A new method using two-tier machine learning is proposed to detect and isolate fault in dynamic distillation process. The residuals between output of network trained by normal condition data and samples are recognized as the threshold for detection. Fault detection is carried out by comparing the deviation between the prediction of one network and the measured value. Once the fault is detected, another network is activated to fit the dynamic distillation process adaptively. When the deviation between simulation output and the measured output of distillation column is less than the threshold, the fitting is considered satisfying. Then the input variables causing output variables’ abnormal fluctuation are found via the analysis of structure parameters of two networks. This method is applied to detect process’s fault and isolate variables relating with fault in the distillation tower simulation, and proved to be effective and veracious.
MAO Hai-tao TIAN Wen-de LIANG Hui-ting
. Fault Detection and Isolation of Dynamic Distillation Process Using Two-tier Machine Learning[J]. The Chinese Journal of Process Engineering, 2017
, 17(2)
: 351
-356
.
DOI: 10.12034/j.issn.1009-606X.216293
[1] Engelien H. K, Skogestad S.Selecting appropriate control variables for a heat-integrated distillation system with prefractionator. Computers & Chemical Engineering[J], 2004, 28(5): 683-691.
[2] Chetouani Y. Model selection and fault detection approach based on Bayes decision theory: Application to changes detection problem in a distillation column. Process Safety and Environmental Protection[J], 2014, 92(3): 215-233.
[3] Xie Lei , Zeng J S, Kruger U, et, al. Fault detection in dynamic systems using the Kullback–Leibler divergence. Control Engineering Practice[J], 2015, 43: 39-48.
[4] Bo C M, Qiao X, Zhang G M, et, al. An integrated method of independent component analysis and support vector machines for industry distillation process monitoring. Journal of Process Control[J], 2010, 20(10): 1133-1140.
[5] Motlaghi S, Jalali F, Ahmadabadi M.N. An expert system design for a crude oil distillation column with the neural networks model and the process optimization using genetic algorithm framework.Expert Systems with Applications[J], 2008, 35(4): 1540-1545.
[6] Liau C K, Yang C K, Tsaib M T. Expert system of a crude oil distillation unit for process optimization using neural networks. Expert Systems with Applications[J], 2004, 26(2): 247-255.
[7] Vanini Z.N, Khorasani K, Meskin N. Fault detection and isolation of a dual spool gas turbine engine using dynamic neural networks and multiple model approach. Information Sciences[J], 2014, 259: 234-251.
[8] Qu Y, Li Z M, Li E.Fault detection and diagnosis for non-Gaussian stochastic distribution systems with time delays via RBF neural networks. ISA Transactions[J], 2012, 51(6): 786-791.
[9] Tian W D, Sun S L, Guo Q J. Fault detection and diagnosis for distillation column using two-tier model. The Canadian Journal of Chemical Engineering[J] ,2013, 91: 1671-1685.
[10]Kamal H, Meisam P, Hosein M M. Fault diagnosis and classification based on wavelet transform and neural network. Progress in Nuclear Energy[J], 2011, 53(1): 41-47.
[11]Gorman C, Valova I.Optimizations of the Gravitationally Organized Related Mapping ANN through Genetic Algorithms. Procedia Computer Science[J], 2014, 36: 529-534.
[12] Leonard E B, Ted P. statistical inference for probabillstic function of finite state Markov chains. The Annals of Mathematical Statistics[J], 1966, 37(6): 1554-1563.
