Neuro-fuzzy-based fault detection of the air flow sensor of an idling gasoline engine

Abstract This paper presents a neuro-fuzzy-based diagnostic system for detecting the faults of an air flow sensor of an idling gasoline engine. Based on the Takagi-cSugeno fuzzy system model, the diagnostic system is formulated by the input-output relationships between symptoms and faults. The system parameters are regulated with the learning datum from experiments, using the steepest descent method and back-propagation algorithm. The proposed diagnostic system consists of two parts: one is to judge the fault of the sensor and the other is to identify the bias degree of the sensor. The experimental results show that the fault source and fault (bias) degree can be identified with the proposed diagnostic system, and indicate that the neuro-fuzzy strategy is an efficient and available approach for fault diagnosis problems of the gasoline engine system.

[1]  Chin-Teng Lin,et al.  A neural fuzzy control system with structure and parameter learning , 1995 .

[2]  Hsing-Chia Kuo,et al.  Neural-fuzzy fault diagnosis in a marine propulsion shaft system , 2002 .

[3]  Ada Fort,et al.  Soft fault detection and isolation in analog circuits: some results and a comparison between a fuzzy approach and radial basis function networks , 2002, IEEE Trans. Instrum. Meas..

[4]  Mattias Nyberg,et al.  Model based diagnosis of the air path of an automotive diesel engine , 2001 .

[5]  Yan Shi,et al.  An improvement of neuro-fuzzy learning algorithm for tuning fuzzy rules , 2001, Fuzzy Sets Syst..

[6]  Mattias Nyberg.  Automatic design of diagnosis systems with application to an automotive engine , 1999 .

[7]  Giorgio Rizzoni,et al.  Fault detection and isolation for an experimental internal combustion engine via fuzzy identification , 1995, IEEE Trans. Control. Syst. Technol..

[8]  Mattias Nyberg,et al.  Using hypothesis testing theory to evaluate principles for leakage diagnosis of automotive engines , 2003 .

[9]  Giorgio Rizzoni,et al.  Diagnosis of an automotive emission control system using fuzzy inference , 1997 .

[10]  Sheng Zhang,et al.  Study of machine fault diagnosis system using neural networks , 2002, Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290).

[11]  Yan Shi,et al.  A new approach of neuro-fuzzy learning algorithm for tuning fuzzy rules , 2000, Fuzzy Sets Syst..

[12]  Zdzislaw Kowalczuk,et al.  Model based diagnosis for automotive engines-algorithm development and testing on a production vehicle , 1995, IEEE Trans. Control. Syst. Technol..

[13]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.