Developmment of an intelligent design methodology for composite materials for various applications

Fiber reinforced polymers have various applications in the automotive and aerospace industries. This project focuses on development of an intelligent material design methodology using machine learning and evolutionary optimization algorithms that will provide the end user optimized permutations of g

Project Title

Developmment of an intelligent design methodology for composite materials for various applications

Project Area of Specialization

Mechanical Engineering

Project Summary

Fiber reinforced polymers have various applications in the automotive and aerospace industries. This project focuses on development of an intelligent material design methodology using machine learning and evolutionary optimization algorithms that will provide the end user optimized permutations of glass and carbon fiber reinforced composite laminates under the given constraints of cost and mechanical properties. The advantages of the proposed work include time and cost savings as it would limit the amount of experimentations required by the end-user during the composite design process as this would help narrow down the choices available to the end-user to the best ones. The steps involved in the process include a market survey to collect data of available composite fabrics, the generation of fabric properties data using a commercial code and the development of a machine learning based optimization methodology for optimizing the composite material design. Afterwards analysis for a medium UAV wing is to be carried out where an optimum laminate configuration will be provided using the algorithm. Furthermore, physically testing will be carried out of the configuration under ASTM standards to give real life results of the configuration.

Project Objectives

The objective of the project is the development of a machine learning based design strategy for layered woven fabric composites. The developed algorithm will be able to optimize the material and layup of the composite given the maximum allowed cost of the composite material. The tentative outcomes of our project include:

• Generation of Properties Data for various Configurations of Composite Materials.
• Selection, Training and Testing of Machine Learning (ML) Algorithms for Composites Design.
• Application of trained ML algorithm for Composite Design for a Selected Application.
• Structural Analysis of the Application Using the Designed Composite Material.

Project Implementation Method

This project application is the use of composite laminate as wing of a medium UAV with a MTOW of 15 kg, an optimized laminate configuration will be predicted using the algorithm for the wing of the UAV and will be analyzed using FEA and physical experimentation.

Benefits of the Project

• Limiting the experimentation involved in the material selection process can reduce development cost and save time and precious resources.
• The end user can achieve best optimized results for simple laminate configurations through the methodology

Technical Details of Final Deliverable

The final deliverable obtained is an algorithm derived from the methodology which is able to provide the optimum bending stiffness for a hybrid composite laminate under a specified cost constraint or provide minimum cost for a specified bending stiffness constraint whichever condition is provided to the algorithm with the laminate comprised of carbon and glass fiber laminae and embedded core material for aerospace, automotive or structural applications.

Final Deliverable of the Project

Software System

Core Industry

Manufacturing

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com