Abstract

Advanced composite materials have become very important in high-performance electronic applications due to their capability of offering variable electric conductivity, high thermal stability, great mechanical strength, lightweight structure, and enhanced electromagnetic interference (EMI) shielding capacity. Possibilities of using polymer-matrix composite, ceramic composites, carbon-based nano-composites, and hybrid-nanofiller composites exist as possible means of circumventing the constraints faced in using conventional electronic packaging and heat dissipation materials. Conventional composite-material design depends greatly on trial and error, empirical formulation, and optimization of individual parameters. Such a process hinders the efficient identification of suitable matrix-reinforcement pairing, appropriate concentration of fillers, interfacial characteristics, and processing variables. Improving the material’s thermal conductance could result in weakening of its electrical insulation, flexibility, and mechanical stability, thereby creating a multi-objective optimization problem. A new methodological framework of Advanced Composite Intelligence and Multi-objective Optimization (ACIMO), which utilizes artificial intelligence (AI), materials property prediction, and multi-objective evolutionary optimization, is proposed. The framework takes into account the material composition, reinforcement geometry, filler concentration, interfacial characteristics, and fabrication conditions to estimate thermal conductivity, electrical resistivity, tensile strength, and EMI shielding efficiency.

Authors

R. Gayathri
Dr. D.Y. Patil Institute of Technology, India

Keywords

Advanced Composite Materials, Artificial Intelligence, Multiobjective Optimization, Thermal Conductivity, Electronic Materials