Title: Optimisation of the multi-pass grinding operation using Evolution Strategy with variable length representation

Authors: Mayank Chadha, Cheol W. Lee

Addresses: Industrial and Manufacturing Systems Engineering, University of Michigan – Dearborn, 4901 Evergreen Road, Dearborn, MI 48128, USA. ' Industrial and Manufacturing Systems Engineering, University of Michigan – Dearborn, 4901 Evergreen Road, Dearborn, MI 48128, USA

Abstract: Conventional optimisation schemes for the multi-pass grinding operation assume an identical condition for each roughing pass and another condition for finishing. The objective of this study is to obtain the globally optimal solution by individually determining the condition of each pass while choosing the total number of passes at the same time. A novel optimisation scheme is developed based on the Variable Length Evolution Strategy (VLES) which works with individuals of varying length. The proposed algorithm is implemented for the cylindrical traverse grinding process. Advantages of the VLES are demonstrated by comparing its performance with that of a conventional scheme. [Received 5 May 2009; Revised 17 December 2009; Accepted 4 January 2010]

Keywords: evolutionary algorithms; process design; process optimisation; process control; multi-pass machining; global optimum; cylindrical traverse grinding; variable length representation.

DOI: 10.1504/IJMR.2010.033468

International Journal of Manufacturing Research, 2010 Vol.5 No.3, pp.286 - 304

Published online: 02 Jun 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article