Prediction and modeling of mechanical properties in fiber reinforced self-compacting concrete using particle swarm optimization algorithm and artificial neural network

• Mechanical behavior of fiber reinforced SCC investigated.
• The rheological, and durability properties were tested and compared.
• The PSOA and ANN were used to predict mechanical properties of fiber reinforced SCC.

Intelligence system is a field of computer science that designs and studies efficient computational methods for solving problem. The purpose of present study is to investigate the effects of fibers on the performance of self compacting concrete (SCC). In this experiment study, 9 concrete mixtures containing two types of fibers (polyphenylene sulfide: 0.1, 0.2, 0.3 and 0.4% by volume and steel: 0.1, 0.2, 0.3 and 0.4% by volume) and unreinforced samples have been tested and compared. Fresh, mechanical and durability properties and ultrasonic pulse velocity of all SCC mixtures were evaluated. Then this experimental data was used to train the feed forward artificial neural network type. Finally the trained ANN (artificial neural network) and PSOA (particle swarm optimization algorithm) are used to generate a polynomial model for predicting SCC properties. The obtained results showed that the mechanical properties can be significantly improved by fiber reinforcement and workability of the SCC decreases with increasing fiber content. Moreover, steel fibers have better performance with relation to mechanical properties than polyphenylene sulfide fibers. In addition, PSOA integrated with the ANN is a flexible and accurate method in prediction of mechanical properties of fiber reinforced SCC properties.