Operating Reliability Assessment Method Based On MED-FVMD Fuzzy Approximate Entropy And FSVDD

The arrival of Industry 4.0 not only requires more intelligent machinery and equipment,but also has higher requirements for the reliability of mechanical equipment.Based on the information of mechanical equipment operation status and signal processing technology as the main means,this paper starts from the three aspects of signal acquisition,state feature extraction and operational reliability assessment,and proposes a minimum entropy-based deconvolution-fast variational mode decomposition.Operating reliability assessment method based on fuzzy approximate entropy and fuzzy support vector data.Firstly,aiming at the problem that the early damage state information of mechanical equipment is vulnerable to strong background noise and is not easy to be extracted,a state feature extraction method based on minimum entropy deconvolution-fast variational mode decomposition and fuzzy approximation entropy is proposed.The method uses the minimum entropy deconvolution to denoise the signal,and enhances the pulse characteristics of the original signal that are originally submerged by noise.The fast-decomposed mode decomposition is used to decompose the denoised signal,and the fuzzy approximation is used to quantify the fast.The modal components obtained after the variational mode decomposition are decomposed and the feature vectors are constructed.Secondly,for the problem of insufficient sample data in the reliability evaluation of mechanical equipment operation,combined with fuzzy support vector data,it only needs the advantage of normal sample data of mechanical equipment when establishing the reliability evaluation model of mechanical equipment operation,and the method can distinguish each sample.Based on the confidence of the data,an operational reliability evaluation model based on fuzzy support vector data description is proposed,and the proposed evaluation model is applied to the measured signals of rolling bearings to verify its effectiveness.Finally,the experimental data of different gears under different working conditions are collected on the experimental platform.The state feature extraction method based on minimum entropy deconvolution-fast variational mode decomposition and fuzzy approximate entropy is used to extract the state features of the acquired signals and construct the state.The feature vector is then evaluated using the operational reliability assessment model based on the fuzzy support vector data description.The results show that the proposed method can effectively evaluate the operational reliability of gears under different working conditions.