Research on localization of Acoustic Emission source based on algebraic neural network and chaotic features

Due to defects of time-difference of arrival localization which influenced by speed differences of various model waveforms and waveform distortion in transmitting process, a neural network technique was introduced to calculate localization of the Acoustic Emission(AE) source. In order to overcome the shortcomings of the traditional BP algorithm such as long training time and low accuracy, we propose the concept of algebraic neural network and introduce the algebraic algorithm in the network training phase which transforms the complex nonlinear optimization problem to a set of simple algebraic equations and achieves the best result directly. Meanwhile the nonlinear dynamic features of the AE signals from rotor rub-impact are analyzed for AE source localization. New nonlinear dynamic features like correlation dimension, maximum Lyapunov exponent and Kolmogorov entropy are proposed to use as the localization features in the inputs of neural network. The experiment results show that rub-impact AE source localization problem is well solved by combining these nonlinear dynamic features and neural network, thus to provide an approach to rotor rub-impact fault diagnosis, the application prospects and further research.