基于ECTLBO-ELM模型的荷电状态估计

State of charge estimation based on ECTLBO-ELM model

电池的荷电状态(SOC)表示电池的可用容量,是电池管理系统重要参数之一。以锂离子电池为例,准确的估计可以提高其性能。为了建立锂离子电池的精确计算模型,提出了一种基于增强混沌教与学优化算法(ECTLBO)优化极限学习机(ELM)的SOC估计模型(ECTLBO-ELM)。在ECTLBO-ELM模型中,一是利用增强混沌优化策略对班级中最优个体进行混沌搜索以增强TLBO算法的全局优化性能;二是采用改进的TLBO算法优化ELM的输入权值和隐含层偏差,提高其估计性能。利用某10AH的锰酸锂电池的三种不同倍率下的放电实验数据集对提出的算法进行测试,为了揭示所提方法的性能,将结果与标准ELM算法进行比较。结果表明,该方法能较好地估计SOC。

The state of charge(SOC)of a battery indicates its usable capacity,which is one of the important parameters of the battery management system.In the case of lithium-ion batteries,accurate estimation can improve their performance.In order to establish an accurate computing model for Li-ion battery,a SOC estimate model is proposed based on enhanced chaos teaching-learning based optimization optimized extreme learning machine,named ECTLBO-ELM.This study is mainly concentrated on two aspects.One is to use enhanced chaos optimization strategy to search the optimal individual in the class to enhance the global optimization performance of TLBO algorithm.The other is to use the improved TLBO algorithm to optimize the input weights and hidden layer biases of ELM to improve the estimation performance.The discharge data sets of a 10AH lithium manganate battery with three different rations are used to test the proposed algorithm.To reveal the performance of the proposed method,the results are compared with ELM methods.The results show that the method can estimate SOC better.