基于主元分析的锂动力电池SOC估算方法研究

发布时间：2018-03-30 03:38

本文选题：电池SOC　切入点：主元分析　出处：《哈尔滨理工大学》2014年硕士论文

【摘要】：近年来，随着人们对环保和能源问题重视程度的日益提高，促使了以电动汽车为主流的节能减排汽车行业的兴起，而作为新兴的电动汽车能量的来源——锂离子动力电池的研究也就受到了更多的重视。电池SOC作为电动汽车动力电池的重要参数，它的准确估算关乎整车的控制策略是否能够得以正确实施，因此，电池SOC的估算研究也就成为了广大学者的重要研究课题。由于现有的估算方法往往很难实现对电池SOC的准确估算，因此，如何准确的估算电池SOC已经成为人们关注的焦点。主元分析算法作为多元统计分析的方法，能够进行数据的简化和数据压缩并提取变量中的重要因子，进而进行数据的分析和结果预测，因此，利用该算法对电池SOC建立估算模型具有很强的理论意义和实际意义。本文通过测试实验对影响电池SOC的性能参数进行了分析，得出影响电池SOC的关键因素。然后，通过对主元分析理论的深入剖析，分析了该算法对于电池SOC估算的可行性，提出了利用主元分析算法结合最小二乘回归法建立电池SOC的估算模型，并通过模型的仿真与实验对该模型进行了实验验证。鉴于PCA算法不能提取参数中的非线性因子而造成电池SOC的估算误差，进一步提出了采用核主元分析算法建立电池SOC估算模型，并通过仿真实验对所建立的模型进行了验证。针对核主元分析电池SOC估算模型不能适应温度多变或者电池劣化程度不同的情况，基于电池工作温度和基于电池劣化程度两种方式对模型进行了改进。改进后的电池SOC估算模型通过实验进行了模型的仿真与实验验证。仿真结果表明，改进后的模型能够适应更复杂的环境，满足实时性和可靠性的要求，估算精度有所提高，平均估算误差为1.46%，，优于安时计量法电池SOC估算模型，具有更好的电池SOC估算效果。
[Abstract]:In recent years, with the increasing attention paid to environmental protection and energy issues, the emergence of energy-saving and emission reduction automobile industry, which is the mainstream of electric vehicles, has been promoted. As a new energy source of electric vehicle, lithium ion battery has been paid more attention. As an important parameter of electric vehicle power battery, the accurate estimation of battery SOC is related to whether the control strategy of the whole vehicle can be implemented correctly. The estimation of battery SOC has become an important research topic for many scholars. Because the existing estimation methods are often difficult to realize the accurate estimation of battery SOC, therefore, How to estimate battery SOC accurately has become the focus of attention. As a method of multivariate statistical analysis, principal component analysis (PCA) algorithm can simplify and compress data and extract important factors from variables. Then the data are analyzed and the results are predicted. Therefore, it is of great theoretical and practical significance to establish the estimation model of battery SOC by using this algorithm. In this paper, the performance parameters of battery SOC are analyzed through testing experiments, and the key factors affecting battery SOC are obtained. Then, the feasibility of the algorithm for estimating battery SOC is analyzed through the in-depth analysis of principal component analysis theory. The estimation model of battery SOC based on principal component analysis (PCA) algorithm and least square regression method is proposed. The model is verified by simulation and experiment. Because the PCA algorithm can not extract the nonlinear factor in the parameters, the estimation error of SOC is caused. Furthermore, the core principal component analysis (KPCA) algorithm is used to establish the battery SOC estimation model, and the model is verified by simulation experiments. The SOC estimation model of nuclear principal component analysis battery can not adapt to the situation of variable temperature or different degradation degree of the cell. The model is improved based on the operating temperature of the battery and the degradation degree of the battery. The improved SOC estimation model of the battery is simulated and verified by experiments. The simulation results show that, The improved model can adapt to more complex environment and meet the requirements of real-time and reliability. The estimation accuracy is improved, and the average estimation error is 1.46, which is superior to the amperometric battery SOC estimation model and has better effect of battery SOC estimation.
【学位授予单位】：哈尔滨理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM912

【参考文献】