重庆理工大学学报（自然科学版）2011，Vol.25Issue(3)：92-96,5.
基于DE优化SVR的锂离子电池剩余容量预测
SVR Optimized by DE Optimization Algorithm for Remaining Capacity Forecasting
摘要
Abstract
In the analysis of support vector machine for regression (SVR) on the model of remaining capacity with nonlinearity, aimed at the puzzle of selection of SVR's parameters, the paper proposed a predictive model for remaining capacity of lithium ion batteries. The model was based on the SVR with parameter optimized by differential evolution (DE) algorithm. DE has powerful global searching ability, which would be applied to the optimization of SVR's parameters, and the predictive precision of the lithium ion batteries capacity was compared between DE algorithm and particle swarm optimization (PSO) algorithm. Simulation results show that DE-SVR is better than PSO-SVR in prediction of lithium ion batteries capacity.关键词
支持向量机回归/微分进化算法/粒子群优化算法/参数选择/锂离子电池/容量预测Key words
support vector machine for regression/ differential evolution algorithm/ particle swarm optimization (PSO) algorithm/ parameter selection/ lithium ion batteries/ capacity prediction分类
动力与电气工程引用本文复制引用
唐超,曹龙汉,赵泽鑫,何俊强,吴珍毅..基于DE优化SVR的锂离子电池剩余容量预测[J].重庆理工大学学报（自然科学版）,2011,25(3):92-96,5.基金项目
国防科研项目(TZ-CQTY-Y-A-2010-002) （TZ-CQTY-Y-A-2010-002）
2010年重庆高校优秀成果转化项目(Kjzh10219) （Kjzh10219）
