基于迁移变分自编码器-标签映射的湿式球磨机负荷参数软测量

doi:10.11949/j.issn.0438-1157.20181069

摘要/Abstract

摘要：

在工况改变时，湿式球磨机的实时数据和建模数据分布不一致，不满足传统软测量建模方法要求的数据同分布假设，导致模型失准和性能恶化。为此，引入迁移学习思想，提出一种基于迁移变分自编码器-标签映射的软测量模型，实现多工况下湿式球磨机负荷参数的准确测量。首先，迁移目标域数据编码得到的隐变量分布参数，对源域数据对应隐变量进行拟合，再解码得到迁移数据；然后采用相似性度量选取相似样本构建标签映射模型，并得到映射标签；最后使用迁移数据和映射标签构建出最终的软测量模型。实验结果表明，该软测量方法显著优于现有方法，适用于多工况下的软测量建模。

关键词: 迁移学习, 变分自编码器, 标签映射, 湿式球磨机负荷参数, 过程控制, 预测, 实验验证

Abstract:

When the working condition of wet ball mill is changed, the distributions of real time data and modeling data are inconsistent, and the i.i.d assumption of traditional soft sensor method would not be satisfied, which leads to the inaccuracy of soft sensor model and the deterioration of the system performance. Therefore, by introducing the transfer learning method, a soft sensor model based on transfer variational autoencoder - label mapping strategy is proposed to achieve the precise measurement of wet ball mill load parameters under multiple working conditions. Firstly, the parameters of hidden variables distribution obtained by encode with target domain data are transferred to fit corresponding hidden variable of source domain data. For then, acquiring transfer data by decoding. Moreover, the similarity measure is used to select similar samples to construct the label mapping model, and the mapped labels are obtained. Then the final soft sensor model is constructed according to the transfer data and mapped labels. The experimental results show that the proposed soft sensor method is significantly better than the existing method, and the proposed method is suitable for soft sensor modeling under the situation of multi working condition.

Key words: transfer learning, variational autoencoder, label mapping, wet ball mill load parameters, process control, prediction, experimental validation

中图分类号:

TP 29

支恩玮, 闫飞, 任密蜂, 阎高伟. 基于迁移变分自编码器-标签映射的湿式球磨机负荷参数软测量[J]. 化工学报, 2019, 70(S1): 150-157.

Enwei ZHI, Fei YAN, Mifeng REN, Gaowei YAN. Soft sensor of wet ball mill load parameters based on transfer variational autoencoder - label mapping[J]. CIESC Journal, 2019, 70(S1): 150-157.

图/表 7

图1 变分自编码器

Fig.1 Variational Autoencoder

图2 隐变量z概率密度曲线

Fig.2 Probability density curves of hidden variables

图3 TVAE-LM算法流程

Fig.3 Flow diagram of TVAE-LM algorithm

图4 引入迁移预测结果对比

Fig.4 Comparison of predicted results with transfer

图5 引入LM预测结果对比

Fig.5 Comparison of predicted results with LM

表1 引入标签映射预测结果对比(RMSE)

Table 1 Comparison of prediction results by introducing label mapping(RMSE)

Working condition	TVAE			TVAE-LM
Working condition	MBVR	PD	CVR	MBVR	PD	CVR
1→2	0.1961	0.0345	0.0543	0.1236	0.0266	0.0164
1→3	0.3569	0.0430	0.1185	0.1398	0.0291	0.0213
3→1	0.5595	0.0791	0.0754	0.3065	0.0579	0.0349
3→2	0.2947	0.0594	0.0450	0.1181	0.0291	0.0157

