Journal
MATERIALS AND MANUFACTURING PROCESSES
Volume 21, Issue 1, Pages 39-45Publisher
TAYLOR & FRANCIS INC
DOI: 10.1081/AMP-200060608
Keywords
back propagation neural network; finish hard turning; residual stress profile; Taguchi method
Ask authors/readers for more resources
Mechanical components shaped by hard turning processes are commonly used under high stress and repeated loading conditions. The physical strength and fatigue life of these components is known to be significantly affected by the residual stress distributions induced by finish hard turning. A thorough understanding of the residual stress profile including both magnitude and direction along the depth of the hard turned workpiece is therefore very important to maximize component life and improve its performance. Many studies have been conducted to determine the effect of cutting tool geometry, cutting parameters, and workpiece material on residual stress distribution in hard turning. However, due to the complexity of hard turning processes, the effect factors considered in these studies are typically insufficient. Knowledge of the most important effect factors such as cooling type, insert grade and tool geometry, tool wear, cutting conditions, and workpiece material has been limited. Very few analytical models are available and accurate enough to predict residual stress profiles in hard turning. In this paper, a series of experiments were designed based on the factorial robust engineering method to explore the full factors affecting the residual stress profiles, and an intelligent model based on back-propagation neural network (BPNN) was developed to predict circumferential and longitudinal residual stress profiles in hard turning. The prediction results match the experimental results well, and much higher performance relative to conventional linear regression method has been achieved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available