Abstract: On the view of the standard genetic algorithm (SGA) having such frequent shortcomings as premature convergence, oscillation and over-randomization in iterative process, an approach for an adaptive genetic algorithm and protection of excellent individuals was presented in this paper. In this method, the sequence of tensioning cables was defined as design variables and the minimum sum of tension acted on cables was taken as the optimized aim. A reasonable sequence of tensioning cables was given by an improved genetic algorithm proposed to optimize the sequence of tensioning cables in the pre-stressed steel structures. This method overcomes the shortage of giving the sequence of tensioning cables empirically in the traditional construction means, which provides a feasible method for optimization of pre-stressed steel structure constructions.