Computational Approaches for Good Alternative Solutions Based Load Resistance Capacity in Design of Steel Moment-Resisting Frames
A modeling technique to generate good design alternatives from optimal design of steel moment resisting frames is proposed. First presented is the development of Genetic Algorithm(GA)-based approaches that enable designers to identify the best locations for rigid connections in a frame. This GA uses a unique cost function that enables the trade-off study between the number of rigid connections and total cost. Second, the optimization formulation is modified to generate alternatives with a cost comparable to that of the optimal solution. It enables engineers to compare different solutions on the basis of the structure’s lateral load resistance capacity along with incidental factors like the location of the rigid connections, number of different section types, column splices etc. An example of a 5-story and 5-bay steel moment resisting frame is provided to illustrate the effectiveness of the proposed study.
L. Garibaldi, C. Surace, K. Holford and W.M. Ostachowicz
B. H. Choi "Computational Approaches for Good Alternative Solutions Based Load Resistance Capacity in Design of Steel Moment-Resisting Frames ", Key Engineering Materials, Vol. 347, pp. 563-568, 2007