| Mine fan system for mine work personnel to the existence of the lifeline,has received more and more attention.Revolves around the mine fan fault diagnosis technology research,this paper puts forward a kind of based on blind source separation and fractal fault diagnosis system of the theory of the new architecture,and based on this architecture,focusing on the mine fan fault signal blind source separation and the fault signal feature extraction had in-depth discussions about the problem of these two aspects.The core idea and the innovation point of this article below summarized as follows:(1)In view of the traditional fan fault diagnosis system in mixed vibration signal,the defect in the diagnosis of low accuracy this paper proposes a new fault diagnosis system architecture.Blind source separation algorithm,introduced the new architecture,the fault signal and interference signal is separated from the mixed signal.At the same time,the new architecture using multiple sets of monitoring data of parallel processing,finally the results in fusion processing,further improve the performance of the system of fault diagnosis.(2)In view of the mine fan fault signal separation problem of low signal-to-noise ratio,non-stationary a based on eigenvalue decomposition is proposed and the time-frequency distribution of blind source separation method.This method firstly divided by eigenvalue decomposition signal subspace and noise subspace,the observation signal onto the signal subspace,so as to achieve noise reduction purposes.And then put forward through joint diagonalization of a set of projection signal of time-frequency distribution matrix to realize the non-stationary signal blind separation.(3)Propose a fan fault feature extraction based on fractal theory and diagnosis methods,and design based on blind source separation and fan fault simulation platform of fractal algorithm,can be a good simulation of the fan fault related situation,to facilitate the fault types of judgment. |
PDF Full Text Request