公开(公告)号：EP2648011A1

公开(公告)日：2013-10-09

申请号：EP12176616.6

申请日：2012-07-16

申请人： Samsung SDI Co., Ltd.

发明人： Baek, Ho Yul ,  Shin, Jeongsoon ,  Park, Cheol Ho ,  Chai, Heekwan ,  Ko, Younghwa ,  Suh, Seung Bum ,  Cheong, Kyeong-Beom ,  Park, Nokap

IPC分类号： G01R31/36 ,  H01M10/48

CPC分类号： G01R31/3679 ,  G01R31/3651 ,  G06N5/00 ,  H01M10/48

摘要： The present invention provides a system for predicting the lifetime of a battery. In one embodiment, the battery lifetime prediction system comprises a learning data input unit adapted to receive as learning data battery factors comprising at least one design factor and/or at least one formation factor and/or at least one process factor of a battery cell targeted for learning, and additionally at least one measurement factor representing a lifetime of the learning battery cell targeting for learning. Further it comprises a target data input unit adapted to receive as target data a corresponding at least one battery factor associated with a prediction battery cell targeted for lifetime prediction and a machine learning unit, adapted to assign weights to the battery factors inputted into the data input unit by performing machine learning on the learning data. Moreover the battery lifetime prediction system comprises a life time prediction unit, adapted to combine the target data of the prediction battery cell with the weights of the machine learning unit resulting in lifetime data indicating a lifetime of the prediction battery cell.

摘要翻译： 本发明提供一种用于预测电池寿命的系统。 在一个实施例中，电池寿命预测系统包括学习数据输入单元，其适于接收作为学习数据的电池因子，所述电池因素包括至少一个设计因素和/或至少一个形成因子和/或至少一个目标电池单元的过程因子 以及另外至少一个测量因子，代表学习电池单元的目标学习的寿命。 此外，它包括目标数据输入单元，其适于作为目标数据接收与用于寿命预测的预测电池单元相关联的相应的至少一个电池系数，以及机器学习单元，适于为输入到数据输入的电池因子分配权重 通过对学习数据进行机器学习来实现。 此外，电池寿命预测系统包括寿命预测单元，其适于将预测电池单元的目标数据与机器学习单元的权重组合，得到指示预测电池单元的寿命的寿命数据。