ANFIS for Predicting Surface Roughness in Turning Operation Performed on CNC Lathe

Abstract:

For mass production, mainly automation is used, in which cutting parameters are set to obtain required surface roughness. The parts like IC Engine piston, cylinders require very smooth surface finish. The same is the case of sleeves, collets etc., of machine parts. These are made by automatic machining operations. To get approximate value of required surface roughness, the cutting parameters that are to be set with help of Adaptive Neuro Fuzzy Inference System (ANFIS) that is designed by using Fuzzy Logic Toolbox. The Fuzzy Logic Toolbox is a collection of functions built on the MATLAB numeric computing environment. It provides tools to create and edit fuzzy inference systems (FIS) within the framework of MATLAB. ANFIS constructs a relation between given parameters (input data and output data), when it is trained with experimentally predetermined values. It consists of different functions, of which bell and triangular membership functions are used for our purpose. The comparison of accuracy of predicted values for both membership functions are performed using testing data. The training and testing data was obtained performing operation on CNC lathe for 50 work pieces of which 40 were used for training ANFIS and the remaining 10 were used for comparing the accuracy of both Bell and Triangular membership functions. The detailed analysis and procedure is presented.

• Periodical: Applied Mechanics and Materials (Volumes 110-116)
• Pages: 1793-1798
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.110-116.1793
• Online since: October 2011
• Authors: M.A. Vinod Kumar
• Keywords: CNC Lathe, Neuro Fuzzy Inference System, Surface Roughness (SR)

References

[1] Shibendu Shekhar Roy, Design of ANFIS, J SCI IND, 64 (2005).

[2] J.S. T Jang, C. T Sun and E. Mizutani, Neuro Fuzzy and Soft computing, Prenctice Hall International, Inc, (1997).

[3] www. mathworks. com.

[4] Chang, L.C., F.J. Chang, and Y. H Tsai, Fuzzy exemplar-based inference system for flood forecasting, Water Resour. Res., 41, (2005).

DOI: 10.1029/2004wr003037

[5] El Bardie M.A. Surface Roughness Model, J Eng Manuf, 207 (1993).

[6] Nagy Bengiamin, Jun Huang, A systematic approach to linguistic fuzzy modeling based on input-output data, Proc. Of winter simulation conference, (2000).

DOI: 10.1109/wsc.2000.899755

[7] C. Elanchezian, T. Sunder Selwyn, G. Shanmuga Sunder, Computer Aided Manufacturing, Lakshmi Publishers.

[8] Sugeno M & Kang G T, Structure identification of fuzzy model, Fuzzy Sets Syst, 28 (1988).