基于IPSO-LightGBM模型的锂离子电池荷电状态预测

Prediction of state of charge for lithium-ion batteries based on IPSO LightGBM model

为有效提升锂离子电池的荷电状态(State of Charge,SOC)预测精度,提出一种基于改进粒子群优化算法(Improved Particle Swarm Optimization,IPSO)和轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)的锂离子电池SOC预测模型,LightGBM模型用于构建锂离子电池SOC的预测,IPSO用于优化LightGBM模型的超参数。首先,对公开数据集进行预处理,并使用多种策略改进基本粒子群优化算法;其次,建立基于IPSO-LightGBM、LightGBM和反向传播(Back Propagation,BP)神经网络的锂离子电池SOC预测模型;最后,使用马里兰大学提供的电池数据集对三种模型进行实验仿真。结果表明,IPSO-LightGBM模型的预测准确率优于未优化的LightGBM模型和BP神经网络模型。

To effectively improve the accuracy of State of Charge(SOC)prediction for lithium-ion batteries,a lithium-ion battery SOC prediction model based on Improved Particle Swarm Optimization(IPSO)algorithm and Light Gradient Boosting Machine(LightGBM)is proposed.The LightGBM model is used to construct SOC prediction for lithium-ion batteries,and IPSO is used to optimize the hyperparameters of the LightGBM model.Firstly,the public dataset is preprocessed,and multiple strategies are used to enhance the basic particle swarm optimization algorithm;secondly,lithium-ion battery SOC prediction models based on IPSO-LightGBM,LightGBM,and Back Propagation(BP)neural networks are established;finally,experimental simulations on the three models are conducted using the battery dataset provided by the University of Maryland.The results indicate that the prediction accuracy of the IPSO-LightGBM model surpasses that of the unoptimized LightGBM model and the BP neural network model.