Predicting surface roughness of hardened AISI 1040 based on cutting parameters using neural networks and multiple regression

详细信息    查看全文

• 作者：?lhan Asiltürk (1)
  
• 关键词：Intelligent control ; Neural network ; CNC turning ; Surface roughness ; Regression model
• 刊名：The International Journal of Advanced Manufacturing Technology
• 出版年：2012
• 出版时间：4 - November 2012
• 年：2012
• 卷：63
• 期：1
• 页码：249-257
• 全文大小：483KB
• 参考文献：1. Dejparvar Derakhshani E, Akbari AA (2008) Experimental investigation of hardness and spindle speed effect on surface roughness in hard turning operation using CBN cutting tool. Iran J Mech Eng 60:40-9
  2. Lela B, Baji? D, Jozi? S (2008) Regression analysis, support vector machines, and Bayesian neural network approaches to modeling surface roughness in face milling. Int J Adv Manuf Technol. doi:10.1007/s00170-008-1678-z
  3. Rahman M, Zhou Q, Hong GS (1995) On-line cutting state recognition in turning using a neural network. Int J Adv Manuf Technol 10(2):87-2 CrossRef
  4. Thiele JD, Melkote SN (1999) Effect of cutting edge geometry and workpiece hardness on surface generation in the finish hard turning of AISI 52100 steel. J Mater Process Technol 94:216-26 CrossRef
  5. Kohli A, Dixit US (2004) A neural-network-based methodology for the prediction of surface roughness in a turning process. Int J Adv Manuf Technol 25:118-29 CrossRef
  6. Feng CX, Wang X (2002) Development of empirical models for surface roughness prediction in finish turning. Int J Manuf Technol Manag 20:348-56
  7. ?zel T, Karpat Y (2005) Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks. Int J Mach Tool Manuf 45:467-79 CrossRef
  8. Sick B (2002) On-line and indirect tool wear monitoring in turning with artificial neural networks: a review of more than a decade of research. Mech Syst Signal Process 16:487-46 CrossRef
  9. Arbizu PI, Pérez CJL (2003) Surface roughness prediction by factorial design of experiments in turning processes. J Mater Process Technol 143-44:390-96 CrossRef
  10. Sharma VS, Suresh D, Rakesh S, Sharma SK (2008) Estimation of cutting forces and surface roughness for hard turning using neural Networks. J Intell Manuf 19:473-83 CrossRef
  11. Asiltürk ?, ünüvar A (2009) Intelligent adaptive control and monitoring of band sawing using a neural - fuzzy system. J Mater Process Technol, 209(5):2302-2313
  12. L?u Y, Wang C (1999) Neural network based adaptive control and optimization in the milling process. Int J Adv Manuf Technol 15:791-95 CrossRef
  13. Rumelhart D, Hinton G, Williams R (1986) Learning representations by backpropagating errors. Nature 323:533-34 CrossRef
  14. Desai KM, Survase SA, Saudagar PS, Lele SS, Singhal RS (2008) Comparison of artificial neural network (ANN) and response surface methodology (RSM) in fermentation media optimization: case study of fermentative production of scleroglucan. Biochem Eng J 41:266-73 CrossRef
  15. Diamantidis NA, Karlis D et al (2000) Unsupervised stratification of cross-validation for accuracy estimation. Artif Intell 116(1-):1-6 CrossRef
  16. Ozdeniz AH, Yilmaz N (2009) Artificial neural network modeling of the spontaneous combustion occurring in the industrial-scale coal stockpiles with 10-8?mm coal grain sizes. Energy Sources Part A Recover Util Environ Eff 31(16):1425-435 CrossRef
  17. Reddy BS, Padmanabhan G, Reddy KVK (2008) Surface roughness prediction techniques for CNC turning. Asian J Sci Res 1(3):256-64 CrossRef
  
• 作者单位：?lhan Asiltürk (1)
  
  1. Faculty of Technology, University of Selcuk, Konya, 42075, Turkey
  
• ISSN：1433-3015

文摘

In this study, models for predicting the surface roughness of AISI 1040 steel material using artificial neural networks (ANN) and multiple regression (MRM) are developed. The models are optimized using cutting parameters as input and corresponding surface roughness values as output. Cutting parameters considered in this study include cutting speed, feed rate, depth of cut, and nose radius. Surface roughness is characterized by the mean (R a) and total (R t) of the recorded roughness values at different locations on the surface. A total of 81 different experiments were performed, each with a different setting of the cutting parameters, and the corresponding R a and R t values for each case are measured. Input–output pairs obtained through these 81 experiments are used to train an ANN is achieved at the 200,00th epoch. Mean squared error of 0.002917120% achieved using the developed ANN outperforms error rates reported in earlier studies and can also be considered admissible for real-time deployment of the developed ANN algorithm for robust prediction of the surface roughness in industrial settings.