Active Learning Intelligent Soft Sensor based on Probability Selection

doi:10.16182/j.issn1004731x.joss.20-0093

Abstract

Abstract: Aiming at lack of tag samples and high cost of sampling tags in complex industrial processes, an active learning algorithm based on probability selection is proposed. Firstly, unlabeled samples are performed subspace integration by using the principal component analysis. Then, the information of unlabeled samples is evaluated by the uncertainty, which is calculated based on the out put of all sub learners. And the most valuable samples are selected to mark manually. Finally, the function of unlabeled samples and labeled samples are analyzed, and the termination conditions are designed by introducing the performance index of training set. Through simulations of industrial processes data, it is verified that the proposed method can improve the accuracy of the model while reducing the cost of marking.

Key words: probability selection, soft sensor, sub learners, uncertainty, termination conditions

CLC Number:

TP274

Dai Xuezhi, Xiong Weili. Active Learning Intelligent Soft Sensor based on Probability Selection[J]. Journal of System Simulation, 2021, 33(6): 1350-1357.

References

[1] Fortuna L, Graziani S, Rizzo A, et al.Soft Sensors for Monitoring and Control of Industrial Processes[M]. Berlin: Springer Science & Business Media, 2007.
[2] Wang H Z, Yu J S.Studies on Soft Sensor Modeling Using Mixtures of Kernels PCR[J]. Control and Instruments in Chemical Industry (S1000-3932), 2005, 32(2): 23-25.
[3] Shao W M, Tian X M.Adaptive Soft Sensor for Quality Prediction of Chemical Processes based on Selective Ensemble of Local Partial Least Squares Models[J]. Chemical Engineering Research and Design (S0263-8762), 2015, 95: 113-132.
[4] 汤健, 柴天佑, 余文, 等. 在线KPLS建模方法及在磨机负荷参数集成建模中的应用[J]. 自动化学报, 2013, 39(5): 471-486.
Tang Jian, Chai Tianyou, Yu Wen, et al.On-line KPLS Algorithm with Application to Ensemble Modeling Parameters of Mill Load[J]. Acta Automatica Sinica, 2013, 39(5): 471-486.
[5] 郑小霞, 钱锋. 基于PCA和最小二乘支持向量机的软测量建模[J]. 系统仿真学报, 2006, 18(3): 217-219.
Zheng Xiaoxia, Qian Feng.Soft Sensor Modeling Based on PCA and Support Vector Machines[J]. Journal of System Simulation, 2006, 18(3): 217-219.
[6] Zhang S, Wang F, He D, et al.Online Quality Prediction for Cobalt Oxalate Synthesis Process using Least Squares Support Vector Regression Approach with Dual Updating[J]. Control Engineering Practice (S0967-0661), 2013, 21(10): 1267-1276.
[7] Zhang Z, Wang T, Liu X.Melt Index Prediction by Aggregated RBF Neural Networks Trained with Chaotic Theory[J]. Neurocomputing (S0925-2312), 2014, 131(9): 368-376.
[8] 杨尔辅, 周强, 胡益锋, 等. 基于PCA-RBF神经网络的工业烈解炉收率在线预测软测量方法[J]. 系统仿真学报, 2001, 13(增1): 194-197.
Yang Erfu, Zhou Qiang, HU Yifeng, et al.A Soft-sensing Approach to On-line Predict the Yields of Industrial Pyrolysis Furnace Based on PCA-RBF Neural Networks[J]. Journal of System Simulation, 2001, 13(S1): 194-197.
[9] 何志昆, 刘光斌, 赵曦晶, 等. 高斯过程回归方法综述[J]. 控制与决策, 2013(8): 1121-1129.
He Zhikun, Liu Guangbin, Zhao Xinjing, et al.Overview of Gaussian Process Regression[J]. Control and Decision, 2013(8): 1121-1129.
[10] 曲昭伟, 吴春叶, 王晓茹. 半监督自训练的方面提取[J]. 智能系统学报, 2019, 14(4): 635-641.
Qu Zhaowei, Wu Chunye, Wang Xiaoru.Aspects Extraction based on Semi-supervised Self-training[J]. CAAI Transactions on Intelligent Systems, 2019, 14(4): 635-641.
[11] Zhou Z H, Li M.Semi-supervised Regression with Cotraining-style Algorithms[J]. IEEE Transactions on Knowledge & Data Engineering (S1041-4347), 2007, 19(11): 1479-1493.
[12] Deng J, Xie L, Chen L, et al.Development and Industrial Application of Soft Sensors with Online Bayesian Model Updating Strategy[J]. Journal of Process Control (S0959-1524), 2013, 23(3): 317-325.
[13] Huang G, Song S, Gupta J N D, et al. Semi-Supervised and Unsupervised Extreme Learning Machines[J]. IEEE Transactions on Cybernetics (S1083-4419), 2017, 44(12): 2405-2417.
[14] Cohn D A, Ghahramani Z, Jordan M I.Active Learning with Statistical Models[J]. Journal of Artificial Intelligence Research (S1076-9757), 1996, 4(1): 705-712.
[15] Zhou Z H, Li M.Semi-supervised Learning by Disagreement[J]. Knowledge and Information Systems (S 0219-1377), 2010, 24(3): 415-439.
[16] Tang Q, Li D, Xi Y.A New Active Learning Strategy for Soft Sensor Modeling based on Feature Reconstruction and Uncertainty Evaluation[J]. Chemometrics and Intelligent Laboratory Systems (S0169-7439), 2018, 172: 43-51.
[17] Ge Z Q.Active Learning Strategy for Smart Soft Sensor Development under a Small Number of Labeled Data Samples[J]. Journal of Process Control (S0959-1524), 2014, 24(9): 1454-1461.
[18] Ge Z Q.Active Probabilistic Sample Selection for Intelligent Soft Sensing of Industrial Processes[J]. Chemometrics and Intelligent Laboratory Systems (S0169-7439), 2016, 151: 181-189.
[19] Shi X D, Xiong W L.Approximate Linear Dependence Criteria with Active Learning for Smart Soft Sensor Design[J]. Chemometrics and Intelligent Laboratory Systems (S0169-7439), 2018, 180: 88-95.
[20] Douak F, Melgani F, Benoudjit N.Kernel Ridge Regression with Active Learning for Wind Speed Prediction[J]. Applied Energy (S0306-2619), 2013, 103(5): 328-340.
[21] 熊伟丽, 张伟, 徐保国. 一种基于EGMM的高斯过程回归软测量建模[J]. 信息与控制, 2016, 45(1): 14-19.
Xiong Weili, Zhang Wei, Xu Baoguo.A Soft Sensor Modeling Method Based on EGMM Using Gaussian Process Regression[J]. Information and Control, 2016, 45(1): 14-19.
[22] Ge Z Q, Song Z H.Subspace Partial Least Squares Model for Multivariate Spectroscopic Calibration[J]. Chemometrics and Intelligent Laboratory Systems (S0169-7439), 2013, 125(Complete): 51-57.
[23] Douaka F, Melgania F, Alajlanc N, et al.Active Learning for Spectroscopic Data Regression[J]. Journal of Chemometrics (S0886-9383), 2012, 26(7): 374-383.