38. Prediction of Natural Frequencies of Reinforced Concrete Slabs Using Artificial Neural Networks

Floor vibration is a critical serviceability issue in reinforced concrete buildings, requiring accurate estimation of slab natural frequencies during the design process. Although mathematical formulas provide reliable results, they are often computationally and time-consumingly expensive, especially for preliminary design and parametric studies. This study presents an artificial neural network (ANN)-based model for rapid prediction of the natural frequencies of concrete slabs. A dataset is generated from mathematical formulas that analyze vibrations by varying key geometric and material parameters. The ANN is trained using normalized data and its performance is evaluated against independent test samples. The results show that the proposed ANN predicts the fundamental natural vibration frequency with high accuracy, achieving excellent fit with mathematical results and a coefficient of determination exceeding 0.99. Owing to its computational efficiency and accuracy, the proposed approach provides a practical tool for quick vibration assessment of concrete floor systems.