Multi-component signal decomposition method with noise has become a research hotspot of equipment condition monitoring. Aiming at the iterative divergence problem of traditional wavelet ridge extraction algorithm for multi-component harmonic signals widely existing in mechanical and electrical systems, in order to achieve the goal of high decomposition accuracy and anti-noise performance of multi-component signals, the relationship between the initial scale and the extracted components is analyzed. Compared with the time domain of noisy harmonic signals, an improved wavelet ridge extraction algorithm is proposed (WRSD) After the instantaneous frequency of a component is obtained by this extraction algorithm, the component can be separated from the original signal and its instantaneous amplitude can be obtained by using the synchronous demodulation method. This method has high accuracy and certain anti-noise performance for instantaneous frequency estimation. Through simulation analysis and engineering application, the key zero point in intelligent manufacturing equipment of high-performance composite parts can be realized Fault detection of components.
Read full abstract