The objective of this study was to establish a technical reliability framework for supplementing concrete with nanoparticles to enhance strength and self-cleaning concrete. The enhancing strength is predicted using the machine learning method of support vector regression (SVR) for self-cleaning concrete. The process was intended to analyze how nanoparticles not only improve the mechanical properties of concretes. The effects of nanoparticles on the ultimate load capacities for two applicable concrete beams applied in building and bridge structures are evaluated by the reliability analysis given by Mont Carlo simulation (MCS). The research mainly aimed at quantifying the effects of nanoparticles on enhancing the safety conditions of beams. Ultimately, the objective is to establish whether nanoparticles can serve high-performance concrete beam structures in terms of enhancing their ultimate conditions.
The Mori–Tanaka model was utilized for evaluating the mechanical properties of reinforced concrete, which are applied in the design of beams using ultimate conditions. By using the data provided based on the nanoparticle inputs as physical and mechanical characteristics, including nanoparticle distributions and their interaction effects, the concrete compressive strength is estimated using SVR due to the reduced computational burden of reliability analysis and its accuracy compared to Kriging and response surface models (RSM). The nonlinear performance function extracted by SVR results and theoretical results are extracted for two types of applications of concrete beams. The failure domains of beams are evaluated using the MCS for extracting the reliability index.
First, the research evaluated elastic modulus and compressive strength for concrete enhanced by nanoparticles; the nanoparticle levels enhanced the compressive performance of concrete materials. Consequently, it has increased the load capacity of the beam by increasing the nanoparticle value fraction. The concrete comparative strength is accurately simulated using SVR compared to Kriging and RSM. The hybrid reliability analysis showed a capable application in the concrete beam structures enhanced by nanoparticles, increasing initial prices of nanoparticles. Nanoparticles proved beneficial performance-enhancing ultimate load capacity for beams by improving the concrete materials as the elasticity modulus increased about 85%, which provided a significant improvement in performance for enhancing reliability levels. The nanoparticles with 0.01%–0.05 in the bridge beam and 0.01–0.1 in the building beam showed the effective enhancement. The depth of beams is the effective variable to increase the safety levels. The area of bars in compression positions is insensitive factor for bridge beams with simply supported conditions while it significantly improves the reliability index in building beams with clumped supports.
This study applies the Mori–Tanaka model to considering the nanoparticles reinforced concrete. Using the Mori–Tanaka model, an evidence-based link between the distribution of nanoparticles and performance improvements in the mechanical properties of nanocomposite concrete are established using the SVR model. The ultimate load capacity of reinforcing beams is approximated by a hybrid analytical-artificial intelligence model which is followed by the experimental results. The safety levels of self-cleaning concrete structures enhanced by the nanoparticles are discussed using a hybrid method. The nanoparticles’ effects are investigated using the reliability index of two applicable concrete beams applied in buildings and bridges with self-cleaning properties given by nanoparticles.
