Quality Parameters Coupling Mechanism And Judgement Mechanism Of Coupling Degree In Multi-stage Machining Process

In the gray process of machining, the coupling effect of quality parameters and stream of variation significantly increase the difficulty in quality prediction and control, which subsequently affects final product quality. In this paper, starting from the data-driven quality prediction, the coupling mechanism of quality parameters and discriminant mechanism of coupling degree are systematically studied based on multi-stage machining process from the surface to the essence. Firstly, regarding of the gray process of machining, Fourier transform was introduced to modify the prediction error of metabolic GM (1,1) model when collecting data from production process. A novel cycling Fourier gray model (CFGM (1,1)) was hereby put forward to improve the prediction accuracy, which is able to compensate systematic random error as well as to filtrate the noise of data. Secondly, coordinate system group of machine-clamp-workpiece-tool was set up to quantitatively describe the quality parameters and workpiece figures and thus to identify and calculate the error arising from quality parameters. Then, based on the theory of stream of variation, we developed and studied the dynamic coupling mechanism of quality parameters and discriminant mechanism of coupling degree when produced in the process of machining. Lastly, according to Mahalanobis distance, the concepts of coupling degree and threshold of coupling degree were proposed and the discriminant mechanism of coupling degree was demonstrated as calculation formulas to characterize the coupling possibility and degree. Thus, it is helpful in complex quality control of multi-stage machining process, which will shift quality control from a single process to multi-stage machining process, from single parameter to multi-parameters coupling, from process to advance, from static to dynamic, from local to system and from surface to nature.