Papers by Author: Dan Zhang

Abstract: In order to eliminate the noise in ECG signal and increase the diagnosis efficiency, a method based on morphological filtering and wavelet algorithm is proposed. The morphological filters is used to filter out the baseline interference signal, and the wavelet transform is applied to remove high frequency interference. The experiment proves that the algorithm is effective.

Abstract: A method for canceling noise in mechanical signals was presented, which was based on adaptive filtering and discrete wavelet transform Through multi-scale decomposition of wavelet transform, the isolated noise components was as the input signals of the adaptive filter. Through the simulated signal, it shows that the method can achieve noise reduction of non-stationary signals. The proposed approach for noise reduction has been successfully applied to fault diagnosis of bearing signals.

Abstract: This paper presents a fault diagnosis method on roller bearings based on adaptive neuro-fuzzy inference system (ANFIS) in combination with feature selection. The class separability index was used as a feature selection criterion to select pertinent features from data set. An adaptive neural-fuzzy inference system was trained and used as a diagnostic classifier. For comparison purposes, the back propagation neural networks (BPN) method was also investigated. The results indicate that the ANFIS model has potential for fault diagnosis of roller bearings.