基于小波变换的VFFRLS锂电池参数辨识

Lithium-ion battery parameter identification based on VFFRLS with wavelet transform

锂离子电池管理系统的可靠性通常是建立在高精度的电池模型参数辨识的基础上,因其在实际运行与测量环境中往往会受到噪声干扰,使采样信号中存在噪声信号,从而导致常规辨识算法下的参数辨识精度受到影响。因此,本文以锂离子电池二阶RC模型为研究对象,分析了遗忘因子更新过程中噪声对常用的可变遗忘因子递推最小二乘法(VFFRLS)的影响,进而提出一种基于小波变换的VFFRLS的参数辨识算法。利用小波变换的多尺度多分辨率特性对采样信号进行分频处理,再采用VFFRLS进行参数辨识,解决了因噪声影响遗忘因子带来的跟踪性能不佳的问题。实验结果表明,本文提出的算法具有更高精度的辨识结果。

The reliability of Li-ion battery management system is usually based on the high-precision identification of battery model parameters. Because of the noise interference in the actual operation and measurement environment, the noise in the sampling signal is usually caused, which results in the influence of the parameter identification accuracy by the conventional identification algorithm. Therefore, in this paper, the second-order RC model of lithium-ion battery is taken as the research object, with the influence of noise on the common variable forgetting factor recursive least square(VFFRLS)in the process of updating forgetting factors analyzed. A parameter identification algorithm is proposed, which is VFFRLS based on wavelet transform. The multi-scale and multi-resolution characteristics of wavelet transform are used to divide the frequency of the sampling signal, with VFFRLS used to identify the parameters. The problem of poor tracking performance caused by the forgetting factor caused by noise is solved. The experimental results show that the proposed algorithm has higher accuracy identification results.