Abstract: In order to solve the problems of shrinkage and cracking of cement-based materials and recycling of polyethylene terephthalate (PET) waste, and to prepare PET fiber reinforced cement composite cementitious materials with good toughness and excellent work performance and mechanical properties, the central composite design (CCD) in response surface methodology (RSM) was used to establish a regression model. The effects of fly ash (FA), magnesium oxide expansive agent (MEA) and PET fiber composite additive on the work performance (fluidity) and mechanical properties (flexural strength, compressive strength and flexural-to-compression strength ratio) of cement mortar and their synergistic effects were systematically studied. The reliability of the regression model is verified by variance analysis, and multi-objective optimization is carried out based on the desirability function. The results show that the fluidity is mainly affected by FA content (A) and PET fiber content (C), and the antagonistic effect of AC interaction is significant. The flexural strength is mainly affected by A and C, and the synergistic effect of AC is significant. The compressive strength is affected by the combined regulation of A, MEA content (B) and C, and the interaction between AC and AB is significant. The flexural-to-compression strength ratio is significantly affected by the synergistic effect of A and B. Three groups of optimal mix ratios are obtained by multi-objective optimization and the verification test is carried out. The relative errors between the predicted values and the measured values of each index are less than 5%, indicating that the regression model can be effectively applied to the optimal design of the mix ratio of PET fiber reinforced cement composite cementitious materials, which provides an empirical reference for the multi-objective optimization of cement-based composites.

Key words: response surface methodology, cement mortar, composite additive, work performance, mechanical property, PET fiber