A Random Forest model accurately predicts the mechanical and thermal properties of welded joints, optimizing their assessment and reducing the need for physical experimentation.
Authors
B. Senthilkumaran, Department of Computer Science and Engineering, School of Computing, Veltech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology (deemed to be university), Tamilnadu, Chennai, India
C. Sushama, Department of CSE, School of Computing, Mohan Babu University (erstwhile Sree Vidyanikethan Engineering College), AP, Tirupathi, India
S. Bhaggiaraj, Department of Information Technology, Sri Ramakrishna Engineering College, Tamilnadu, Coimbatore, India
Ashish Kaushal, Assistant Professor, Jindal Global Business School, O. P. Jindal Global University, Haryana, Sonipat, India
P Kamalarajan, Department of Chemistry, R.M.D. Engineering College, Tamilnadu, Kavaraipettai, India
M. Dinesh Babu, Department of Mechanical Engineering, Rajalakshmi Institute of Technology, Tamilnadu, Chennai, India
Summary
The research conducted in this study utilizes a comprehensive dataset that incorporates various materials, welding procedures, and climatic variables in order to train the Random Forest model. Our objective is to conduct a full evaluation of the performance of welded joints, specifically for their applicability in certain applications, by integrating predictions of both mechanical and thermal properties.
The Random Forest model exhibits amazing predictive skills, allowing for accurate estimation of crucial parameters such as tensile strength, impact toughness, and thermal conductivity. This methodology not only optimizes the assessment of welded joint efficacy but also mitigates the necessity for significant and laborious physical experimentation. Our research findings emphasize the significance of a number of factors, such as welding parameters, base material selection, and joint shape, in determining the performance of joints.
Published in: 2024 15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024
To read the full paper, please click here.
