The invention provides an electric vehicle lithium battery SOH online prediction method facing edge federated learning. The method comprises the following steps: 1, deploying the same battery health state prediction model at a vehicle-mounted terminal and a cloud server side; 2, collecting real-time driving data of the vehicle-mounted terminal, and preprocessing real vehicle data; 3, carrying out local model training on each vehicle-mounted terminal; 4, uploading model parameters to the cloud server after each vehicle-mounted terminal completes training; 5, the third step and the fourth step are cycled, and iterative updating is carried out; and 6, predicting the health state of the battery. Aiming at the problems of vehicle-mounted terminal data privacy leakage and network congestion caused by a central machine learning model, federated learning allows a vehicle-mounted terminal not to send local data to a cloud server any more, and joint optimization of a plurality of isolated data centers on the machine learning model is realized while data localization is ensured, so that the data processing efficiency is improved. And meanwhile, the model parameters replace the user data to be uploaded to the cloud, so that the problem of network congestion is effectively solved.

    本发明提出一种面向边缘联邦学习的电动汽车锂电池SOH在线预测方法,包括如下步骤:一:在车载终端及云服务器端部署相同的电池健康状态预测模型;二:收集车载终端的实时行驶数据,并对实车数据进行预处理;三:各车载终端进行本地模型训练;四:各车载终端完成训练后,将模型参数上传至云服务器;五:循环三和四,进行迭代更新;六:电池健康状态预测。本发明针对中心式机器学习模型导致的车载终端数据隐私泄露,以及网络堵塞的问题,联邦学习允许车载终端不再向云服务器发送本地数据,在保证数据本地化的同时实现多个孤立数据中心对机器学习模型进行联合优化,保证了数据的隐私性,同时模型参数代替了用户数据上传至云端,有效解决网络堵塞的问题。


    Access

    Download


    Export, share and cite



    Title :

    Edge federated learning-oriented online prediction method for SOH (state of health) of lithium battery of electric vehicle


    Additional title:

    面向边缘联邦学习的电动汽车锂电池SOH在线预测方法


    Contributors:
    XIAO FEI (author) / WU LIFENG (author)

    Publication date :

    2022-07-01


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G01R Messen elektrischer Größen , MEASURING ELECTRIC VARIABLES / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung





    Online State-of-Health Prediction of Lithium-Ion Battery with a Multiscale Filtering Technique

    Youn, B.D. / Hu, C. / Kim, T. et al. | British Library Conference Proceedings | 2012



    Modular machine learning structure for electric vehicle battery state of health estimation

    NEGOITA GIANINA ALINA / TESKEY WESLEY / RENN JEAN-BAPTISTE et al. | European Patent Office | 2024

    Free access