http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Jinwei Fan,Xingfei Ren,Ri Pan,Peitong Wang,Haohao Tao 한국정밀공학회 2022 International Journal of Precision Engineering and Vol.23 No.9
In the magnetic composite fluid (MCF) polishing process, appropriate polishing parameters are the basis of achieving high-quality polishing without damage. Appropriate polishing parameters are mainly based on an accurate polishing model and an excellent polishing parameters optimization algorithm. However, due to the complicated principle of MCF polishing and various influencing elements, traditional modeling methods have the limitations of low accuracy, poor application, and difficulty in correcting. Therefore, it is challenging to obtain the optimal polishing quality by optimizing the polishing parameters based on the traditional model. This study proposed an online modeling approach considering data cleaning based on machine learning modeling, and the particle swarm optimization (PSO) algorithm was used to optimize polishing parameters. Then, copper polishing experiments were carried out to validate the modeling and optimization methods. The results demonstrate that the proposed machine learning online modeling method can establish an accurate MCF polishing model, and the nano-scale fine polishing of copper can be achieved by the optimized polishing parameters of PSO, and the surface roughness of the copper sample was reduced by 85% to 0.031 μm.