Design of Alumina Reinforced Aluminium Alloy Composites with Improved Tribo-Mechanical Properties: A Machine Learning Ap
- PDF / 1,029,363 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 18 Downloads / 270 Views
ORIGINAL ARTICLE
Design of Alumina Reinforced Aluminium Alloy Composites with Improved Tribo-Mechanical Properties: A Machine Learning Approach Titov Banerjee1,3 • Swati Dey2 • Aluru Praveen Sekhar3 • Shubhabrata Datta4 Debdulal Das3
•
Received: 16 March 2020 / Accepted: 7 October 2020 Ó The Indian Institute of Metals - IIM 2020
Abstract Artificial intelligence approach for data-driven design is employed to design an alumina reinforced aluminium matrix composite (AMC) with improved tribomechanical properties. Machine learning tool, viz. Artificial neural network (ANN), is used as a tool to create a set of models describing the properties of the AMC. The database required for the ANN modelling was extracted from published literature. The objective functions to search the optimum combinations of composition, size and morphological properties were provided from those ANN models. Since the objectives are conflicting in nature, a multi-objective optimization is introduced using genetic algorithm as a tool and the achieved Pareto solutions are used for designing the composite with tailored properties. Keywords Metal matrix composite Aluminium Alumina Mechanical behavior Wear Artificial neural network Genetic algorithm Multi-objective optimization Pareto front
& Shubhabrata Datta [email protected] 1
Birbhum Institute of Engineering and Technology, Siuri, India
2
Department of Aerospace Engineering and Applied Mechanic, Indian Institute of Engineering Science and Technology, Shibpur, Howrah 711103, India
3
Department of Metallurgy and Materials Engineering, Indian Institute of Engineering Science and Technology, Shibpur, Howrah 711103, India
4
Department of Mechanical Engineering, SRM Institute of Science and Technology, Kattankulathur, Chennai 603203, India
1 Introduction Aluminium alloy has high strength-to-weight ratio, adequate ductility but poor wear performance. To meet the perpetual need of transport industries, i.e. in aerospace, aircraft and automobile industries [1–4], its application is restricted to fabricate such components like engine head, disc brake, engine cylinder, piston etc. where friction and wear are important issues. To overcome those difficulties, from last four decades a constant effort is given to focus interest in design and development of aluminium alloy based metal matrix composites. These materials can provide increased strength and hardness as well as better wear performance. In addition to the metal matrix, hard ceramics (called as reinforcement) are incorporated to obtain composites which exhibit improved tribo-mechanical properties. It is quite understandable that many variables are responsible to guide the wear process in a tribo-system. It may also be admitted that no general statement is possible which can explain the nature of wear rate in relation to matrix behavior [5]. Narayan et al. [6] reported that factors controlling the matrix behavior together with tribological parameters are responsible for wear process. Wear is overly complex in nature. It i
Data Loading...