Surface roughness prediction in end milling by using predicted point oriented local linear estimation method. Issue 9 (June 2016)

• Record Type: Journal Article
• Authors: Wen, Long; Li, Xinyu; Gao, Liang; Yi, Wenchao
• Is Part Of: International journal of advanced manufacturing technology. Volume 84:Issue 9/12(2016)
• Page Start: 2523
• Page End: 2535
• Publication Date: 2016-06
• Subjects: Local linear estimator -- Surface roughness prediction -- Pseudo data generation -- End milling; Manufacturing processes -- Periodicals; Production engineering -- Periodicals; 670.427
• DOI: 10.1007/s00170-015-7884-6
• Languages: English
• ISSNs: 0268-3768

Abstract

Abstract Surface roughness is one of the most important indexes of the product quality, and the accurate prediction on surface roughness can provide a more economical way to determine the machining process using computer numerical control machines. In this paper, the effects between the machining parameters and the surface roughness are investigated, and one method named predicted point oriented local linear estimator (PPOLLE) is proposed for the surface roughness prediction. The PPOLLE contains two prediction stages: the first one is to generate effective pseudo data and the other one focuses on the final prediction based on the large amount of pseudo samples. The performance of PPOLLE is evaluated by using a set of benchmark functions. The results show that PPOLLE is an effective modeling approach within limited time consumption. Furthermore, two sets of experimental data are obtained to validate the PPOLLE on the surface roughness prediction in end milling. Three machining parameters, the spindle speed, the feed rate, and the depth of cut, are selected as the input. Several comparisons among proposed PPOLLE and several traditional methods, adaptive network-based fuzzy inference system, artificial neural network, and gene expression programming, are also conducted. The results indicate that PPOLLE outperforms those methods and is very effective for the surface roughness prediction.