Research On Blind Source Separation Of Multi-mixed Vibration Signal Of Aircraft Engine

Modern signal processing methods can't separate the complicated multi-mixed vibration signals effectively without any pre-information. As one of the hottest and most exciting topics in the fields of signal processing at present, blind source sepation (BSS) is paied attention for solving this problem. But the researches are mostly about the theory and simulation, little in in the real signal application.The major work of this paper is to research the existing methods of BSS then get some improved methods fitted for separating the vibration signals, which will provide a new way for the state stiring and diagnosis of aircraft engines. The specific work as flowering:(1). Analysed and explained the independent component analysis (ICA) theory and algorithm in detail; gave the class of core function and a way of its selection.(2). Proposed an improved algorithm called ICA with reference (ICA-R), which is combined with characters of the vibration signal to pursuit the expected one. Through simulation analysis, this technique has not only more separating precision but also faster convergence than traditional methods.(3). Analysed sorts of vibration experiments using BSS and achieved some available conclusions and adaptive methods, which was successful to confirm the validity of BSS applied to separating vibration signal.(4). A BSS package using MATLAB and VC++ softwares was designed based on the theory of setting parameters. This package made the operation of BSS simulations and applications conveniently under a graphical interfaces control.