| 基于支持向量回归的电池SOC估计方法研究 |
| |
| 引用本文： | 裴晟,陈全世,林成涛. 基于支持向量回归的电池SOC估计方法研究[J]. 电源技术, 2007, 31(3): 242-243,252 |
| |
| 作者姓名： | 裴晟 陈全世 林成涛 |
| |
| 作者单位： | 清华大学,汽车安全与节能国家重点实验室,北京,100084;清华大学,汽车安全与节能国家重点实验室,北京,100084;清华大学,汽车安全与节能国家重点实验室,北京,100084 |
| |
| 摘 要： | 电池荷电状态(SOC)估计的准确性对于混合动力汽车至关重要.将支持向量回归方法用于电动汽车电池SOC的估计.方法中考虑了电池温度、电压、电流、净安时数等因素,对于电动汽车典型工况试验数据得到了小于0.04的误差.比较研究表明:支持向量回归方法比神经网络方法有更好的鲁棒性.
|
| 关 键 词： | 电动汽车 SOC 支持向量回归 |
| 文章编号： | 1002-087X(2007)03-0242-03 |
| 修稿时间： | 2006-09-25 |
Study on estimating method for battery state of charge based on support vector regression |
| |
| PEI Sheng,CHEN Quan-shi,LIN Cheng-tao. Study on estimating method for battery state of charge based on support vector regression[J]. Chinese Journal of Power Sources, 2007, 31(3): 242-243,252 |
| |
| Authors: | PEI Sheng CHEN Quan-shi LIN Cheng-tao |
| |
| Abstract: | The state of charge (SOC) determination of battery for hybrid electric vehicle (HEV) is very important. A novel way to estimate SOC by support vector regression (SVR) was presented. Temperature, voltage, current and ampere-hour were considered. The error between training data and test data is less than 0.04. The comparison results demonstrate that support vector regression method has better robust capability than artificial neural network method. |
| |
| Keywords: | electric vehicle state of charge(SOC) support vector regression |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
