Sparsity adaptive matching pursuit and spectrum line interpolation method for measuring radial and axial error motions of spindle rotation

This paper proposes a sparsity adaptive matching pursuit and spectrum line interpolation (SAMP-SLI) method for measuring the radial and axial error motions of spindle rotation of machine tool. The method measures the displacements from the outside surface of a custom-fabricated cylindrical precision artifact using capacitive displacement sensors. The measured error motions are reconstructed with sparse spectrum lines in the frequency domain by SAMP algorithm. The frequencies, amplitudes, and initial phases of these sparse spectrum lines are corrected using the spectrum line interpolation algorithm to reduce the effect of the spectral leakage and picket fence. Then, the relevant error motions of spindle rotation are separated from these corrected sparse spectra. A simulation is conducted to compare the calculation performance of the proposed method and the existing methods, the results of which shows that the proposed method has higher accuracy in decomposition of error motions. The method is then applied to the measurements of spindle of a numerical control (NC) vertical machining center. The measurement results of the proposed method are compared with that of the API spindle measurement system, which shows that the proposed method has obviously higher accuracy in spindle speed identification. Finally, the measurement uncertainty of the proposed method is analyzed.