nav emailalert searchbtn searchbox tablepage yinyongbenwen piczone journalimg journalInfo searchdiv qikanlogo popupnotification paper paperNew
2023, 09, v.50 8-15
基于Mixup数据增强的CNN-GRU深度学习电火花线切割放电状态识别
基金项目(Foundation): 国家自然科学基金面上项目(51275098)
邮箱(Email): su_guokang@163.com;
DOI:
摘要:

针对电火花线切割放电状态识别中,数据集样本较少导致训练模型准确率不高的问题,提出基于Mixup数据增强的CNN-GRU网络算法。该算法首先使用Mixup数据增强对原始的电火花波形数据进行数据增强,通过线性插值对数据进行混合,得到新的扩容之后的数据集;随后使用增强的数据集训练CNN-GRU模型,并用该模型进行分类。经实验表明,使用Mixup数据增强的CNN-GRU模型能有效的识别出数据中的“时序特征”与“局部特征”,且模型的准确率达到了96%。

Abstract:

In order to solve the problem of low accuracy of training models due to small sample data in the recognition of discharge state of WEDM, a CNN-GRU network algorithm based on Mixup data enhancement is proposed. In this algorithm, the original EDM waveform data is enhanced by Mixup data enhancement, and the data is mixed by linear interpolation to obtain a new expanded data set. The CNN-GRU model is then trained using the enhanced data set, and the model is used for classification. The experiment show that the CNN-GRU model trained by Mixup data enhancement can effectively identify the "temporal features" and "local features" of the data, and the accuracy of the model reaches 96%.

参考文献

[1]李立青,郭艳玲,白基成,等.电火花加工技术研究的发展趋势预测[J].机床与液压,2008(2):174-178.

[2]潘伯郁.往复走丝电火花线切割机床智能自适应采样控制系统和纳秒级高频电源的研发及应用[J].电加工与模具,2020(6):25-28.

[3]刘志东,王振兴,张艳,等.高速走丝电火花线切割高效切割技术研究[J].中国机械工程,2011,22(4):385-389.

[4]霍孟友,张建华,艾兴.电火花放电加工间隙状态检测方法综述[J].电加工与模具,2003(3):17-19.

[5]陈吉红,胡鹏程,周会成,等.走向智能机床[J]. Engineering,2019,5(4):186-210.

[6]Kao J Y,Tarng Y S. A neutral-network approach for the on-line monitoring of the electrical discharge machining process[J]. Journal of Materials Processing Technology,1997,69(1-3):112-119.

[7]TARNG Y S,TSENG C M,CHUNG L K. A fuzzy pulse discriminating system for electrical discharge machining[J].International Journal of Machine Tools and Manufacture,1997,37(4):511-522.

[8]李云龙.基于粒子群算法的线切割放电状态检测研究[D].哈尔滨:哈尔滨理工大学,2014.

[9]闫雯雯.基于神经网络的电火花线切割加工放电状态检测研究[D].哈尔滨:哈尔滨理工大学,2013.

[10]唐琦.基于PNN算法的电火花线切割间隙放电状态检测研究[D].哈尔滨:哈尔滨理工大学,2018.

[11]刘长红.基于声光信号和深度学习的电火花线切割伺服控制系统研究[D].广州:广东工业大学,2022.

[12]沈旭东.基于深度学习的时间序列算法综述[J].信息技术与信息化,2019(1):71-76.

[13]WEN Q,SUN L,YANG F,et al. Time series data augmentation for deep learning:A survey[J]. arXiv preprint,2020:12478.

[14]SHORTEN C,KHOSHGOFTAAR T M. A survey on imagedata augmentation for deep learning[J]. Journal of big data,2019,6(1):1-48.

[15]FIELDS T,HSIEH G,CHENOU J. Mitigating drift intimeseries data with noise augmentation[C]. 2019 International Conference on Computational Science and Computational Intelligence(CSC),IEEE,2019:227-230.

[16]IWANA B K,UCHIDA S. Time series data augmentation for neural networks by time warpingwith a discriminative teacher[C].2020 25th International Conference on Pattern Recognition(ICPR),IEEE,2021:3558-3565.

基本信息:

DOI:

中图分类号:TG484

引用信息:

[1]叶之骞,钟紫鹏,邓永聪等.基于Mixup数据增强的CNN-GRU深度学习电火花线切割放电状态识别[J].机械,2023,50(09):8-15.

基金信息:

国家自然科学基金面上项目(51275098)

引用

GB/T 7714-2015 格式引文
MLA格式引文
APA格式引文
检 索 高级检索