Options
Modelling of 3D topographic parameters of machined surfaces using Artificial Neural Network regression approach
Journal
Materials Today: Proceedings
Date Issued
2022-01-01
Author(s)
Patil, Dushyant
Dhisale, Manthan
Gandhshreewar, Chinmay
Deshpande, Prasad
Verma, Amber
Shah, Bimlesh
Abstract
Surface Topography is very important to understand many manufacturing and industrial processes. Traditionally 2D surface characterization has been carried out in research laboratories and industries to understand the surface topography of material. However later, sophisticated instruments were developed to measure 3D surface topography parameters. Compared to 2D surface topography of material, 3D surface topography holds more detailed information. It provides insight into the 3D nature of material surface rather than its 2D cross section. Measurement of 2D parameters require less effort and investment than 3D surface parameters. A need thus exists to explore ways to identify 3D surface topography parameters from the 2D parameters which can be obtained more easily. The main objective of this paper is to estimate 3D surface topography parameters of a machined material from 2D parameters using Machine Learning techniques. The current paper focuses on the collection of data of 2D and 3D surface topography parameters for various machining processes. For the inspection of surface parameters, sophisticated 3D surface profilometer was used. The Artificial Neural Network Regression based Machine Learning model is thus proposed, implemented and rigorously tested for various machined surfaces for different operating parameters.
Volume
62
Subjects