Non-linear modelling using central composite design to predict green sand mould properties

Abstract The quality of cast products in green sand moulds is largely influenced by the mould properties, such as green compression strength, permeability, hardness, and others, which depend on the input (process) parameters. The relationships of these properties with input parameters such as sand grain size and shape, amount of binder and water, number of strokes in ramming, etc., are complex in nature. Design of experiments (DOE) with response surface methodology was used to develop the said input-output relationships. An attempt was also made to relate green compression strength to mould hardness. Moreover, responses such as permeability, hardness, and green compression strength were separately expressed as functions of bulk density. The relationship between green compression strength and permeability in a clay-bonded sand system is generally reverse in nature. To ensure good-quality castings, the green sand mould is required to have both adequate permeability as well as strength. This situation will lead to the formulation of a multiobjective optimization problem. A Pareto optimal front of solutions was developed for strength and permeability, using a multiobjective optimization tool - the so-called non-dominated sorting genetic algorithm (NSGA). The Pareto optimal front contains a number of optimal solutions. One optimal solution differs from another on account of the fact that different sets of weightages are given to the objective functions. Thus, the user has to select one set of optimal solutions out of different available sets.