A Generalized formulation for the discrete optimum design of steel framed structures under the multiple constraints including member buckling constraints is presented as a successive binary programming problem using elastic analysis. This study is an ...
A Generalized formulation for the discrete optimum design of steel framed structures under the multiple constraints including member buckling constraints is presented as a successive binary programming problem using elastic analysis. This study is an extension of the previous research (reference 2) on discrete optimization of steel framed structures considering only the linear combined stress interaction formula which do not include the instability of each member. The design problem is the allocation of member sizes from a catalog of commercially available sections in such a way as to minimize the cost of a structure within the buckling constraints, working stress constraints and/or deflection constraints.
A matrix set of nonlinear buckling constraints combined with working stress constraints is derived by the stiffness method, following the network-topological approach and incorporating the nonlinear working stress interaction formula of the AISC specifications based on the basic theory of column buckling, which is taken from the CRC column formula which, in turn, is derived from the Euler elastic buckling formula and the parabolic adaptation of inelastic buckling theory. This nonlinear constraint set is successively linearized using the Tailor series expansion. The linearized set of constraints is transformed into the constraint set of a successive binary programming problen.
The binary objective function of the successive binary programming formulation for minimum cost design is obtained by associating estimated unit costs of sections with binary variables.
A computer program is developed for the implementation of above formulation. From the results of design examples using this program, it has been concluded that in the discrete optimum design of steel framed structures the buckling constraints should be included and could be easily incorporated into the general frame work of the formulation, and yet it has the same level of efficiency as the previous research on discrete optimization which is formulated without considering instability.