A neural network approach for the real-time detection of faults. (English) Zbl 1365.92008
Summary: Fault detection is an essential part of the operation of any chemical plant. Early detection of faults is important in chemical industry since a lot of damage and loss can result before a fault present in the system is detected. Even though fault detection algorithms are designed and implemented for quickly detecting incidents, most these algorithms do not have an optimal property in terms of detection delay with respect to false alarm rate. Based on the optimization property of cumulative sum (CUSUM), a real-time system for detecting changes in dynamic systems is designed in this paper. This work is motivated by combining two fault detection (FD) strategies; a simplified procedure of the incident detection problem is formulated by using both the artificial neural networks (ANN) and the CUSUM statistical test (Page-Hinkley test). The design of a model-based residual generator is intended to reveal any drift from the normal behavior of the process. In order to obtain a reliable model for the normal process dynamics, the neural black-box modeling by means of a nonlinear auto-regressive with eXogenous input (NARX) model has been chosen in this study. This paper also shows the choice and the performance of the neural network in the training and test phases. After describing the system architecture and the proposed methodology of the fault detection, we present a realistic application in order to show the technique’s potential. The purpose is to develop and test the fault detection method on a real incident data, to detect the change presence, and pinpoint the moment it occurred. The experimental results demonstrate the robustness of the FD method.
MSC:
| 92B20 | Neural networks for/in biological studies, artificial life and related topics |
| 92E99 | Chemistry |
| 62N05 | Reliability and life testing |
References:
| [1] | Adankon MM, Cheriet M (2007) Optimizing resources in model selection for support vector machine, Pattern Recognit 40:953-963 · Zbl 1119.68143 · doi:10.1016/j.patcog.2006.06.012 |
| [2] | Basseville B (1986) On line detection of jumps in mean. Lect Notes Contr Inf Sci 77:12-26 |
| [3] | Billings SA, Voon WSF (1986) Correlation based model validity tests for nonlinear models. Int J Control 44:235-244 · Zbl 0597.93058 · doi:10.1080/00207178608933633 |
| [4] | Cammarata L, Fichera A, Pagano A (2002) Neural prediction of combustion instability. Appl Energy 72:513-528 · doi:10.1016/S0306-2619(02)00024-7 |
| [5] | Chen S, Billings SA (1989) Representation of nonlinear systems—The NARMAX model. Int J Control 49:1013-1032 · Zbl 0674.93009 |
| [6] | Cheng C-S, Cheng S-S (2001) A neural network-based procedure for the monitoring of exponential mean. Comput Ind Eng 40:309-321 · doi:10.1016/S0360-8352(01)00031-6 |
| [7] | Cheng S, Shih FY (2007) An improved incremental training algorithm for support vector machines using active query. Pattern Recognit 40:964-971 · Zbl 1119.68148 · doi:10.1016/j.patcog.2006.06.016 |
| [8] | Chetouani Y (2004) Fault detection by using the innovation signal: application to an exothermic reaction. Chem Eng Process 43:1579-1585 · doi:10.1016/j.cep.2004.02.002 |
| [9] | Chetouani Y (2006a) Application of the generalized likelihood ratio test for detecting changes in a chemical reactor. Process Saf Environ Protect 84:371-377 · doi:10.1205/psep05013 |
| [10] | Chetouani Y (2006b) Fault detection in a chemical reactor by using the standardized innovation. Process Saf Environ Protect 84:27-32 · doi:10.1205/psep.04285 |
| [11] | Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Math Control Signals Syst 4:303-312 · Zbl 0679.94019 · doi:10.1007/BF02551274 |
