467436 Model-Based Fault Detection for Nonlinear Process Systems
This work presents a model-based fault detection methodology for nonlinear process systems. The objective is to detect faults by estimating the model parameters online by minimising the error between model predictions and the observed plant data. Algorithms for parameter estimation of dynamic systems require integration of the ordinary differential equations (ODEs), which is performed using artificial neural network (ANN) in this work [5]. The parameters are estimated through an augmented optimization problem where the objective is to simultaneously solve the ODEs as well as estimate the parameters. The results are compared with the parameters of the observed plant data under fault-free situation. The estimated parameters should be close to observed parameters when no fault is present. Any substantial discrepancy indicates changes in the process and may be interpreted as a fault if the discrepancy goes above a certain threshold. One key issue in fault detection is to estimate the model parameters precisely (accuracy) and as fast as possible (speed). Accuracy is important to avoid false-positives, while speed ensures that the faults and their location are identified quick enough to be able to take corrective actions in a timely manner. It is shown that this issue can be addressed by using the proposed approach. Details of the proposed approach are illustrated for an isothermal continuous stirred tank reactor (CSTR) where it is assumed that the faults can be represented by changes in values of reaction rate constants. To demonstrate the proposed approach, different scenarios are presented: a fault-free situation and some faulty situations. The scenarios considered show the effectiveness of the proposed approach.
References
[1] V. Venkatasubramanian, R. Rengaswamy, and K. Yin, “A Review of Process Fault Detection and Diagnosis Part I : Quantitative Model-Based Methods,” Comput. Chem. Eng., vol. 27, pp. 293–311, 2003.
[2] R. Isermann, “Model-Based Fault-Detection and Diagnosis - Status and Applications,” Annu. Rev. Control, vol. 29, pp. 71–85, 2005.
[3] P. Mhaskar, A. Gani, N. H. El-farra, C. Mcfall, P. D. Christofides, and J. F. Davis, “Integrated Fault-Detection and Fault-Tolerant Control of Process Systems,” Am. Inst. Chem. Eng., vol. 52, no. 6, pp. 2129–2148, 2006.
[4] I. Hwang, S. Kim, Y. Kim, and C. Seah, “A Survey of Fault Detection, Isolation, and Reconfiguration Methods,” IEEE Trans. Control Syst. Technol., vol. 18, no. 3, pp. 636–653, 2010
[5] V. Dua and P. Dua, “A Simultaneous Approach for Parameter Estimation of a System of Ordinary Differential Equations, Using Artificial Neural Network Approximation,” Ind. Eng. Chem. Res., vol. 51, no. 4, pp. 1809–1814, 2011.
See more of this Group/Topical: Computing and Systems Technology Division