Optimum cost design of reinforced concrete slabs using neural dynamics model

Abstract

For structural optimization algorithms to find widespread usage among practicing engineering they must be formulated as cost optimization and applied to realistic structures subjected to the actual constraints of commonly used design codes such as the ACI code. In this article, a general formulation is presented for cost optimization of single- and multiple-span RC slabs with various end conditions (simply supported, one end continuous, both ends continuous, and cantilever) subjected to all the constraints of the ACI code. The problem is formulated as a mixed integer-discrete variable optimization problem with three design variables: thickness of slab, steel bar diameter, and bar spacing. The solution is obtained in two stages. In the first stage, the neural dynamics model of Adeli and Park is used to obtain an optimum solution assuming continuous variables. Next, the problem is formulated as a mixed integer-discrete optimization problem and solved using a perturbation technique in order to find practical values for the design variables. Practicality, robustness, and excellent convergence properties of the algorithm are demonstrated by application to four examples.