Identyfikatory
Warianty tytułu
Detekcja i lokalizacja uszkodzeń procesów dynamicznych z użyciem sieci rekurencyjnych
Języki publikacji
Abstrakty
The paper focuses on the problem of fault detection and isolation for dynamic processes using selected recurrent neural networks. The main objective is to show how to employ some discoveries of the chaos theory for modeling processes by means of globally and locally recurrent neural networks. Both types of neural models are used in fault detection and isolation block. The performance of the FDI system is examined using two types of neural models: Jordan/Elman tower neural networks and networks with dynamic neural units. The paper contains numerical examples that illustrate the merits and limits of these two approaches.
Treść artykuł wiąże się z problemem detekcji i lokalizacji uszkodzeń dla szerokiej gamy procesów dynamicznych z użyciem wybranych rekurencyjnych sieci neuronowych. Głównym celem jest pokazanie w jaki sposób mogą zostać zastosowane niektóre z odkryć teorii chaosu do modelowania procesów z użyciem globalnych i lokalnych struktur neuronowych. Oba typy modeli neuronowych zostały użyte w bloku detekcji i lokalizacji uszkodzeń. Sprawność układu diagnostycznego porównana została dla modeli procesów z zastosowaniem: sieci wielo-kontekstowych Jordana/Elmana i sieci z neuronami dynamicznymi. W artykule zamieszczono przykłady numeryczne wskazujące na zalety i wady obu podejść.
Czasopismo
Rocznik
Tom
Strony
33--40
Opis fizyczny
Bibliogr. 24 poz., rys., tab.
Twórcy
autor
- Silesian University of Technology, Department of Fundamentals of Machinery Design, 18a Konarskiego Str., 44-100 Gliwice, pp@polsl.pl
Bibliografia
- [1] Ayoubi M.: Nonlinear dynamic systems identification with dynamic neural networks for fault diagnosis in technical processes. Humans, Information and Technology, pages 2120-2125, October 1994.
- [2] Blanke M., Kinnaert M., Lunze J. and Staroswiecki M.: Diagnosis and fault-tolerant control. Springer-Verlag Berlin Heidelberg, 2003.
- [3] Caccavale F. and Villani L.: Fault Diagnosis and Fault Tolerance for Mechatronic Systems: Recent Advances. Springer Tracts in Advanced Robotics. Springer Berlin / Heidelberg, 2003.
- [4] Elman J.: Finding structure in time. Cognitive Science, 14(2):179-211, 1990.
- [5] Jordan M.: Serial Order: A Parallel Distributed Processing Approach. Technical report. California Univ., San Diego, La Jolla. Inst. For Cognitive Science, 1986.
- [6] De Jesus O. and Hagan M.: Backpropagation Algorithms for a Broad Class of Dynamic Networks. IEEE Transactions on Neural Networks, pages 14-27, January 2007.
- [7] Kamiya A., Seppo J., Rajkumar R. and Shigenobu K.: Fusion of soft computing and hard computing for large-scale plants: a general model. Applied Soft Computing, 5(3):265-279, 2005.
- [8] Korbicz J.: Robust fault detection using analytical and soft computing methods. Bulletin of the Polish Academy of Sciences : Technical Sciences, 54(1):75-88, 2006.
- [9] Korbicz J., Kościelny J., Kowalczuk Z. and Cholewa W. (Eds.): Fault Diagnosis. Models, Artificial Intelligence, Applications. Springer Berlin / Heidelberg, 2004.
- [10] Patton R., Frank P. and Robert C.: Issues of Fault Diagnosis for Dynamic Systems. Springer-Verlag Berlin and Heidelberg, 2000.
- [11] Patan K.: Stability Analysis and the Stabilization of a Class of Discrete-Time Dynamic Neural Networks, IEEE Transactions on Neural Networks, May 2007.
- [12] Patan K.: Approximation of state-space trajectories by locally recurrent globally feed-forward neural networks. Neural Networks, pages doi:10.1016/j.neunet.2007.10.004, 2007.
- [13] Przystałka P.: Model-based fault detection and isolation using locally recurrent neural networks. Proceedings of the 9th International Conference on Artificial Intelligence and Soft Computing, Zakopane, Poland, June 22-26, 2008, (to be printed).
- [14] Przystałka P.: Hybrid learning algorithm for locally recurrent neural networks. Fault diagnosis and fault tolerant control, Academic Publishing House EXIT, pages 255-262, September 2007.
- [15] Przystałka P.: Heuristic modeling using recurrent neural networks: simulated and real-data experiments. Computer Assisted Mechanics and Engineering Sciences, 14(4):715-727, November 2007.
- [16] Przystałka, P.: Heuristic modeling of objects and processes using dynamic neural networks. Diagnostyka, 2/38:15–18, 2006.
- [17] Schlang M., Lang B., Poppe T., Runkler T. and Weinzierl K.: Current and future development in neural computation in steel processing. Control Engineering Practice, 9(9):975-986, 2001.
- [18] Tomanek A., Przystałka P. and Wyczółkowski R.: Optimization of Jordan and Elman neural networks through distributed computing environment. Methods of Artificial Intelligence, pages 69-70, November 2007.
- [19] William W.: A Comparison of Architectural Alternatives for Recurrent Networks. Proceedings of the fourth Australian Conference on Neural Networks, pages 189-192, February 1993.
- [20] William W.: Stability of learning in classes of recurrent and feedforward networks. Proceedings of the sixth Australian Conference on Neural Networks, pages 142-145, February 1995.
- [21] Xiaohua Y., Zhifeng Y., Xinan Y. and Jianqiang L.: Chaos gray-coded genetic algorithm and its application for pollution source identifications in convection diffusion equation. Elsevier B.V. pp. 1676-1688, 2008.
- [22] Yao D., Zeng M. and Yongjie Li: Chaotic simulated annealing algorithm applied to ERP dipole localization. Proceedings of the International Conference on Communications, Circuits and Systems, pp 908-911, May 2005.
- [23] Zadeh, Lotfi A.: Fuzzy Logic, Neural Networks and Soft Computing. Communications of the ACM, 37(3):77-84, March 1994.
- [24] Zhong L., Halang W. and Chen G.: Integration of Fuzzy Logic and Chaos Theory. Studies in Fuzziness and Soft Computing. Springer-Verlag Berlin / Heidelberg, 2006.
Typ dokumentu
Bibliografia
Identyfikator YADDA
bwmeta1.element.baztech-article-BAR0-0039-0034