Study on prediction model of surface roughness of SiCp/Al composites based on Neural Network. Issue 1 (1st January 2022)

• Record Type: Journal Article
• Authors: Sun, Hao; Zhang, Chaochao; Li, Yikai; Yin, Tingting; Zhang, Hanming; Pu, Jin
• Is Part Of: Journal of physics. Volume 2174:Issue 1(2022)
• Publication Date: 2022-01-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2174/1/012091
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: In order to effectively meet the actual industrial production standards and improve the prediction accuracy of composite surface roughness, a prediction model of SiCp/Al composite surface roughness based on neural network is proposed. The influence parameters of surface roughness of SiCp/Al composites are analyzed from the cutting tool parameters, and the mathematical calculation of surface roughness of SiCp/Al composites is carried out. Using neural network technology, by determining various parameters of neural network, collecting and processing various data of material surface, the surface roughness prediction model of SiCp/Al composite is constructed to realize the surface roughness prediction of SiCp/Al composite. The experimental results show that the maximum error between the actual value and the predicted value of the surface roughness of composite materials from the prediction model established in this paper is only 0.013, and the average error percentage between the actual value and the predicted value is 0.705%, which can effectively improve the prediction accuracy of the surface roughness of composite materials and meet the standards of actual industrial production.