基于IGWO-SVR的锂电池健康状态预测

Prediction of Lithium Battery Health Status Based on IGWO-SVR

锂离子电池以其高效能和无污染等优势成为我们生活中非常重要的储能元件,而锂电池的健康状态(SOH)是保证系统稳定的决定性因素。该文基于此提出一种改进的灰狼算法优化支持向量回归(IGWO-SVR)来提高锂电池SOH的预测精度。首先提取与锂电池退化有一定相关性的健康因子,然后提出基于正切的收敛因子和Levy策略的灰狼算法,对于模型的泛化能力有着很大的提升,收敛能力和搜索范围的评估也有着不错的提高,采用改进的GWO进行参数寻优,解决SVR模型参数选择困难的问题。根据NASA提供的数据集进行验证,有效地提高了电池健康状态的预测准确性且具有一定的实用性。

With advantages such as high efficiency and no pollution,lithium-ion batteries have become a very essential energy storage component in our lives,and the state of health(SOH)of lithium batteries is a critical factor in ensuring system stability.This research offers an improved grey wolf algorithm optimized support vector regression(IGWO-SVR)to improve lithium battery SOH prediction accuracy.First,health factors with a known relationship to lithium battery degradation are identified,and then the gray wolf algorithm based on the tangent convergence factor and the Levy strategy is proposed,which has a significant improvement in the model’s generalization ability as well as a good improvement in the convergence ability and search evaluation,and the improved GWO is used for parameter search to solve the problem of difficult selection of SVR model parameters.Validation based on the data set provided by NASA,which effectively improves the accuracy of battery health state prediction and has certain practicality.