The synthesis error is mainly related to the machine structure and the cutting process. Because the errors induced by the cutting process are related to the working conditions and the cutting parameters, the error model for the synthesis error should be a dynamic model which is suitable
for various working conditions. To analyze the synthesis error, a series of twenty cutting tests were carried out, and the errors of the machined shafts were simulated by the back-propagation neural network (NN). Because the trained NN stores distributed information of the synthesis errors,
the weights and bias of the trained NN were extracted and were programmed by the visual basic (VB). The macro program is employed to realize the synthesis error compensation under the various working conditions. Using the error compensation method, the maximum synthesis error of the machined
shafts is reduced from 13.6 μm to 2.1 μm. The error is compensated by 92% compared with no compensation.
ADVANCED SCIENCE LETTERS is an international peer-reviewed journal with a very wide-ranging coverage, consolidates research activities in all areas of (1) Physical Sciences, (2) Biological Sciences, (3) Mathematical Sciences, (4) Engineering, (5) Computer and Information Sciences, and (6) Geosciences to publish original short communications, full research papers and timely brief (mini) reviews with authors photo and biography encompassing the basic and applied research and current developments in educational aspects of these scientific areas.