图6 不同建模方法软测量实验误差比较

Fig.6 Comparison of prediction results in different modeling methods

参考文献 31

1	汤健, 赵立杰, 柴天佑, 等. 基于振动频谱的磨机负荷在线软测量建模[J]. 信息与控制, 2012, 41(1): 123-128.
	TangJ, ZhaoL J, ChaiT Y, et al. On-line soft-sensing modelling of mill load based on vibration spectrum[J]. Information & Control, 2012, 41(1): 123-128.
2	王恒, 贾民平, 陈左亮. 基于LS-SVM和机理模型的球磨机料位软测量[J]. 电力自动化设备, 2010, 30(7): 92-95.
	WangH, JiaM P, ChenZ L. Soft measurement based on mechanism model and LS-SVM for fill level of ball mill[J]. Electric Power Automation Equipment, 2010, 30(7): 92-95.
3	汤健. 基于频谱数据驱动的旋转机械设备负荷软测量[M]. 北京: 国防工业出版社, 2015: 25-28.
	TangJ. Soft Sensing of Rotating Machinery Equipment Load Based on Spectrum Data Drive[M]. Beijing: National Defense Industry Press, 2015: 25-28.
4	ZhouP, ChaiT Y, WangH. Intelligent optimal-setting control for grinding circuits of mineral processing process[J]. IEEE Transactions on Automation Science & Engineering, 2009, 6(4): 730-743.
5	XieS W, XieY F, LiF B, et al. Optimal setting and control for iron removal process based on adaptive neural network soft-sensor[J]. IEEE Transactions on Systems Man & Cybernetics Systems, 2018, (99): 1-13.
6	程瑞辉, 阎高伟. 基于OBE-ELM的球磨机料位软测量[J]. 中北大学学报(自然科学版), 2017, 38(5): 574-579.
	ChengR H, YanG W. Soft sensor for ball mill fill level based on OBE-ELM model[J]. Journal of North University of China (Natural Science Edition), 2017, 38(5): 574-579.
7	TangJ, WangD H, ChaiT Y. Predicting mill load using partial least squares and extreme learning machines[J]. Soft Computing, 2012, 16(9): 1585-1594.
8	XiaL Y, WangH N, ZhuP F, et al. Soft-sensor modeling method using kernel principal component analysis bagging ensemble neural network[J]. Information and Control, 2015, 44(5): 519-524.
9	牛大鹏, 刘元清. 基于改进即时学习算法的湿法冶金浸出过程建模[J]. 化工学报, 2017, 68(7): 2873-2879.
	NiuD P, LiuY Q. Modeling hydrometallurgical leaching process based on improved just-in-time learning algorithm[J]. CIESC Journal, 2017, 68(7): 2873-2879.
10	YaoL, GeZ Q. Deep learning of semi-supervised process data with hierarchical extreme learning machine and soft sensor application[J]. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1490-1498.
11	汤健, 柴天佑, 丛秋梅, 等. 基于EMD和选择性集成学习算法的磨机负荷参数软测量[J]. 自动化学报, 2014, 40(9): 1853-1866.
	TangJ, ChaiT Y, CongQ M, et al. Soft sensor approach for modeling mill load parameters based on EMD and selective ensemble learning algorithm[J]. Acta Automatica Sinica, 2014, 40(9): 1853-1866.
12	刘佳, 邵诚, 朱理. 基于迁移学习工况划分的裂解炉收率PSO-LS-SVM建模[J]. 化工学报, 2016, 67(5): 1982-1988.
	LiuJ, ShaoC, ZhuL. Modeling of cracking furnace yields with PSO-LS-SVM based on operating condition classification by transfer learning[J]. CIESC Journal, 2016, 67(5): 1982-1988.
13	GrubingerT, ChasparisG C, NatschlaegerT. Generalized online transfer learning for climate control in residential buildings[J]. Energy and Buildings, 2017, 139: 63-71.
14	YanK, ZhangD. Calibration transfer and drift compensation of e-noses via coupled task learning[J]. Sensors and Actuators B Chemical, 2016, 225: 288-297.
15	PanS J, YangQ. A survey on transfer learning[J]. IEEE Transactions on Knowledge & Data Engineering, 2010, 22(10): 1345-1359.
16	庄福振, 罗平, 何清, 等. 迁移学习研究进展[J]. 软件学报, 2015, 26(1): 26-39.
	ZhuangF Z, LuoP, HeQ, et al. Survey on transfer learning research[J]. Journal of Software, 2015, 26:26-39.
17	KanM, WuJ, ShanS, et al. Domain adaptation for face recognition: targetize source domain bridged by common subspace[J]. International Journal of Computer Vision, 2014, 109(1/2):94-109.
18	ZhuangF Z, LuoP, ShenZ Y, et al. Collaborative dual-PLSA: mining distinction and commonality across multiple domains for text classification[C]//Proceedings of the 19th ACM Conference on Information and Knowledge Management. Toronto, Canada: ACM, 2010: 359-368.
19	LongM S, WangJ M, DingG G, et al. Transfer joint matching for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Columbus, USA: IEEE, 2014: 1410-1417.
20	LongM S, WangJ M, DingG G, et al. Transfer learning with graph co-regularization[J]. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(7): 1805-1818.
21	LongM S, WangJ M, DingG G, et al. Adaptation regularization: a general framework for transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(5): 1076-1089.
22	LongM, WangJ, DingG, et al. Transfer feature learning with joint distribution adaptation[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision. Sydney, Australia: IEEE, 2013: 2200-2207.
23	PanS J, TsangI W, KwokJ T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks, 2011, 22(2): 199-210.
24	KingmaD P, WellingM. Auto-encoding variational Bayes[C]//Proceedings of the International Conference on Learning Representations. Ithaca, NY: arXiv:1312.6114v10 [stat.ML], 2014.
25	BodinE, MalikI, EkC H, et al. Nonparametric inference for auto-encoding variational Bayes[J]. arXiv:1712.06536v1 [stat.ML], 2017. ()
26	恩擎. 基于图像自编码的神经网络特征学习研究及应用[D]. 北京: 北京工业大学, 2017.
	EnQ. Research and application on autoencoder based feature learning model of neural network[D]. Beijing: Beijing University of Technology, 2017.
27	虢齐. 基于深度学习的图像生成技术研究与应用[D]. 成都: 电子科技大学, 2017.
	GuoQ. Research and application of image generation based on deep learning[D]. Chengdu: University of Electronic Science and Technology of China, 2017.
28	王守相, 陈海文, 李小平, 等. 风电和光伏随机场景生成的条件变分自动编码器方法[J]. 电网技术, 2018, 42(6): 1860-1867.
	WangS X, ChenH W, LiX P, et al. Conditional variational automatic encoder method for stochastic scenario generation of wind power and photovoltaic system[J]. Power System Technology, 2018, 42(6): 1860-1867.
29	SinaiS, KelsicE, ChurchG M, et al. Variational auto-encoding of protein sequences[J]. arXiv: 1712.03346v3 [q-bio.QM], 2017. ()
30	HsuC C, HwangH T, WuY C, et al. Voice conversion from unaligned corpora using variational autoencoding wasserstein generative adversarial networks[C]//INTERSPEECH. Stockholm, Sweden: ISCA. 2017: 3364-3368.
31	WalkerJ, DoerschC, GuptaA, et al. An uncertain future: forecasting from static images using variational autoencoders[C]//14th European Conference on Computer Vision Proceedings. Part VII.Amsterdam, Netherlands, 2016.

