¹School of Transportation Science and Engineering, Beihang University, Beijing, P.R.China
²College of Vehicle and Motive Power Engineering, Henan University of Science and Technology, Luoyang, P.R.China

---

Abstract—Incipient cavitations identification is very practical and academic significance for cavity research in cooling pump of engine but it is very complicated. In this paper, a Hilbert-Huang transform(HHT) fuzzy wavelet neural network (FWNN) is proposed for incipient cavitations identification. The main incipient cavitations feature was extracted from entrance pressure fluctuation by the HHT. This FWNN uses wavelet basis function as membership function which shape can be adjusted on line so that the networks have better learning and adaptive ability and at the same time combine the wavelet neural network with fuzzy logical theory to deal with complicated nonlinear, uncertain and fuzzy problem. At last the experiment showed that this identification model can provide fast and reliable incipient cavitations identification with minimum assumptions and minimum requirements for modeling skills.

Index Terms—incipient cavitations identification; Hilbert- Huang transform; fuzzy wavelet neural network; wavelet basis function

[PDF]

Cite: LI Li-hong, XU Xiang-yang, LIU Yan-fang, GUO Qian-jin, LI Xiao-li, "HHT Fuzzy Wavelet Neural Network to Identify Incipient Cavitations in Cooling pump of Engine," Journal of Computers vol. 6, no. 3, pp. 506-513, 2011.