Topology optimization for structures under uncertain loads

This paper treats the problem of optimal topology under uncertain loads on structures. The design procedure for the problem is proposed by introducing statistic parameters of the uncertain loads into the newly redefined objective function. Genetic algorithms are used for obtaining optimal topologies for the selected number of the index loads representing the uncertain loads. To determine the optimal topology for uncertain loads, sensitivity analysis. is applied to obtain the weight parameters for the optimal topologies corresponding to the index loads. The optimal problem for minimizing deformation of plate structures is demonstrated to investigate the validity of the proposed design procedure. Random simulations show that the proposed procedure can design the structural topology more reliable for uncertain loads than the optimal design with the deterministic design procedure.