Energy Efficiency State Identification Of Milling Process Based On EEMD-PCA-ICA

On-line identification of energy efficiency in cutting process is one of the important topics in energy efficient cutting.The research target is to judge whether the cutting state is in a high energy efficiency state online.Aiming at the problem of energy efficiency state identification in cutting process,the characteristics and extraction algorithm of energy efficiency states were studied in monitoring signal of machining system.From the angle of milling force,this article extracts the information related to energy efficiency status in milling force signal,explores the attributes and changing rules of energy efficiency status in milling process,and provides theoretical basis and technical solutions for on-line monitoring of energy efficiency status in milling process system.Energy efficiency state identification system for machine tools is built under no-load and load conditions.Energy efficiency state identification test is designed under no-load and load conditions.The energy efficiency state of noload and milling process is classified by the empirical model.Because the milling force signal is a kind of mixed multi-source state signal,an algorithm for separating and extracting characteristic components of energy efficiency state in milling force signals is proposed based on ensemble empirical mode decomposition(EEMD),principal component analysis(PCA)and independent component analysis(ICA)in this article.Firstly,the pre-processed singlechannel milling force signals are separated into multi-component components by ensemble empirical mode decomposition,in order to decouple the mixed signals.Then principal component analysis is used to remove the correlation between the components,extract the main components related to the source signal,and reduce the dimension of the data,for speeding up the iteration rate of the algorithm and the efficiency of subsequent signal separation.After dimension reduction,the components are processed by fast independent component analysis,and the corresponding independent source signals composed of multi-component components are separated.The characteristic components representing the energy efficiency state are extracted by analyzing the time-frequency characteristics of the source signals of milling force separation.It is found that in the high energy efficiency state,the proportion of spindle frequency doubling component is higher than that in the low energy efficiency state,and the frequency of the third and fourth order components of the source signals of milling force separation is higher than that of the spindle frequency doubling component.The experimental results show that the EEMD-PCA-ICA identification algorithm can separate the components representing the energy efficiency state in the milling force signal,and identify the energy efficiency state in the milling process by analyzing the position of the energy efficiency state component and the proportion of the principal components,so as to provide an evaluation basis for the energy efficiency state recognition in the milling process.