Genetic-Algorithm-Based Optimization of an Industrial Age-Hardening Operation for Packed Bundles of Aluminum Rods

Abstract

Industrial thermal processing operations are highly energy intensive, and must be efficiently operated, especially under the current global energy crisis. However, optimization of such complex operations is a nontrivial task and must be formally approached. In the present work, an integrated model has been used in conjunction with a genetic algorithm to optimize industrial age-hardening of packed bundles of aluminum alloy rods. The presence of peak yield strength and the strong dependence of peak time on age-hardening temperature make this operation interesting for optimization studies. Multiobjective optimization problems have been formulated and solved for energy minimization, productivity maximization, and variability reduction during age-hardening by an elitist nondominated sorting genetic algorithm. The Pareto-front for optimum furnace temperature, time, and quality variation has been determined for isothermal as well as multisegment nonisothermal furnace temperature profiles. Subsequently, optimum multisegment furnace temperature-time set-points were obtained and compared with equivalent isothermal profile. The advantage of multisegment nonisothermal profile over isothermal profile in reducing the energy consumption has been demonstrated.

Keywords:

• Age-hardening
• Aluminum alloy
• Energy consumption
• Genetic algorithm
• Integrated model
• Modeling
• Multiobjective
• Nondominated sorting genetic algorithm
• Optimization
• Pareto-optimality
• Precipitation model
• Productivity
• Product quality
• Thermal model

ACKNOWLEDGMENTS

The authors are grateful to Professor Mathai Joseph, Executive Director, Tata Research Development and Design Center, Pune, for approving and supporting this research project. The second author (RM) is grateful to Professor K. Deb, Indian Institute of Technology, Kanpur in training as well as providing him resources on NSGA II technique used in this work.