An evolutionary approach for modeling and optimization of gelcasting of ceramics

An integrated evolutionary based approach is presented for the modeling and optimization of gelcasting of ceramics. Gelcasting is a well-established colloidal processing method with a short forming time, high yields, high green capacity and low-cost machining, and has been used to prepare high-quality and complex-shaped dense/porous ceramic parts. The gelcasting constituents are reactive chemicals, which directly influences the characteristic properties of the product. Fused Silica (SiO2) ceramics has been prepared at different mix-proportions of solid loading, monomer content and monomer to cross linker ratio. Accurate prediction models to estimate flexural strength, and porosity were evolved from the experimental data using a new potential evolutionary algorithm called multigene genetic programming (MGGP). Subsequently, the developed model has been used for optimization of the mix-proportion of gelcasting constituents. The problem was formulated as a multiobjective optimization problem and a popular evolutionary algorithm, non-dominated sorting genetic algorithm-II (NSGA-II), was used and thereby retrieves the Pareto-optimal solutions set. (C) 2017 Elsevier Ltd. All rights reserved.