基于遗传算法最小二乘支持向量机蓄电池SOC估测

Estimation model of battery SOC based on least square support vector machine optimized by genetic algorithm

介绍了最小二乘支持向量机(LS-SVM)和遗传算法(GA)的基本理论,建立了基于遗传算法的最小二乘支持向量机蓄电池SOC估测模型。通过数据验证选择了模型的最优核函数,同时利用遗传算法对模型的参数进行了寻优。将寻优结果代入模型进行验证,结果表明,该模型具有很高的预测精度,应用在装甲车辆铅酸蓄电池SOC测上具有很高的实用价值。

The basic theories of the LS-SVM （Least Square Support Vector Machine） and GA （Genetic Algorithm） were introduced, then the battery SOC estimate model based on LS-SVM and GA was built. The best kernel of the model was chosen through the data verification and searching best to the model parameter using GA at the same time. The searching result was introduced into the model. The verification result shows that the model has a high precision, and a high practical value when using in the battery SOC estimation.