFLN-FLN模型;利用该模型对电池SOC进行预测,并将预测结果与其他模型预测结果相比较。结果显示,MOSFLA-FLN绝对误差不超过2.71,预测精度高,为SOC的精确计算提供了一种有效方法。
In order to solve the problem that the state of charge(SOC)of Li-ion batteries is difficult to calculate accurately,we propose a SOC prediction model based on enhanced shuffled frog leaping algorithm(MOSFLA)and fast learning network(FLN).Geometric center mutation strategy and opposition-based learning strategy were introduced into shuffled frog leaping algorithm to enhance the global optimization performance of the algorithm.In order to improve the prediction performance of FLN,MOSFLA was used to optimize the parameters of FLN model and MOSFLN-FLN model was established.Finally,we used this model to predict the SOC of batteries,and compared the predicted results with other models.The results show that the absolute error of MOSFLA-FLN is less than 2.71,and the prediction accuracy is high,which provides an effective method for the accurate calculation of SOC.
作者
周勇
Zhou Yong(College of Software,Chongqing Institute of Engineering,Chongqing 401320,China)
出处
《计算机应用与软件》
北大核心
2020年第4期329-333,共5页
Computer Applications and Software
关键词
荷电状态
快速学习网
混合蛙跳算法
几何变异
预测
State of charge
Fast learning network
Shuffled frog leaping algorithm
Geometric variation
Prediction