| [12] | Dash S, Venkatasubramanian V (2000) Challenges in the industrial applications of fault diagnostic systems. Comput Chem Eng 24:785-791 · doi:10.1016/S0098-1354(00)00374-4 |
| [13] | Engell S, Fernholz, G (2003) Control of a reactive separation process. Chem Eng Process 42:201-210 · doi:10.1016/S0255-2701(02)00089-2 |
| [14] | Ferentinos KP (2005) Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms. Neural Netw 18:934-950 · doi:10.1016/j.neunet.2005.03.010 |
| [15] | Fouladirad M, Nikiforov I (2005) Optimal statistical fault detection with nuisance parameters. Automatica 41:1157-1171 · Zbl 1116.62111 · doi:10.1016/j.automatica.2005.02.004 |
| [16] | Fung EHK, Wong YK, Ho HF, Mignolet MP (2003) Modelling and prediction of machining errors using ARMAX and NARMAX structures. Appl Math Model 27:611-627 · Zbl 1040.62113 · doi:10.1016/S0307-904X(03)00071-4 |
| [17] | Gertler JJ (1998) Fault detection and diagnosis in engineering systems. Marcel Dekker Inc, New York |
| [18] | Hertz J, Krogh A, Palmer RG (1991) Introduction to the theory of neural computation. Addison-Wesley, Redwood City |
| [19] | Himmelblau DM (2000) Applications of artificial neural networks in chemical engineering. Korean J Chem Eng 17:373-392 · doi:10.1007/BF02706848 |
| [20] | Hinkley DV (1971) Inference about the change-point from cumulative sum tests. Biometrika 58:509-523 · Zbl 0254.62019 · doi:10.1093/biomet/58.3.509 |
| [21] | Huang L-L, Shimizu A (2006) A multi-expert approach for robust face detection. Pattern Recognit 39:1695-1703 · Zbl 1096.68717 · doi:10.1016/j.patcog.2005.11.020 |
| [22] | Huang Y, Reklaitis GV, Venkatasubramanian V (2000) Dynamic optimization based fault accommodation. Comput Chem Eng 24:439-444 · doi:10.1016/S0098-1354(00)00435-X |
| [23] | Isermann R (2005) Model-based fault-detection and diagnosis—status and applications. Annu Rev Control 29:71-85 · doi:10.1016/j.arcontrol.2004.12.002 |
| [24] | Kinnaert M, Vrancic D, Denolin E, Juricic D, Petrovcic J (2000) Model-based fault detection and isolation for a gas-liquid separation unit. Control Eng Pract 8:1273-1283 · doi:10.1016/S0967-0661(00)00064-2 |
| [25] | Liu C (2003) A Bayesian discriminating features method for face detection. IEEE Trans Pattern Anal Mach Intell 25:725-740 · doi:10.1109/TPAMI.2003.1201822 |
| [26] | Ljung L (1999) System identification, theory for the user. Prentice-Hall, Englewood Cliffs |
| [27] | Lorden G (1971) Procedures for reacting to a change in distribution. Annu Math Stat 42:1897-1908 · Zbl 0255.62067 · doi:10.1214/aoms/1177693055 |
| [28] | Luh G-C., Cheng W-C (2005) Immune model-based fault diagnosis. Math Comput Simul 67:515-539 · Zbl 1180.92041 · doi:10.1016/j.matcom.2004.07.004 |
| [29] | Moatar F, Fessant F, Poirel A (1999) pH modelling by neural networks: application of control and validation data series in the Middle Loire river. Ecol Model 120:141-156 · doi:10.1016/S0304-3800(99)00098-8 |
| [30] | Mu J, Rees D, Liu GP (2005) Advanced controller design for aircraft gas turbine engines. Control Eng Pract 13:1001-1015 · doi:10.1016/j.conengprac.2004.11.001 |
| [31] | Nanayakkara VK, Ikegami Y, Uehara H (2002) Evolutionary design of dynamic neural networks for evaporator control. Int J Refrig 25:813-826 · doi:10.1016/S0140-7007(01)00090-1 |
| [32] | Narendra KS, Parthasarathy K (1990) Identification and control of dynamical systems using neural networks. IEEE Trans Neural Netw 1:4-21 · doi:10.1109/72.80202 |
| [33] | Nguyen D, Widrow B (1990) Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights. Int Jt Conf Neural Netw 3:21-26 · doi:10.1109/IJCNN.1990.137819 |
