This paper proposes a neural network-based optimization scheme for predicting localized stable cutting states in inward-turning operation. A set of cutting experiments are performed in inward orthogonal turning operation. The cutting forces and critical chatter locations are predicted as a function of operating variables including tool overhang length. A neural network model is employed to develop the generalized relations. The optimum cutting parameters are predicted from the model with the help of binary-coded Genetic Algorithms (GA). The results are illustrated with the data of four different work materials.

The most detrimental phenomenon to productivity is unstable cutting. This reduces the tool life and surface quality of the workpiece. Many theoretical investigations are available in literature for prediction of stable and unstable cutting states in orthogonal cutting. In most of the cases, the stability lobe diagram is generated from an analytical linear model by varying one operating parameter at a time. However, cutting processes possess highly nonlinear relationships among the input and output parameters. In orthogonal turning, it is well known that the cutting forces depend on the operating variables such as feed, depth of cut and speed. These variables are often used to control the forces or machining stability by establishing appropriate regression relations. Rao and Shin (1999) studied and found that tool geometry and flank wear have great influence on cutting dynamics. Chiou and Liang (1998) studied the chatter stability of a slender cutting tool and tool wear effect on cutting dynamics. Chandiramani and Pothala (2006) presented dynamics of regenerative chatter during turning operation after considering the variations in shear angle. Berados et al. (2006), Chen and Tsao (2006), and Martinez et al. (2008) have found that the compliance of the work piece has a great influence on cutting dynamics. Azouzi and Guillot (1997), and Risbood et al. (2003) have carried out an experimental investigation to distinguish the stability states of cutting on the basis of output features, such as surface roughness and vibrations in turning operation. Tangjitsitcharoen and Moriwaki (2007) found that the type of chips plays an important role and it can be employed effectively in addition to the cutting force data and stability states.

In practice, there are several other operating parameters like tool overhanging length and type of material, which could also alter the critical operating conditions in parallel. For example, variation of tool overhang length changes the stiffness of the tool holder which, in turn, affects the tool life under unstable conditions. Likewise, the effects of cutting fluids on the surface roughness and tool wear have been predicted (Dhar and Kamruzzaman, 2006). Gaitonde et al. (2008) studied the overall influence of the amount of lubrication along with cutting speed and feed rates on the surface roughness and specific cutting forces which directly effects the stability of the process. Shetty et al. (2008) found that variables like steam pressure influence the surface roughness of the workpiece. Based on this finding they developed a model of cutting tool dynamics.

Mechanical Engineering Journal, Inward-Turning Operation, Genetic Algorithms, Neural Network Model, Mean Square Error, Evolution Strategies, Natural Genetic System, Genetic Operations, Orthogonal Turning Operation, Mechanical Systems, Biological Evolution.