电气技术  2024, Vol. 25 Issue (5): 1-10    DOI:
研究与开发 |
基于贝叶斯优化-卷积神经网络-双向长短期记忆神经网络的锂电池健康状态评估
衣思彤1, 刘雅浓2, 马耀浥1, 李文婕3, 孔航3
1.大连交通大学自动化与电气工程学院,辽宁 大连 116028;
2.大连交通大学机车车辆工程学院,辽宁 大连 116028;
3.大连交通大学计算机与通信工程学院,辽宁 大连 116028
State of health assessment of lithium battery based on Bayesian optimization-convolution neural network-bi-directional long short term memory neural network
YI Sitong1, LIU Yanong2, MA Yaoyi1, LI Wenjie3, KONG Hang3
1. School of Automation and Electrical Engineering, Dalian Jiaotong University, Dalian, Liaoning 116028;
2. School of Rolling Stock Engineering, Dalian Jiaotong University, Dalian, Liaoning 116028;
3. School of Computer and Communication Engineering, Dalian Jiaotong University, Dalian, Liaoning 116028
全文: PDF (4908 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 准确估计电池健康状态是设备稳定运行的关键。针对当前健康状态研究中容量难以直接测量、估计模型调参费时等问题,提出基于多健康特征的贝叶斯优化(BO)算法优化卷积神经网络(CNN)与双向长短期记忆(BiLSTM)神经网络预测模型。基于NASA公开锂电池数据,提取3种健康特征。将CNN与BiLSTM结合,提高时间序列数据处理能力,加入BO算法自动搜寻最优参数集,避免组合网络模型陷入局部最优,从而减少评估时间。对比分析相关神经网络模型,结果表明所提方法预测准确度最高,可有效估计锂电池的健康状态,平均绝对误差和方均根误差均在1%以内。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
衣思彤
刘雅浓
马耀浥
李文婕
孔航
关键词 锂电池健康状态(SOH)贝叶斯优化(BO)算法卷积神经网络(CNN)双向长短期记忆(BiLSTM)神经网络    
Abstract:Accurate estimation of battery state of health (SOH) is the key to the stable operation of the device. In order to solve the problems in the current SOH research, such as the difficulty to measure the volume directly and the time required to adjust the model parameters, a prediction model based on the multi-health features of Bayesian optimization (BO) optimized convolution neural network (CNN) and bi-directional long short term memory (BiLSTM) neural network is proposed. Based on NASA’s publicly available lithium battery data, three health characteristics are extracted. The combination of CNN and BiLSTM improves the processing ability of time series data, and adds BO algorithm to automatically search the optimal parameter set, which avoids the combination network model falling into the local optimal and reduces the estimation time. The results show that the proposed method has the highest prediction accuracy and can aeffectively estimate the SOH of lithium batteries. The mean absolute error and root mean square error are both within 1%.
Key wordslithium battery    state of health (SOH)    Bayesian optimization (BO)    convolutional neural network (CNN)    bi-directional long short term memory (BiLSTM) neural network   
收稿日期: 2024-01-20     
基金资助:辽宁省自然科学基金(2021-MS-298)
作者简介: 衣思彤(1999—),女,辽宁省葫芦岛市人,硕士研究生,主要从事智能交通控制与安全技术方面的研究工作。
引用本文:   
衣思彤, 刘雅浓, 马耀浥, 李文婕, 孔航. 基于贝叶斯优化-卷积神经网络-双向长短期记忆神经网络的锂电池健康状态评估[J]. 电气技术, 2024, 25(5): 1-10. YI Sitong, LIU Yanong, MA Yaoyi, LI Wenjie, KONG Hang. State of health assessment of lithium battery based on Bayesian optimization-convolution neural network-bi-directional long short term memory neural network. Electrical Engineering, 2024, 25(5): 1-10.
链接本文:  
http://dqjs.cesmedia.cn/CN/Y2024/V25/I5/1