| [34] | Patton RJ, Frank PM, Clark RN (2000) Issues of fault diagnosis for dynamic systems. Springer, Berlin |
| [35] | Pollak M, Siegmund D (1985) A diffusion process and its application to detecting a change in the drift of a Brownian motion. Biometrika 72:267-280 · Zbl 0571.60084 · doi:10.1093/biomet/72.2.267 |
| [36] | Previdi F (2002) Identification of black-box nonlinear models for lower limb movement control using functional electrical stimulation. Control Eng Pract 10:91-99 · doi:10.1016/S0967-0661(01)00128-9 |
| [37] | Qin SJ, McAvoy TJ (1996) Nonlinear fir modeling via a neural net PLS approach. Comput chem Eng 20:147-159 · doi:10.1016/0098-1354(95)00011-P |
| [38] | Ragot J, Darouach M, Maquin D, Bloch G (1990) Validation de données et diagnostic. Hermès, Paris |
| [39] | Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323:533-536 · Zbl 1369.68284 · doi:10.1038/323533a0 |
| [40] | Scenna N, Benz S, Drozdowicz B, Lamas E (2000) A diagnosis system for fault diagnosis in batch distillation columns, ESCAPE10. Comput Aided Process Eng 8:805-810 · doi:10.1016/S1570-7946(00)80136-4 |
| [41] | Schneiderman H, Kanade T (2004) Object detection using the statistic of parts. Int J Comput Vis 56:151-177 · doi:10.1023/B:VISI.0000011202.85607.00 |
| [42] | Sharma R, Singh K, Singhal D, Ghosh R (2004) Neural network applications for detecting process faults in packed towers. Chem Eng Process 43:841-847 · doi:10.1016/S0255-2701(03)00103-X |
| [43] | Skogestad S (2003) Self-optimizing control: the missing link between steady state optimization and control. Comput Chem Eng 24:569-575 · doi:10.1016/S0098-1354(00)00405-1 |
| [44] | Subasi A (2005) Automatic recognition of alertness level from EEG by using neural network and wavelet coefficients. Expert Syst Appl 28:701-711 · doi:10.1016/j.eswa.2004.12.027 |
| [45] | Subasi A, Erçelebi E (2005) Classification of EEG signals using neural network and logistic regression. Comput Methods Programs Biomed 78:87-99 · doi:10.1016/j.cmpb.2004.10.009 |
| [46] | Thiria S, Lechevalier Y, Gascuel O, Canu S (1997) Statistique et méthodes neuronales. Dunod, Paris |
| [47] | Venkatasubramanian V, Rengaswamy R, Yin K, Kavuri SN (2003a) A review of process fault detection and diagnosis: Part I: Quantitative model-based methods. Comput Chem Eng 27:293-311 · doi:10.1016/S0098-1354(02)00160-6 |
| [48] | Venkatasubramanian V, Rengaswamy K, Kavuri SN (2003b) A review of process fault detection and diagnosis: Part II: Qualitative models and search strategies. Comput Chem Eng 27:313-326 · doi:10.1016/S0098-1354(02)00161-8 |
| [49] | Venkatasubramanian V, Rengaswamy K, Kavuri SN, Yin K (2003c) A review of process fault detection and diagnosis: Part III: Process history based methods. Comput Chem Eng 27:327-346 · doi:10.1016/S0098-1354(02)00162-X |
| [50] | Vogl TP, Mangis JK, Rigler AK, Zink WT, Alkon DL (1988) Accelerating the convergence of the backpropagation method. Biol Cybern 59:256-264 · doi:10.1007/BF00332914 |
| [51] | Wang H, Oh Y, Yoon, E (1998) Strategies for modeling and control of nonlinear chemical processes using neural networks. Comput Chem Eng 22:823 · doi:10.1016/S0098-1354(98)00157-4 |
| [52] | Warnes MR, Glassey J, Montague GA, Kara B (1996) On data-based modelling techniques for fermentation processes. Process Biochem 31:147-155 · doi:10.1016/0032-9592(95)00043-7 |
| [53] | Yang MH, Kriegman D, Ahuja N (2001) Face detection using multimodal density models. Comput Vis Image Underst 84:264-284 · Zbl 1033.68614 · doi:10.1006/cviu.2001.0937 |
| [54] | Zaknich A (2003) Neural networks for intelligent signal processing. World Scientific, Singapore · Zbl 1029.68123 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
