Xiaosong Hu, Yang Xin, F. Feng, Kailong Liu, Xianke Lin
{"title":"A Particle Filter and Long Short-Term Memory Fusion Technique for Lithium-Ion Battery Remaining Useful Life Prediction","authors":"Xiaosong Hu, Yang Xin, F. Feng, Kailong Liu, Xianke Lin","doi":"10.1115/1.4049234","DOIUrl":null,"url":null,"abstract":"\n Accurate prediction of the remaining useful life (RUL) of lithium-ion batteries can improve the durability, reliability, and maintainability of battery system operation in electric vehicles. To achieve high-accuracy RUL predictions, it is necessary to develop an effective method for long-term nonlinear degradation prediction and quantify the uncertainty of the prediction results. To this end, this paper proposes a hybrid approach for lithium-ion battery RUL prediction based on particle filter (PF) and long short-term memory (LSTM) neural network. First, based on the training set, the model parameters are iteratively updated using the PF algorithm. Second, the LSTM model parameters are obtained using the training set. The mean and standard deviation in the prediction stage are obtained through Monte Carlo (MC) dropout. Finally, the mean value predicted by MC-dropout is used as the measurement for the PF in the prediction phase, the standard deviation represents the uncertainty of the prediction result, and the mean and standard deviation are integrated into the measurement equation of the model. The experimental results show that the proposed hybrid approach has better prediction accuracy than the PF, LSTM algorithm, and two other types of hybrid approaches. The hybrid approach can obtain a narrower confidence interval.","PeriodicalId":54846,"journal":{"name":"Journal of Dynamic Systems Measurement and Control-Transactions of the Asme","volume":null,"pages":null},"PeriodicalIF":1.7000,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Dynamic Systems Measurement and Control-Transactions of the Asme","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1115/1.4049234","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
引用次数: 13
Abstract
Accurate prediction of the remaining useful life (RUL) of lithium-ion batteries can improve the durability, reliability, and maintainability of battery system operation in electric vehicles. To achieve high-accuracy RUL predictions, it is necessary to develop an effective method for long-term nonlinear degradation prediction and quantify the uncertainty of the prediction results. To this end, this paper proposes a hybrid approach for lithium-ion battery RUL prediction based on particle filter (PF) and long short-term memory (LSTM) neural network. First, based on the training set, the model parameters are iteratively updated using the PF algorithm. Second, the LSTM model parameters are obtained using the training set. The mean and standard deviation in the prediction stage are obtained through Monte Carlo (MC) dropout. Finally, the mean value predicted by MC-dropout is used as the measurement for the PF in the prediction phase, the standard deviation represents the uncertainty of the prediction result, and the mean and standard deviation are integrated into the measurement equation of the model. The experimental results show that the proposed hybrid approach has better prediction accuracy than the PF, LSTM algorithm, and two other types of hybrid approaches. The hybrid approach can obtain a narrower confidence interval.
期刊介绍:
The Journal of Dynamic Systems, Measurement, and Control publishes theoretical and applied original papers in the traditional areas implied by its name, as well as papers in interdisciplinary areas. Theoretical papers should present new theoretical developments and knowledge for controls of dynamical systems together with clear engineering motivation for the new theory. New theory or results that are only of mathematical interest without a clear engineering motivation or have a cursory relevance only are discouraged. "Application" is understood to include modeling, simulation of realistic systems, and corroboration of theory with emphasis on demonstrated practicality.