Application of Hilbert–Huang Transform to acoustic emission signal for burn feature extraction in surface grinding process

•A sensor system was constructed using various sensors.•HHT was used to digest AE signal for grinding burn feature extraction.•Both the frequency and amplitude of the processed AE signal change in marginal spectra.•RMS and spectral centroid value of the marginal spectra was defined as grinding burn features.•The application of HHT on AE in grinding burn detection is of great potential. This paper presents a sensor system using motor current sensors, voltage sensors, accelerator and acoustic emission sensor for grinding burn feature extraction. The new method, Hilbert–Huang transform (HHT), was applied as a signal processing tool to digest the raw acoustic emission and accelerator signals and to extract grinding burn features. A filtering criterion using average energy percentage of IMF components was proposed in order to simplify the calculation. Five IMF components were selected based on this criterion and their marginal spectra were calculated. The marginal spectral amplitude of the first three IMF components and the spectral centroid of the last two IMF components clearly reflected the occurrence of grinding burn. Results indicate that the application of HHT to acoustic emission signals in grinding burn detection is of great potential. Besides, the wheel rotation speed can be successfully uncovered through the intrinsic mode function (IMF), which verified the physical meaning of the EMD method.