Selection of elemental composition of a Nickel-based super alloy through neural networks and multi-criteria decision making

ABSTRACT

The present work aims to develop nickel-based superalloys by studying the impact of their alloying elements in providing the desired physical and mechanical characteristics. At present, materials are being developed using experiment-driven trial and improvement. The long time duration in their development rules out the planning of new materials and thus leading engineers to select the commercially available materials with a compromise towards the properties. The computationally developed materials have the potential to meet the needs of the materials engineers, and now the materials are being brought into the centre of the planning method. The selection of suitable elemental composition seems to be a multi-criterion decision-making problem with conflicting and diverse objectives as different elements execute different properties at varied conditions. Nickel is an adaptable element that can alloy with most metals. The present study deals with the procedure that can be applied to identify suitable elemental composition for a Ni based super alloy, allowing them to adjust for required properties in order to enhance its use in a particular application. The methodology deals with the selection of ‘n’ number of alternatives to give desired properties and applying multi-criteria decision-making algorithms to rank and select one among the alternatives developed. The Spearman rank correlation coefficient in the range of 0.74-1.00 shown a strong correlation between the methods adopted. From the analysis Cr (wt. 12%), Co (wt. 16.5%), Mo (wt. 2.5%), W (wt. 1.5%), Nb (wt. 1.5%), Al (wt. 2%), Ti (wt. 3%), Fe (wt. 1.5%), C (wt. 0.2%), and B (wt.0.03%) can be developed to get an yield strength of 504.94 MPa, Ultimate tensile strength of 678.62 MPa, % elongation 18.53 and Creep resistance at 1000 hours work life of 263 MPa to get better ductile properties. Also Cr (wt. 18%), Co (wt. 17%), Mo (wt. 4%), W (wt. 2%), Al (wt. 2%), Ti (wt. 1.5%), Fe (wt. 2%), C (wt. 0.1%) and B (wt.0.02%) can be developed to get an yield strength of 451.25 MPa, Ultimate tensile strength of 669.57 MPa, % elongation 25.18 and Creep resistance at 1000 hours work life of 309 MPa to get better strength properties.

KEYWORDS:

• Compositional effects
• neural networks
• MCDM techniques
• selection of an alternative

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Dileep Kumar Ganji

Mr. Dileep Kumar Ganji is a Research scholar of VIT, post graduated in Industrial engineering specialization.  Currently working on development of Nickel based super alloys for gas turbine applications. Worked as Assistant professor in Vidyanikethan Engineering College. 

Rajyalakshmi Gajjela

Dr. Rajyalakshmi G Completed Master’s degree in Industrial Engineering from Sri Venkateswara University, Tirupati, India in 2003. Received PhD degree in Production Engineering –from Sri Venkateswara University, Tirupati, India in 2015. Working as Associate Professor in School of Mechanical Engineering, VIT, and Vellore. Having a total of 17 years of teaching experience and 10 years of research experience. Authored and co-authored, more than 50 refereed journal and 30 conference proceedings papers.  Collaborating with many professors from various institutes coordinates for several masters and research projects on Production, Optimization and Materials area. Main Research interests are related with Advanced Machining, Optimization, Development of hybrid algorithms, Industry 4.0, development of materials and materials processing techniques, Tool design and Sustainable manufacturing etc. Having membership with IISE. Completed a funded project from DST-SERB entitled “Improving the performance of WEDM by developing wire electrodes”. Produced three PhD and guiding 5 students. Applied for a patent on “Tool for making Non Circular holes”.