Ultra-high-performance concrete (UHPC) results from the mixture of several constituents, leading to a highly complex material in both, fresh and hardened state. The higher number of constituents, together with a higher number of possible combinations, relative proportioning and characteristics, makes the behavior of this type of concrete more difficult to predict. The objective of the research is to build four analytical models, based on artificial neural networks (ANN), to predict the 1-day, 7-day, and 28-day compressive strengths and slump flow. Recycled glass powder milled to different particle size, fluid catalytic cracking residue (FCC) and different particle size limestone powder was used as partial replacements for Portland cement and silica fume. The ANN models predicted the 1-day, 7-day, and 28-day compressive strengths and slump flow of the test set with prediction error values (RMSE) of 2.400 MPa, 2.638 MPa, 2.064 MPa and 7.245 mm respectively. The results indicated that the developed ANN models are an efficient tool for predicting the slump flow and compressive strengths of UHPC while incorporating silica fume, limestone powder, recycled glass powder and FCC.

Ultra-high-performance concrete (UHPC) is a kind of high-tech cementitious material with superb mechanical and durability properties compared to other types of concrete. However, due to the high content of cement and silica fume used, the cost and environmental impact of UHPC is considerably higher than conventional concrete. For this reason, several efforts around the world have been made to develop UHPC with greener and less expensive local pozzolans. This study aimed to design and produce UHPC using local fly ash available in Colombia. A numerical optimization, based on Design of Experiments (DoE) and multi-objective criteria, was performed to obtain a mixture with the proper flow and highest compressive strength, while simultaneously having the minimum content of cement. The results showed that, despite the low quality of local fly ashes in Colombia, compressive strength values of 150 MPa without any heat treatment can be achieved.

Over the last 20 years, remarkable advances have taken place in the research on reactive powder concrete (RPC). However, because of the high contents of cement and silica fume (SF) usually used in those types of concrete, the cost and environmental impact of RPC is considerably higher than conventional concrete. Hence, the use of supplementary cementitious materials as partial substitution of cement and SF has been an object of great interest by the scientific community. However, the replacement of cement and SF can result in the deterioration of certain properties of RPC, such as the early strength; however, RPC usually needs great amounts of cement and SF. This work presents a study to analyze the effect of metakaolin (MK) as a partial substitute of cement in a previously optimized mixture of RPC using statistical tools such central composite design, main effect plot analysis, and response surface methodology. In addition to MK, supplementary cementitious materials such as SF, limestone powder and recycled glass powder, and fine Type III cement were used. Based on the laboratory experiments results and statistical analysis, it was concluded than MK develops a high activity in the hydration process of RPC, helping it reach high strength at early ages, such as 1 and 7 days, which may be of interest for applications such as the connection of prefabricated elements or accelerated bridge construction. However, the effect of the partial substitution of Type III cement by MK on 28-day compressive strength was nonsignificant. Moreover, the MK inclusion in RPC provides a significant decrease in workability as the amount of MK increases.