Intelligent Modeling of Abnormal Vibration for Large-Complex Machine Based on Chaos and Wavelet Neural Networks

Abstract

This paper analyses the chaotic characteristics of a large temper rolling millpsilas abnormal vibration signals, and studies phase space reconstruction techniques of the signals. Then, combining the theory of chaotic dynamics and wavelet neural networks, a new vibration model is set up, through inversion method. The property of the model is tested and compared with the model of backpropagation(BP) neural networks, respectively. The result shows that the wavelet neural networks have an advantage over the backpropagation neural networks in rapid convergence and high accuracy.