The Research Of Surface Roughness Prediction In Precision Turning Based On Copula Theory

As the reindustrialization strategy of America and industry 4.0 of Germany were proposed, China also put forward the manufacturing industry platform for action “made in china 2025”, and established that intelligent manufacturing was the main direction of future manufacturing industry. In order to build the intelligent factory and realize the intelligent production, effective monitoring method and surface quality prediction was essential, so the relevance between surface quality and monitoring parameters was the key problem needed to solve. Under the background of 7075 aluminum alloy precision turning, the new surface roughness prediction method based on Copula function was put forward in this research, the main research work was as follows:1. The correlation between cutting parameters（cutting speed, feed rate, depth of cut） and cutting force as well as surface roughness were studied, which indicated that the cutting parameters against the cutting force showed the same trend as the relations between cutting parameters and surface roughness, and significant correlation was existed between cutting force and surface roughness.2. Combining with the limitations of traditional relevance analysis method and features of mechanical processing, a new relevance analysis method based on Copula function was proposed. Meanwhile, sample selection was done to construct the marginal distribution function, best Copula function as well as joint distribution function.3. The relationship between cutting force and surface roughness was deeply analyzed with Copula function, which revealed that the positive and nonlinear correlation was found between them, and their dependency structure was centro symmetric. Besides, the influence degree of cutting forces in the different direction on the surface roughness was not same, the results revealed that the radial cutting force performed the best and axial cutting force showed the worst, and the relevance degree between cutting force and surface roughness varied with their magnitude.4. A new surface roughness prediction model was constructed based on the Copula function and probability theory. Then, this model was verified with the experiment results, and it was found that this model provided good prediction accuracy（94.3464% average accuracy） and has a great application value in practice.This study was supported by the national major projects of science and technology “high speed CNC cutting process monitoring and assessment method”（№2013ZX04009031-4） and the funding of Jiangsu innovation program for graduate education（CXZZ12₀658）. The surface roughness prediction method based on Copula function proposed in this study provided a new vision for the on-line monitoring technology in the process of automatic and mechanical manufacture.