Predicting the elastic modulus of normal and high strength concretes using hybrid ANN-PSO

In the design and analysis stages, the modulus of elasticity plays a crucial role in influencing the lateral deflection of a reinforced concrete structure. The elastic modulus (EM) of concrete is affected by a wide variety of parameters, some of which are the compressive capacity of the sample, its age, the type of aggregate used, the type of cement used, the loading rate, and the size of the test sample. A precise estimation of the modulus of elasticity in accordance with accepted guidelines is a challenging process requiring specialized loading protocols and continuous strain monitoring. Intelligent systems, particularly artificial neural networks, are now widely used to be considered general and efficient tools for applied research. As the brain does, neural networks process information similarly. To solve a particular problem, a large number of closely connected processing components work simultaneously. To identify weighting factors, other approaches may be utilized; however, the particle swarm optimization (PSO) algorithm was chosen in this research. The algorithm belongs to the field of swarm intelligence and is a method of global minimization that can be applied to problems where the answer is a point or surface in n-dimensional space. An innovative hybrid neural network-based model for determining the EM of samples is presented in this study. In a similar manner to conventional regulations, the proposed models have been determined using the concrete compressive strength parameter simply. To confirm the correctness of the model, a thorough comparison was conducted between the laboratory values, the model findings from this research, and the outcomes of previously established relationships. For comparison, code relations such as ACI 318, ACI 363, FIB, and NS 3473 were used. The suggested model operates excellently and helps determine the EM of concrete, as shown by the results.