We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
COMPARISON OF REGRESSION AND ARTIFICIAL NEURAL NETWORK MODELS FOR PREDICTION OF SURFACE ROUGHNESS IN CREEP FEED GRINDING.
- Authors
AFSHARI, D.; RAHMANI, V.
- Abstract
The main objective of this paper is to compare Regression Method with Artificial Neural Network Method in prediction of surface roughness in creep feed grinding. Data used in this paper has been extracted from creep feed grinding of a cobalt based superalloy (MAR-M-509) turbine blade. The input parameters for surface roughness prediction are Wheel Speed (m/sec), Feedrate (mm/min) and Depth of Cut (mm). In this work, Surface Roughness average R was considered a to evaluate surface roughness quality. Result shows amongst the input parameters, the interaction between wheel speed and feed rate has the highest influence on the surface roughness (Ra) and depth of cut has the least effect on Ra and so its effect can be neglected in regression method. Comparison of the results of two models with experiments was satisfactory. In addition, the results of Neural Network and Regression Models were compared to each other and Neural Network was found to be more precise than Regression Model in predicting surface roughness.
- Subjects
SURFACE roughness; ARTIFICIAL neural networks; REGRESSION analysis; GRINDING &; polishing; INTERFACIAL roughness
- Publication
i-manager's Journal on Mechanical Engineering, 2015, Vol 5, Issue 3, p1
- ISSN
2230-9055
- Publication type
Article
- DOI
10.26634/jme.5.3.3440