Study on deflections of reinforced concrete flat plates under service load

Abstract

Excessive deflections of reinforced concrete slabs can cause severe serviceability problems. In recent years, realistic estimation of slab deflection under service loads has become more important due to the increasing use of high strength materials and due to the ultimate limit state design that generally leads to thinner members. Shortterm deflection of concrete slab is affected by level of cracki ng and cannot be calculated by elastic analysis. Long-term deflection is further increased due to the effect of creep and shrinkage of concrete. Deflection calculations of slabs using nonlinear Finite Element (FE) analysis are complicated and time consuming. The main objective of this work is to develop an easy method of deflection estimation that will be useful in selecting thickness of flat plate slabs. The effect of cracking in flat plate slab can be simulated using a number of nonlinear FE models of which ACI/Branson equation is the simplest. A global plate sti ffness approach incorporating this equation has been used in this cunent work. The performance of ACI/Branson crack model in predicting short- and long-term experimental deflections is reasonably good. However, use of this rigorous nonlinear FE analysis is not particularly suitable for everyday use of the designer. With an aim to develop a simplified method of deflection calculation, a general purpose Artificial Neural Network (ANN) prediction software has been selected. A large number of FE analysis has been carried out on slabs with varying spans, column sizes, loads, material properties etc. and a database have been created for training the ANN prediction tool. The ANN program has been trained using this database until the amount of enor in predicting deflection become very small. Once trained, the prediction tool has been validated against experimental and numel1cal results from previous FE analysis. Use of the trained ANN software to estimate short-and longterm deflections have been demonstrated with example. Now, a designer will be able to estimate deflection of flat plate slab easily for different span, column size, loading, material properties etc. by using the developed ANN prediction tool and will be able to select the appropriate thickness for the flat plate slab.