Manufacturing Technology Today

M. Pradeep Kumar ¹, K. S. Vijay Sekar ¹, A. Rajadurai ²

Affiliations
1 Dept. of Mech. Engg., CEG, IN
2 Dept. of Production Engg, MIT, Anna University, Chennai, IN

Abstract

This paper presents a comparative study of the effect of two different flow stress models on the orthogonal metal cutting process using the FEM. The flow stress models used were the Oxley's model and the modified Johnson and Cook model. The orthogonal cutting experiments were conducted with AISI 1045 steel work material and tungsten carbide cutting tool. The FEA results with modified Johnson's model for the cutting force, feed force, chip thickness, shear angle and shear strain compared well with the experimental values with only a marginal deviation of 10-20% for feed rates of 0.16mm/rev and above while the deviation for Oxley's model was higher across most feed rates. The Oxley model gave better results for the FE predictions of contact temperature. This study suggested that the modified Johnson and Cook model was an efficient alternative tool for FE simulation of the orthogonal metal cutting process.

Full Text

     

S. Ramesh ¹, M. Pradeep Kumar ², L. Karunamoorthy ², K. S. Vijay Sekar ²

Affiliations
1 Dept. of Mechanical & Prod. Engg., Sathyabama University, Chennai, IN
2 Dept. of Mechanical Engg., Anna University, Chennai, IN

Abstract

Titanium and its alloys are attractive materials due to their unique high strength to weight ratio and their exceptional corrosion resistance. This paper combines the predictive machining approach with neural network modeling of cutting parameters. Experimental work has been performed in orthogonal cutting of Ti-6Al-4V using Plain Carbide tool. At the selected cutting conditions the forces have been measured. The experimental data were utilized to train the developed simulation environment based on back propagation neural network modeling. The cutting speed, feed, depth of cut have been considered as the input parameters and cutting force, feed force as output parameters to develop the model. The trained neural network was used in predicting the cutting parameters. Predictive ANN models were found to be capable of better predictions of forces about 93 to 97% accuracy within the range that they had been trained.

Full Text