TY - JOUR ID - 149290 TI - Prediction of Structural Response for HSSCC Deep Beams Implementing a Machine Learning Approach JO - International Journal Of Coastal, Offshore And Environmental Engineering(ijcoe) JA - IJCOE LA - en SN - AU - Mohammadhassani, Mohammad AU - Zarrini, Mahdi AU - Noroozinejad Farsangi, Ehsan AU - Khadem Gerayli, Neda AD - Academic Staff of Seismology Engineering & Risk Department, Road, Housing & Urban Development Research Center (BHRC) AD - Academic Staff, Islamic Azad University, Astanee-Ashrafiye Branch AD - Academic Staff, Department of Earthquake Engineering, Graduate University of Advanced Technology, Kerman AD - Technology management, technology transfer, master of science, transportation research institute, road, housing and urban development research (BHRC) Y1 - 2018 PY - 2018 VL - 3 IS - 1 SP - 35 EP - 43 KW - Deep beam KW - Artificial intelligence KW - Deflection KW - HSSCC DO - 10.29252/ijcoe.2.1.35 N2 - High Strength Concrete (HSC) is a complex type of concrete, that meets the combination of performance and uniformity at the same time. This paper demonstrates the use of artificial neural networks (ANN) to predict the deflection of high strength reinforced concrete deep beams, which are one of the main elements in offshore structures. More than one thousand test data were collected from the experimental investigation of 6 deep beams for the case of study. The data was arranged in a format of 10 input parameters, 2 hidden layers, and 1 output as network architecture to cover the geometrical and material properties of the high strength self-compacting concrete (HSSCC) deep beam. The corresponding output value is the deflection prediction. It is found that the feed forward back-propagation neural network, 15 & 5 neurons in first and second, TRAINBR training function, could predict the load-deflection diagram with minimum error of less than 1% and maximum correlation coefficient close to 1. UR - https://www.ijcoe.org/article_149290.html L1 - https://www.ijcoe.org/article_149290_a6d098c791633bb392a7f51d0893337f.pdf ER -