Finite Element Analysis of Induced Residual Stresses during Turning of Carbon Steels and Optimization of the Process Parameters

Abstract

Machining processes are essential to manufacturing industries worldwide, playing a crucial role in shaping and finishing components to precise specifications. Turning is one of the most widely utilized machining processes for carbon steels owing to its versatility and efficiency. However, residual stresses are induced into the workpiece materials during turning, significantly affecting their strength, dimensional stability, durability, and performance. Conventional methods for measuring residual stresses, such as hole drilling and X-ray diffraction, are often costly and time-consuming. They provide limited data to optimize the parameter selection and metal-cutting conditions, which can influence the residual stress distribution and magnitude. Moreover, these methods are limited by the sample size, surface preparation, and penetration depth. Therefore, there is a need for alternative approaches that can provide more accurate and reliable predictions of residual stresses in turning operations at a reduced cost. This research focused on developing a finite element model for predicting induced residual stress and optimizing turning parameters: speed, depth of cut, and feed rates. The study analyzed induced residual stress profiles on the surface and subsurface layers generated during turning operations on carbon steels. The Johnson-Cook constitutive model was applied to model the effects of workpiece material properties, and simulations were conducted using ANSYS (explicit dynamics). Two workpiece materials, AISI 1020 and AISI 1040, were selected for the study. These materials are commonly used in manufacturing cylindrical shafts, pipes, crankshafts, and couplings produced through turning and are known for their ease of machinability. Experimental turning operations were conducted on carbon steel workpieces to validate the results. X-ray diffractometer measured the induced residual stresses on the workpiece materials. In addition, the research investigated the influence of speed, depth of cut, and feed rates on residual stresses and cutting forces and performed optimization of process parameters using basic Taguchi design, Grey Relational Analysis (GRA), Surface Response Methodology (RSM), and Genetic Algorithm (GA). The key findings highlighted the substantial influence of cutting speed on resultant cutting forces for both AISI 1020 and AISI 1040. The models developed were validated by comparing the experimental results, with minor discrepancies of 5.21% for AISI 1020 and 5.47% for AISI 1040, particularly at a depth of 60 µm. ANOVA results for the resultant cutting forces revealed that for AISI 1020, cutting speed (v) had the most significant impact (95.23%), while feed rate (f) and depth of cut (d) demonstrated relatively minor effects. Similarly, for AISI 1040, cutting speed (v) was the dominant factor (94.96%), with feed rate (f) and depth of cut (d) exerting minimal influence. Optimal turning parameters were successfully identified for both steel types, with a cutting speed of 80 m/min, a feed rate of 0.2 mm/rev, and a depth of cut ranging from 0.2 mm to 0.4 mm, yielding the most effective results. AISI 1020 steel displayed surface compressive residual stresses from -308.0 MPa to -267.5 MPa, while AISI 1040 steel showed similar compressive stresses ranging from -650.1 MPa to -140.5 MPa. ANOVA revealed that speed (v) significantly influenced induced residual stresses for both materials, contributing 95.24% and 69.43% of the total sum of squares, respectively. Interactions between speed and feed rate (v*f) were also significant (P-value ≤0.05) for both steels, alongside feed rate (f) for AISI 1040 and depth of cut (d) for AISI 1020. Chip thickness deviations were within 6.01% for AISI 1020 and 2.35% for AISI 1040, affirming the accuracy of the finite element model (FEM). In this study, a robust 3D FEM was developed to accurately predict induced residual stresses, which were rigorously validated through experimental measurements using an X-ray diffractometer. The minimal percentage errors between the simulated and experimental datasets highlighted the precision of the model. Furthermore, applying the Johnson-Cook model to simulate material behavior, friction, and flow stresses enhanced the robustness of the FE model, making it highly applicable for practical industrial use in the machining sector. This research is unique because it precisely determines the optimal cutting parameters (cutting speed, feed rate, and depth of cut) that minimize residual stress, significantly reducing to -136.23 MPa. This result not only underscores the effectiveness of combining FEM and RSM but also provides critical, empirically validated insights for industry professionals seeking to improve part quality and reduce defects in post-machining processes. The model represents a reliable and validated approach to predicting and controlling induced residual stresses in machining operations, contributing to enhanced product performance and longevity.