State-of-health estimation of lithium-ion batteries based on improved long short-term memory algorithm. (September 2022)

• Record Type: Journal Article
• Authors: Gong, Yadong; Zhang, Xiaoyong; Gao, Dianzhu; Li, Heng; Yan, Lisen; Peng, Jun; Huang, Zhiwu
• Is Part Of: Journal of energy storage. Volume 53(2022)
• Publication Date: 2022-09
• Subjects: Lithium-ion battery -- State of health -- Long short-term memory network -- Particle swarm optimization -- Grey relational analysis; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2022.105046
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: Long short-term memory network (LSTM) is a popular deep learning network method for estimating the state of health (SOH) of lithium-ion batteries. However, the hyperparameters in the network are usually difficult to pre-define, which degrades the estimation accuracy in applications. To address this challenge, this paper proposes a data-driven estimation method based on the improved LSTM, where the network topology is estimated by the particle swarm optimization (PSO) algorithm. First, four health indicators in the charging and discharging process are selected. Grey relational analysis is further employed to quantify their correlations with the battery SOH. Then, an LSTM model is established to map the relationship between health factors and battery SOH. To tackle the parameter determination and slow convergence problems of the classical LSTM method, the particle swarm optimization algorithm is adopted to determine the key hyperparameters in the neural network, and the RMSProp training method and dropout technique are introduced to accelerate the convergence speed and avoid over-fitting problems. The experimental results show that the prediction accuracy is improved by at least 5 % when compared with the classical LSTM method. Highlights: An improved PSO-LSTM model is proposed to accurately estimate the battery SOH. The RMSProp algorithm is utilized to accelerate the training speed and dropout method is used to avoid overfitting problem. Health factor related to the battery temperature is considered during the modeling process. The cycle aging experiments of lithium-ion batteries are designed.