Intelligent optimisation of an ultra-high-performance concrete (UHPC) multi-objective mixture ratio based on particle swarm optimisation

ABSTRACT

To achieve specific construction requirements, such as the desired strength and flowability, the proportion of ultra-high performance concrete (UHPC) constituents must be carefully balanced. However, traditional concrete mixture design requires comprehensive trial-and-error experiments, which are cumbersome and costly. The requirements and objectives of a project may be contradictory, rendering the experimental design much more difficult. Therefore, to optimize the mixing ratio of UHPC, particle swarm optimization (PSO) was used to perform multi-objective optimization of the mechanical and flow properties of UHPC. The premise of comprehensive consideration of compressive strength and fluidity was given based on the Pareto optimality theory. The optimal ratio was validated through a combination of the macroscopic force phenomenon and microscopic mechanism. The findings show that the generated PSO algorithm model can simulate and analyze the complete relationship among all functional UHPC components, compressive strength, and fluidity and thus demonstrates its quite high accuracy and efficiency. The macro-micro test results indicate that the optimized UHPC ratio based on the PSO algorithm satisfies the best compressive strength and fluidity. The PSO algorithm can identify the direction for the optimization of the combination of UHPC multiple performance ratios.

KEYWORDS:

• UHPC
• particle swarm optimisation
• multi-objective optimisation
• compressive strength
• fluidity

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China: [Grant Number 51578325].