Application of nonlinear PCR for optimization of hybrid binder used in construction materials

The research is aimed at optimization of a hybrid binder formulation that includes water solution of sodium silicate (water glass) and polyisocyanate. Optimization is performed with respect to ten output quality characteristics. Calibration modeling is done as a two-step procedure. At first, PCA is applied to the X block for variable reduction. Then nonlinear regression is used to predict a particular quality characteristic as a function of score vectors. The input variables reduction enables to choose an optimal binder formulation that meets the predefined quality requirements. The prediction has been confirmed by the verification experiments.