caa a22 4500 605469296 CHVBK 20210128100321.0 cr unu---uuuuu 210128e20151201xx s 000 0 eng 10.1007/s00500-014-1281-1 doi (NATIONALLICENCE)springer-10.1007/s00500-014-1281-1 A fuzzy-filtered grey network technique for system state forecasting [Elektronische Daten] [Dezhi Li, Wilson Wang, Fathy Ismail] A fuzzy-filtered grey network (FFGN) technique is proposed in this paper for time series forecasting and material fatigue prognosis. In the FFGN, the fuzzy-filtered reasoning mechanism is proposed to formulate fuzzy rules corresponding to different data characteristics; grey models are used to carry out short-term forecasting corresponding to different rules. A novel hybrid training method is proposed to adaptively update model parameters and improve training efficiency. The effectiveness of the developed FFGN is demonstrated by a series of simulation tests. It is also implemented for material fatigue prognosis. Test results show that the developed FFGN predictor can capture data characteristics effectively and forecast data trend accurately. Springer-Verlag Berlin Heidelberg, 2014 Fuzzy-filtered neural networks nationallicence Grey model nationallicence Material fatigue prognosis nationallicence Li Dezhi Department of Mechanical and Mechatronics Engineering, University of Waterloo, N2L 3G1, Waterloo, ON, Canada aut Wang Wilson Department of Mechanical Engineering, Lakehead University, P7B 5E1, Thunder Bay, ON, Canada aut Ismail Fathy Department of Mechanical and Mechatronics Engineering, University of Waterloo, N2L 3G1, Waterloo, ON, Canada aut Soft Computing Springer Berlin Heidelberg 19/12(2015-12-01), 3497-3505 1432-7643 19:12<3497 2015 19 500 https://doi.org/10.1007/s00500-014-1281-1 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s00500-014-1281-1 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Li Dezhi Department of Mechanical and Mechatronics Engineering, University of Waterloo, N2L 3G1, Waterloo, ON, Canada aut NATIONALLICENCE 700 1- Wang Wilson Department of Mechanical Engineering, Lakehead University, P7B 5E1, Thunder Bay, ON, Canada aut NATIONALLICENCE 700 1- Ismail Fathy Department of Mechanical and Mechatronics Engineering, University of Waterloo, N2L 3G1, Waterloo, ON, Canada aut NATIONALLICENCE 773 0- Soft Computing Springer Berlin Heidelberg 19/12(2015-12-01), 3497-3505 1432-7643 19:12<3497 2015 19 500