Abstract: To address the structural optimization problem of concrete-filled square steel tubular truss with discrete variables, the imitative full internal force genetic algorithm was used to optimize the structural size of the member section aimed at minimizing the total cost of truss member. The imitative full internal force algorithm was introduced into the imitative full internal force genetic algorithm as an operator, and some initial populations were generated by the imitative full internal force algorithm to increase population diversity. Meanwhile, the penalty function of the traditional genetic algorithm was improved to enhance the optimization efficiency of the algorithm. Comparative analysis of two case studies demonstrates that the cost of the square steel tube concrete truss structure is reduced by 15%, 15% and 45%, 53% respectively by using the imitative full internal force genetic algorithm compared with the imitative full internal force algorithm and the traditional genetic algorithm. It is effective and feasible of the imitative full internal force genetic algorithm for optimizing the concrete-filled square steel tubular truss.

Key words: concrete-filled square steel tubular, truss, imitative full internal force algorithm, structural optimization, hybrid genetic algorithm