LCD denoise and the vector mutual information method in the application of the gear fault diagnosis under different working conditions

2017 5th International Conference on Mechanical Engineering, Materials Science and Civil Engineering

机械制造;材料科学;土木建筑工程

Xiangfeng, Zhang^1 ; Hong, Jiang^1

School of Mechanical Engineering, Xinjiang University, Urumqi

830047, China^1

关键词: Cross-correlation coefficient;    Fault characteristics;    Gear fault diagnosis;    Instantaneous frequency;    Mutual information method;    Mutual informations;    Physical significance;    Subjective experiences;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/324/1/012072/pdf
DOI  :  10.1088/1757-899X/324/1/012072

来源: IOP

【 摘 要 】

In this paper, the full vector LCD method is proposed to solve the misjudgment problem caused by the change of the working condition. First, the signal from different working condition is decomposed by LCD, to obtain the Intrinsic Scale Component (ISC)whose instantaneous frequency with physical significance. Then, calculate of the cross correlation coefficient between ISC and the original signal, signal denoising based on the principle of mutual information minimum. At last, calculate the sum of absolute Vector mutual information of the sample under different working condition and the denoised ISC as the characteristics to classify by use of Support vector machine (SVM). The wind turbines vibration platform gear box experiment proves that this method can identify fault characteristics under different working conditions. The advantage of this method is that it reduce dependence of man's subjective experience, identify fault directly from the original data of vibration signal. It will has high engineering value.