Authors: Petia Koprinkova-Hristova; Nikolay Tontchev; Silviya Popova
Addresses: Institute of Information and Communication Technologies, Bulgarian Academy of Sciences, Acad G. Bonchev str. bl.25A, Sofia 1113, Bulgaria ' Department of Transport Management, University of Transport, 158 Geo Milev str., bl.2, Sofia 1574, Bulgaria ' Institute of System Engineering and Robotics, Bulgarian Academy of Sciences, Acad G. Bonchev str. bl.2, Sofia 1113, Bulgaria
Abstract: The paper presents the application of multi-criteria optimisation to steel alloy composition determination aimed at obtaining improved properties material for crankshafts production. Neural network model of steel mechanical characteristics in dependence on amounts of its alloying elements was used for simulation purpose. Two optimisation approaches were applied to find optimal alloys composition. In the first one, several Pareto optimal solutions were found based on those obtained by simulation numerous compositions in the investigated region of interest. The second one used Taguchi method for robust design. The obtained optimal solutions were compared and decision about further production experiments was taken.
Keywords: neural networks; modelling; multicriteria optimisation; Pareto front; Taguchi methods; steel alloys; vehicle crankshafts; crankshaft production; simulation; robust design.
International Journal of Reasoning-based Intelligent Systems, 2013 Vol.5 No.2, pp.96 - 103
Received: 08 May 2021
Accepted: 12 May 2021
Published online: 20 Oct 2013 *