A Prediction Model for Concrete Performance of High-Rise Buildings Based on Particle Swarm Optimization Algorithm

As one of the basic materials of construction, the performance of concrete has a great influence on the safety of buildings. In order to improve the performance of concrete in civil high-rise buildings and increase its performance without changing the cost of concrete preparation, the study proposes a concrete performance prediction model, the core algorithm of which is the Back Propagation Neural Network (BPNN) algorithm, and the hyperparameters of BPNN are optimized using the improved Particle Swarm Optimization (PSO) algorithm. The results showed that the multi-objective optimization model for mixing ratio can improve the performance of concrete without changing or reducing the preparation cost of concrete. After optimizing the mixing ratio, the compressive strength of groups A, B, and C increased, and the slump of most mixing methods decreased, while the preparation cost of concrete remained unchanged. The concrete performance prediction model constructed through research can accurately predict the compressive strength and slump of concrete, and optimize the proportion of concrete.