A study on vibration recognition of nano-imaging system based on wavelet analysis

In order to intelligently diagnose the vibration types corresponding to various errors in nano-imaging process, so that the experimental personnel could take corresponding measures, in this paper, firstly, all types of vibration signals were decomposed and reconstructed in nano-imaging process based on the wavelet transform, thus extracting feature vectors of all types of vibration signals. Secondly, BP neural network model was established, and network training was carried out with the obtained feature vectors as the input information of network and all types of vibration sources as the output information of the network, which was finally passed through the actual inspection. The results showed that, the feature value of all types of vibration signals extracted and obtained by wavelet feature has merged together with BP neural network model, whose network recognition result are basically consistent with actual vibration signals. According to the results, it could effectively recognize the all types of vibration signals during the nano-imaging process and has a higher practical guiding significance. © Liu et al.