A hybrid ANN-ES system for dynamic fault diagnosis of hydrocracking process

Abstract In this paper, the wavelet-sigmoid basis neural network(WSBN) proposed by authors earlier is integrated with expert system(ES) for dynamic fault diagnosis(DFD). Meanwhile, a two-level integration strategy(TLIS) for DFD is also presented. At the first level of TLIS, a WSBN detects the fault sources through data from the dynamic processes, and outputs the corresponding fault degrees. At the second level, ES interprets the results from the WSBN and predicts the time period for fully developing the faults and the product qualities as well as the reactor states at the moment when the fault is fully developed by using a dynamic simulation package. If the predicted values exceed their acceptable ranges, then ES gives some related proposals for removing the fault causes or canceling their effects by using the reasoning context consisted of production rules. DFD is applied to a hydrocracking process showing the efficiency of the TLIS.

[1]  Venkat Venkatasubramanian,et al.  Model-based reasoning in diagnostic expert systems for chemical process plants , 1987 .

[2]  David M. Himmelblau,et al.  Fault detection in a single-stage evaporator via parameter estimation using the Kalman filter , 1987 .

[3]  Bhavik R. Bakshi,et al.  Wave‐net: a multiresolution, hierarchical neural network with localized learning , 1993 .

[4]  Qinghua Zhang,et al.  Wavelet networks , 1992, IEEE Trans. Neural Networks.

[5]  J.J. Gertler,et al.  Survey of model-based failure detection and isolation in complex plants , 1988, IEEE Control Systems Magazine.

[6]  Ware Myers.  Interview: Elane Rich of MCC , 1990 .

[7]  Jim Klein,et al.  Diagnostic model processor: Using deep knowledge for process fault diagnosis , 1990 .

[8]  Venkat Venkatasubramanian,et al.  Representing bounded fault classes using neural networks with ellipsoidal activation functions , 1993 .

[9]  A. E. Farell,et al.  Framework for enhancing fault diagnosis capabilities of artificial neural networks , 1994 .

[10]  J. A. Leonard,et al.  Radial basis function networks for classifying process faults , 1991, IEEE Control Systems.

[11]  P. A. Clark,et al.  A distributed fault diagnosis method based on digraph models : steady-state analysis , 1993 .

[12]  Venkat Venkatasubramanian,et al.  A neural network methodology for process fault diagnosis , 1989 .

[13]  Peter L. Lee,et al.  An integrated neural network/expert system approach for fault diagnosis , 1993 .

[14]  Bernard Delyon,et al.  Accuracy analysis for wavelet approximations , 1995, IEEE Trans. Neural Networks.

[15]  Mark A. Kramer,et al.  Diagnosis using backpropagation neural networks—analysis and criticism , 1990 .