编辑推荐 0

Metrics

阅读次数

全文

1133

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	4	0	0	1129

来源	本网站	其他网站

次数	191	942
比例	17%	83%

摘要

412

最新录用	在线预览	正式出版

0	0	412

来源	本网站	其他网站

次数	260	152
比例	63%	37%

[1]	康飞, 吕伟光, 巨锋, 孙峙. 废锂离子电池放电路径与评价研究[J]. 化工学报, 2023, 74(9): 3903-3911.
[2]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[3]	闫琳琦, 王振雷. 基于STA-BiLSTM-LightGBM组合模型的多步预测软测量建模[J]. 化工学报, 2023, 74(8): 3407-3418.
[4]	尹刚, 李伊惠, 何飞, 曹文琦, 王民, 颜非亚, 向禹, 卢剑, 罗斌, 卢润廷. 基于KPCA和SVM的铝电解槽漏槽事故预警方法[J]. 化工学报, 2023, 74(8): 3419-3428.
[5]	郭雨莹, 敬加强, 黄婉妮, 张平, 孙杰, 朱宇, 冯君炫, 陆洪江. 稠油管道水润滑减阻及压降预测模型修正[J]. 化工学报, 2023, 74(7): 2898-2907.
[6]	李艳辉, 丁邵明, 白周央, 张一楠, 于智红, 邢利梅, 高鹏飞, 王永贞. 非常规服役超临界锅炉的微纳尺度腐蚀动力学模型建立及应用[J]. 化工学报, 2023, 74(6): 2436-2446.
[7]	于源, 陈薇薇, 付俊杰, 刘家祥, 焦志伟. 几何相似涡流空气分级机环形区流场变化规律研究及预测[J]. 化工学报, 2023, 74(6): 2363-2373.
[8]	苏晓丹, 朱干宇, 李会泉, 郑光明, 孟子衡, 李防, 杨云瑞, 习本军, 崔玉. 湿法磷酸半水工艺考察与石膏结晶过程研究[J]. 化工学报, 2023, 74(4): 1805-1817.
[9]	宋冰, 郑城风, 侍洪波, 陶阳, 谭帅. 基于VAE-OCCA的质量相关故障检测方法研究[J]. 化工学报, 2023, 74(4): 1630-1638.
[10]	罗来明, 张劲, 郭志斌, 王海宁, 卢善富, 相艳. 1~5 kW高温聚合物电解质膜燃料电池堆的理论模拟与组装测试[J]. 化工学报, 2023, 74(4): 1724-1734.
[11]	许文烜, 江锦波, 彭新, 门日秀, 刘畅, 彭旭东. 宽速域三种典型型槽油气密封泄漏与成膜特性对比研究[J]. 化工学报, 2023, 74(4): 1660-1679.
[12]	吴心远, 刘奇磊, 曹博渊, 张磊, 都健. Group2vec：基于无监督机器学习的基团向量表示及其物性预测应用[J]. 化工学报, 2023, 74(3): 1187-1194.
[13]	张中秋, 李宏光, 石逸林. 基于人工预测调控策略的复杂化工过程多任务学习方法[J]. 化工学报, 2023, 74(3): 1195-1204.
[14]	张江淮, 赵众. 碳三加氢装置鲁棒最小协方差约束控制及应用[J]. 化工学报, 2023, 74(3): 1216-1227.
[15]	史克年, 郑景元, 钱宇, 杨思宇. 基于马尔可夫链的蒸汽动力系统两阶段随机规划[J]. 化工学报, 2023, 74(2): 807-817.