This study comprehensively investigates the rheological properties of self-compacting concrete (SCC) and their impact on critical parameters, including the migration coefficient, penetration depth of chlorine ions, specific electrical resistance, and compressive strength. A total of 43 mix designs were meticulously examined to explore the relationships between these properties. Quantitative analysis employed a backpropagation neural network model with a single hidden layer to accurately predict the resistant and durable characteristics of self-compacting concrete. The optimal number of neurons in the hidden layer was determined using a fitting component selection method, implemented in MATLAB software(2021b). Additionally, qualitative analysis was conducted using sensitivity analysis and expert opinions to determine the priority of research additives. The main contributions of this paper lie in the exploration of SCC properties, the utilization of a neural network model for accurate prediction, and the prioritization of research additives through sensitivity analysis. The neural network model demonstrated exceptional performance in predicting test results, achieving a high accuracy rate using 14 neurons for predicting parameters such as chlorine penetration depth, compressive strength, migration coefficient, and specific electrical resistance. Sensitivity analysis revealed that xanthan gum emerged as the most influential additive, accounting for 43% of the observed effects, followed by nanomaterials at 35% and micro-silica at 21%.
Considering easy determination of natural frequency in structures leads researchers to focus on detecting the damage through the dynamic parameters using combination of various artificial intelligence algorithms. The main contribution of this research is to detect damage in structures (including its depth and location) for the deep beams with Timoshenko behavior using optimization via simulation (OVS). This method is established based on the first three natural frequencies of the deep or semi-deep beams. The finite element method (FEM) is conducted to obtain essential inputs parameters for OVS. The exact location and depth of the structural damage are determined, using combination of multi-objective optimization algorithms, multi-objective genetic algorithm (MOGA), and modified multi-objective genetic algorithm (MMOGA). This research also remarkably concerns about detecting the location of the defect in the beams with several cracks. In order to verify the results obtained from numerical analysis, several experimental specimens are presented. The dynamics parameters of the beams are experimentally identified using modal hammer. The responses obtained from the numerical method, proposed in this research, are also compared with the results obtained from previous studies. Practically, a beam with real dimensions is examined for different boundary conditions. In addition, the results obtained from MOGA and MMOGA are compared with the other researchers' achievements. Finally, it was observed that the proposed method, (OVS), can be satisfyingly determined the exact location and depth of damage with the high accuracy.