WORK ROLL GRINDING –MODELING AND OPTIMIZATION USING ARTIFICIAL NEURAL NETWORK AND GENETIC ALGORITHM
Abstract
Selection of optimum parameters in work roll grinding mainly relies on the experience and expertise of individuals working in grinding industries. Systematic knowledge accumulation regarding the manufacturing process is essential in order to obtain optimal process conditions. It is not safe a priori to presume that rules of thumb, which are widely used on the shop floor, always lead to fast production and to increased productivity. Thus, neural network Meta models are suggested in this work in order to generalize from examples connecting input process parameters, such as wheel speed work speed, in-feed, Traverse speed, dress depth and dress lead. These examples or knowledge are gathered from experiments from industrial practice, which are designed systematically using orthogonal arrays (DOE). Neural network model thus developed yields a more accurate process than the regression method. Furthermore, they can be employed in the fitness function of a genetic algorithm that can optimize the grinding conditions.
Downloads
References
K.Matsushima and T. Sata, ‘Development of intelligent machine tool’, Jnl. of Faculty Eng., University, Tokyo, No.35, pp: 299-314 (1980).
L.Monostori and D. Barschddorff, ‘Artificial neural networks in Intelligent Manufacturing ‘, Robotics computer Integrated Manufacturing No.9, pp: 412-436 (1992).
R.P. Lippmann, ‘An introduction in computing with neural network’, IEEE ASSP Magazine, pp: 4-22, (1982).
G. Chryssolouris and M.Guillot, ‘A comparison of statistical and AI approaches to the selection of process parameters in intelligent machining, Jnl. Engg. Industry. No.112, pp: 122-131 (1990).
S.Ranagwala and David A. Dornfield, ‘Learning and optimization of machining operation using computing abilities of neural network’, IEEE Trans. Sys. MangementNo.19, pp: 299-314,(1989).
Pignatiello J.J., ‘An overview of the strategy and tactics of Taguchi’, Inst. Ind. Engg. Transactions Vol.20; pp.247-254 (1988).
Ross J., ‘Taguchi technique for quality Engineering’, McGraw-Hill, Singapore, (1999).
McGraw-Hill, Singapore, (1999).
Godlberg.D.E. ‘Genetic algorithms in search, optimization and machine learning’, Addison Wesley, Reading, MA. (1989).
Benardos P.G. and Vosniakas G.C. (2003), ‘Predicting surface roughness in machining: a review’, Int. J. Machine Tools Manufacture, Vol.43.No.8.pp.833-844. (2003).
Deb K. (1995), ‘Optimization in Engineering Design: Algorithms and examples’, Prentice-Hall, New Delhi.
