基于自编码器和PCA的滚筒烘丝机多块过程监测方法
Multi block process monitoring method of drum dryer based on autoencoder and PCA
-
摘要: 鉴于滚筒烘丝干燥过程的多变量、强耦合、非线性的复杂特征，而传统主成分分析（PCA）过程监测方法不具有非线性表达能力及全局建模无法精确检测故障的问题，提出自动编码器特征提取和多块主成分分析联合驱动的多块建模方法。首先，依据叶丝干燥工作原理对变量进行分块，以突出过程局部特征；其次，使用自编码器提取每个子块的非线性特征；再次，分别建立相应的PCA监测模型；最后，通过贝叶斯推理对多个子空间的监测结果进行融合决策。以两个实际叶丝干燥案例进行验证，结果表明：该方法的报警率分别高达91.67%和98.21%，相比传统PCA和AE-PCA检测方法，该方法能准确揭示并表征干燥过程的整体运行状态及局部特征信息，提高对滚筒叶丝干燥生产过程的异常检测精度，实现对质量异常情况的准确报警，有利于保证滚筒叶丝干燥过程的稳定生产。Abstract: The process of drum drying silk has complex characteristics of multivariability, strong coupling, and nonlinearity. Traditional Principal Component Analysis (PCA) method lacks strong nonlinear ability, and global modeling method is difficult to achieve accurate fault detection of the process.Therefore, this paper proposed a novel multi-block approach by integrating Autoencoder (AE) for feature extraction and PCA model.Initially, in order to capture local features, the process variables were divided into blocks according to the drying process principle of tobacco leaf.Secondly, autoencoder was used to extract the nonlinear features of each sub-block.Then, the corresponding PCA models are established for each sub-block, respectively.Lastly, the monitoring results of multiple subspaces were fused for decision-making by Bayesian inference. Two actual leaf silk drying cases were used for verification, and the results showed that the alarm rates of this method were as high as 91.67% and 98.21%. Compared to traditional PCA and AE-PCA detection methods, this algorithm could accurately reveal and characterize the overall operating status and local feature information of the drying process, improve the accuracy of anomaly detection in the drum leaf silk drying production process and achieve accurate alarm for quality anomalies, to ensure stable production of the drum leaf silk drying process.
-
Key words:
- drum dryer /
- autoencoder /
- principal component analysis /
- multi-block modeling /
- process monitoring
-
-
[1]
李永生, 何文苗, 李石头, 等.烟气水分、感官舒适性与烘丝工艺参数的关系研究[J].郑州轻工业学院学报(自然科学版), 2014, 29(6):13-16.
-
[2]
黄胜, 李建辉, 张永川.长沙卷烟厂SPC系统的应用实践[J].中国烟草学报, 2008, 14(S1):14-17.
-
[3]
李文泉, 赵文田, 李文斌.统计过程控制技术SPC在烟草制丝生产中的应用[J].机械工程与自动化, 2009(5):116-118.
-
[4]
张敏, 童亿刚, 戴志渊, 等.SPC技术在制丝质量管理中的初步应用[J].烟草科技, 2004(9):10-11.
-
[5]
张雷, 李金学, 堵劲松, 等.基于DGRU网络的烘丝机筒壁温度动态预测[J].轻工学报, 2022, 37(6):85-91
, 100. -
[6]
商庆杰, 潘宏菽, 张力江, 等.实施统计过程控制(SPC)中常见问题的探讨[J].现代制造技术与装备, 2016(2):126-128, 130.
-
[7]
杨锦忠, 宋希云.多元统计分析及其在烟草学中的应用[J].中国烟草学报, 2014, 20(5):134-138.
-
[8]
马洁, 党爱民, 李刚, 等.基于MSPM的故障诊断技术研究现状与展望[J].华侨大学学报(自然科学版), 2012, 33(6):601-607.
-
[9]
王伟, 赵春晖.基于PCA的卷烟制丝过程监测与故障诊断[J].控制工程, 2017, 24(12):2435-2442.
-
[10]
WANG L, DENG X G.Multi-block principal component analysis based on variable weight information and its application to multivariate process monitoring [J].Canadian Journal of Chemical Engineering, 2018, 96(5):1127-1141.
-
[11]
王伟, 张利宏, 黎明星, 等.卷烟制叶丝段批次过程的多阶段分布式监测与异常诊断[J].烟草科技, 2018, 51(2):69-76.
-
[12]
鲍宇, 程硕, 王靖涛.基于深度学习的化工过程故障检测与诊断研究综述[J].化学工业与工程, 2022, 39(2):9-22.
-
[13]
LIU C L, WANG Y L, WANG K, et al.Deep learning with nonlocal and local structure preserving stacked autoencoder for soft sensor in industrial processes[J].Engineering Applications of Artificial Intelligence, 2021, 104:104341.
-
[14]
WEHRENS R.Chemometrics with R:Multivariate data analysis in the natural sciences and life sciences[D].[S.l.:s.n.], 2011:43-66.
-
[15]
ZHANG J X, CHEN M Y, HONG X.Nonlinear process monitoring using a mixture of probabilistic PCA with clusterings [J].Neurocomputing, 2021, 458:319-326.
-
[16]
DONG J, SUN R Q, PENG K X, et al.Quality monitoring and root cause diagnosis for industrial processes based on Lasso-SAE-CCA[J].IEEE Access, 2019, 7:90230-90242.
-
[17]
GU B B, XIONG W L.Multi-mode process monitoring based on multi-block information extraction PCA method with local neighbourhood standardization[J].International Journal of Modelling Identification and Control, 2019, 32(3/4):264-273.
-
[18]
郑静, 熊伟丽.基于互信息的多块k近邻故障监测及诊断[J].智能系统学报, 2021, 16(4):717-728.
-
[19]
童楚东, 蓝艇, 史旭华.基于互信息的分散式动态PCA故障检测方法[J].化工学报, 2016, 67(10):4317-4323.
-
[1]
-
点击查看大图
计量
- PDF下载量: 34
- 文章访问数: 1010
- 引证文献数: 0
下载:
