A surface quality prediction model considering the machine-tool-material interactions

Surface topography is an important factor in evaluating surface integrity and surface roughness of the processed material to ensure the processing quality and shorten the processing cycle. In this paper, a prediction model of milling surface quality, considering the influence of the machine tool factors, the milling tool factors, and the workpiece material factors, is proposed. Based on the dynamic path trajectory, the model takes first the imbalance of the machine tool spindle as a critical machine factor affecting the tool-workpiece displacement and the surface topography. Then, the tool parameters are predicted using a backpropagation neural network (BPNN); the tool wear behaviors are taken into account. To expand the different workpiece application, material properties, such as the elastoplasticity, are introduced to improve the prediction accuracy. Thus, a series of experiments with various machine tools, milling tools, materials, and processing parameters validate the surface quality prediction model based on machine-tool-material interactions. Finally, the results demonstrate that the simulated outcomes accord well with the experiment; the error rate of the best result is only 3.60%, and the average error rate is 9.6%.