An Optimal-Pruned Extreme Learning Machine based modelling of surface roughness

No Thumbnail Available
Janahiraman T.V.
Ahmad N.
Journal Title
Journal ISSN
Volume Title
Institute of Electrical and Electronics Engineers Inc.
Research Projects
Organizational Units
Journal Issue
A computer based modelling and prediction method is vital in the field of Computer Numerical Control based cutting operation. The final quality of finished surface is mainly influenced by the interaction between the work piece, cutting tool and machining system. Therefore, many researchers attempted to develop an efficient prediction systems for surface roughness before machining. In this paper, Optimal Pruned Extreme Learning Machine (OPELM) is proposed for modelling and predicting surface roughness with respect to its cutting parameters in turning based machining process. The surface roughness models obtained from other methods such as Response Surface Method, Neural Network and Extreme Learning Machine were compared with the experimental results. Our experimental study consist of 15 workpieces that were used for cutting using turning operation. The correlation between the input parameters such as feed rate, cutting speed and depth of cut with surface roughness was modelled using OPELM. Based on our study, OPELM performed the best in modelling and predicting based on unknown set of input. � 2014 IEEE.
Backpropagation; Computer control systems; Cutting tools; Forecasting; Knowledge acquisition; Learning systems; Neural networks; Numerical methods; Surface properties; Turning; Back propagation neural networks; Computer numerical control; Cutting operations; Efficient predictions; Extreme learning machine; Response surface method; Response surface methodology; Surface roughness model; Surface roughness