A multi-objective optimization algorithm for forecasting the compressive strength of RAC with pozzolanic materials

Strengthened recycled aggregate concrete (RAC) with pozzolanic materials is considered eco-friendly concrete and the early assessment of its hardened characteristics is vital for the design and implementation procedures. Therefore, developing an accurate approach for modeling the compressive strength of strengthened RAC is essential since the compressive strength value is a necessary variable in different design codes. In this regard, a multi-objective metaheuristic (MSSA-DE) algorithm is proposed to forecast the compressive strength of the strengthened RAC at 28 days of curing age. This algorithm integrates the salp swarm algorithm (SSA) with the differential evolution technique (DE) via a multi-objective fitness function. To this end, the DE algorithm is used for enhancing the ability of feature exploitation in the SSA, where optimized parameters and structural learning are combined in the learning process to simultaneously boost the generalization performance. Then, a random population is produced, and the archive is generated. Finally, the current populations are updated and the solutions are determined. To study the effectiveness of the developed algorithm, a series of experimental and statistical tests are conducted. Results showed that the proposed MSSA-DE model is highly competitive and achieves better prediction performance compared with other metaheuristic models. A sensitivity analysis observed that the cement, water, and fine aggregate contents, as well as the physical properties of recycled aggregate, played a vital role in the model prediction.

Automated summary

A method for accurately predicting the compressive strength of strengthened recycled aggregate concrete was developed, integrating the SSA and DE algorithms to achieve better prediction performance. Sensitivity analysis showed that certain material contents and properties played vital roles in the model prediction.