Multiple-particle-swarm-optimization-based crowding distance.
Published in Chapter:
Application of Statistical Modelling and Evolutionary Optimization Tools in Resin-Bonded Molding Sand System
Ganesh R. Chate (K. L. S. Gogte Institute of Technology, India), Manjunath Patel G. C. (Sahyadri College of Engineering and Management, India), Mahesh B. Parappagoudar (Padre Conceicao College of Engineering, India), and Anand S. Deshpande (K. L. S. Gogte Institute of Technology, India)
Copyright: © 2018
|Pages: 30
DOI: 10.4018/978-1-5225-5396-0.ch007
Abstract
Trial-and-error methods in foundries to determine optimum molding sand properties consume more time and result in reduced productivity, high rejection, and cost. Hence, current research is focused towards development and application of modelling and optimization tools. In foundry, there is requirement of mound properties with conflicting nature (that is, minimize: gas evolution and collapsibility; maximize: compression strength, mound hardness, and permeability) and determining best combination among them is often a difficult task. Optimization of resin-bonded molding sand system is discussed in this book chapter. Six different case studies are considered by assigning different combination of weight fractions for multiple objective functions and corresponding desirability (Do) values are determined for DFA, GA, PSO, and MOPSO-CD. The obtained highest desirability value is considered as the optimum solution. Better performance of non-traditional tools might be due to parallel computing approach. GA and PSO have yielded almost similar results, whereas MOPSO-CD produced better results.