Genetic Algorithm for Material Optimization

Simulated the creation of composite materials with tailored effective properties through a genetic algorithm framework. The model integrates mechanical, electrical, and thermal governing equations to compute composite behavior at varying doping concentrations. Using cost functions and weighted parameters, the algorithm iteratively optimized material compositions within Hashin–Shtrikman bounds to predict realistic upper and lower performance limits. A 3D finite-element thermal model (21×21×21 mesh, 9,261 nodes) was implemented to validate the simulation’s physical accuracy, visualizing laser-induced temperature distributions and confirming model reliability. Assumptions regarding homogeneous mixing and linear property averaging were analyzed to understand limitations and trade-offs. The project demonstrated how multi-objective optimization can guide the design of next-generation composite materials with custom thermal, mechanical, and electrical responses.