Study On corrosion Acoustic Emission Separation Based On Blind Source Separation

摘要

When corrosion signals of tank bottom is detected by online method,it is essential to identify corrosion signals of different corrosion pots.It is a new method based on blind source separation.Blind source separation has produced many arithetics,among which entropy maximization is more mature.The aim of this paper is to separate corrosion signals by using entropy maximization arithmetic.Furthermore,the separation of linear mixed acoustic emission signals is achieved through simulation.The results indicate that blind source separation is an effective method for the separation of corrosion signals.