Neural networks modeling of honing surface roughness parameters defined by ISO 13565

Abstract For decades, the arithmetic average (AA or R a ) and root sum of squares (RSS or R q ) have been the two major surface roughness measures to define a broad range of surfaces for a mechanical product. A number of drawbacks have been identified in recent years with the above measures. To map more scientifically and closely the surface roughness to the product functions and performances, ISO 13565 has defined a different set of measures, including R k , R pk , R vk , M r1 , and M r2 . This has not only made process planning different and much more difficult, but also made modeling of the relationship between these roughness measures and the machining parameters a multiple-input and multiple-output problem. While some companies are trying the traditional trial-and-error method to implement the ISO 13565 standard, this study applies artificial neural networks to develop an empirical model for the honing process of engine cylinder liners in order to help reduce emissions, improve oil efficiency, and prolong engine life. Threefold cross-validation is applied to develop the models. Hypothesis testing and the prediction error statistics are employed to select the best model. Data from industrial experiments based on fractional factorial design will illustrate th goodness of the modeling approach and the models.

[1]  R. Palmer,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[2]  Alice E. Smith,et al.  Bias and variance of validation methods for function approximation neural networks under conditions of sparse data , 1998, IEEE Trans. Syst. Man Cybern. Part C.

[3]  Neil Gershenfeld,et al.  The nature of mathematical modeling , 1998 .

[4]  George E. P. Box,et al.  Empirical Model‐Building and Response Surfaces , 1988 .

[5]  Dennis W. King.  Statistical Quality Design and Control , 1993 .

[6]  Robert Groth,et al.  Data mining - a hands-on approach for business professionals , 1997 .

[7]  Douglas C. Montgomery,et al.  Applied Statistics and Probability for Engineers, Third edition , 1994 .

[8]  Chang-Xue Feng,et al.  Surface roughness predictive modeling: neural networks versus regression , 2003 .

[9]  Alan J. Miller.  Subset Selection in Regression , 1992 .

[10]  G. Mirchandani,et al.  On hidden nodes for neural nets , 1989 .

[11]  Xiaowen Wang,et al.  Development of Empirical Models for Surface Roughness Prediction in Finish Turning , 2002 .

[12]  Daniel Kersten,et al.  Introduction to neural networks , 1993 .

[13]  Douglas C. Montgomery,et al.  Introduction to Statistical Quality Control , 1986 .

[14]  Sarah S. Y. Lam.  PROCESS MONITORING OF ABRASIVE FLOW MACHINING USING A NEURAL NETWORK PREDICTIVE MODEL , 1998 .

[15]  Mike Stewart,et al.  Some Effects of Liner Finish on Diesel Engine Operating Characteristics , 1993 .

[16]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.

[17]  S. R. S. Kalpakjian.  Manufacturing Processes for Engineering Materials , 1984 .

[18]  John A. Schey,et al.  Introduction to manufacturing processes , 1977 .

[19]  Kevin Swingler,et al.  Applying neural networks - a practical guide , 1996 .

[20]  Andrew Kusiak,et al.  Decomposition in data mining: an industrial case study , 2000 .

[21]  R. H. Myers,et al.  Probability and Statistics for Engineers and Scientists , 1978 .

[22]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[23]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[24]  M. Stone.  An Asymptotic Equivalence of Choice of Model by Cross‐Validation and Akaike's Criterion , 1977 .

[25]  Dong Young Jang,et al.  Study of the correlation between surface roughness and cutting vibrations to develop an on-line roughness measuring technique in hard turning , 1996 .

[26]  Chang-Xue Feng,et al.  Digitizing uncertainty modeling for reverse engineering applications: regression versus neural networks , 2002, J. Intell. Manuf..

[27]  David W. Coit,et al.  STATIC NEURAL NETWORK PROCESS MODELS : CONSIDERATIONS AND CASE STUDIES , 1998 .

[28]  M. C. Shaw.  Metal Cutting Principles , 1960 .

[29]  Jun Wang.  Computational Intelligence In Manufacturing Handbook , 2000 .

[30]  N. Draper,et al.  Applied Regression Analysis , 1967 .

[31]  Andrew Kusiak,et al.  Computational Intelligence in Design and Manufacturing , 2000 .

[32]  Alice E. Smith,et al.  COST ESTIMATION PREDICTIVE MODELING: REGRESSION VERSUS NEURAL NETWORK , 1997 .

[33]  J. S. Hunter,et al.  Statistics for experimenters : an introduction to design, data analysis, and model building , 1979 .

[34]  Andrew Kusiak,et al.  Data mining of printed-circuit board defects , 2001, IEEE Trans. Robotics Autom..

[35]  A. R. Crathorne,et al.  Economic Control of Quality of Manufactured Product. , 1933 .

[36]  A. McQuarrie,et al.  Regression and Time Series Model Selection , 1998 .