摘要
为提高汽车电池包密封性检测效率,达到气体泄漏快速、在线测试的目的,提出一种基于特征提取与多层感知器(MLP)神经网络的电池包密封性检测建模方法。首先使用Lasso方法对气体泄漏数据集进行特征提取,然后用MLP神经网络对数据集进行辨识,两者共同构成电池包密封性检测模型。研究结果表明:该检测模型可以缩短气体泄漏检测周期,实现电池包密封性快速在线检测、提升整个工序生产效率的目标。
In order to improve the efficiency of sealing detection in automobile battery pack and achieve rapid and online testing of gas leakage,an efficient modeling method of sealing detection based on feature extraction and multilayer perceptron(MLP)neural network is proposed.First,the Lasso in the embedded method is used to extract the feature of the gas leak data set.Then use MLP neural network to identify the data set,constitute the battery pack sealing detection model together with the Lasso.The research results show that:the model can reduce the duration of gas leakage detection,realize rapid and on-line detection,and improve the production efficiency of the entire process.
作者
杨仕堂
罗磊
时轮
YANG Shitang;LUO Lei;SHI Lun(School of Mechanical Engineering,ShanghaiJiaotong University,Shanghai 200240,China;Shanghai SmartState Technology Co.Ltd,Shanghai 201306,China)
出处
《机械设计与研究》
CSCD
北大核心
2020年第6期139-142,176,共5页
Machine Design And Research
基金
国家科技重大专项资助(2019ZX04027-001)。