Design and optimization of rare-earth free hard magnetic alloys and Nickel-based superalloys for high temperature applications

Rajesh Jha¹; George Stavo Dulikravich¹; Marcelo Jose Colaco²

¹ Florida International University; ² Federal University of Rio de Janeiro

doi:10.20906/CPS/COB-2015-1284

Resumo

Developing a new material or even improving properties of an existing material is a complex and time-consuming task. In recent years, materials scientists around the globe proposed a number of ways to speed up the alloy development process by using various computational tools[1,2]. In this work, we made an attempt to demonstrate the efficacy of using computational tools in design optimization of materials, especially for high-temperature applications. We addressed two different material systems: Alnico alloys (magnetic) and Nickel-based superalloys. Alnico type alloys are hard magnetic alloys and well known for high-temperature applications. In this work, we defined the variable range of various elements and generated an initial set of alloys by a quasi-random sequence generation algorithm. These alloys were synthesized and tested for determining various material properties. We used a response surface methodology approach to develop surrogate models (meta-models) that approximately linked alloy chemistry with desired properties for these multi-component systems while being computationally affordable. These models were further used for multi-objective optimization of desired (conflicting) properties by using a number of algorithms based on evolutionary approaches, as well as our hybrid optimizer[2,3]. Pareto-optimized predictions were experimentally validated and results over the cycles show significant improvement in properties of these alloys. Nickel-based superalloys are used for high-temperature applications in aerospace, nuclear, and petrochemical industry[1]. In this work, we also developed meta-models for two conflicting objectives, namely Stress to rupture and Time to rupture. Thereafter, we made an attempt to improve these properties by multi-objective optimization. Optimization results show significant improvement in these properties for Nickel-based superalloys. In both tasks, we used two commercial optimization packages, "modeFRONTIER" and "IOSO", as well as our hybrid optimizer[1,2,3]. [1] R

Palavras-chave: Response Surfaces; Multi-objective Optimization; Alnico; Material Properties; Pareto-optimized Predictions