Abstract:The numerical simulations of wind loads are critical in the design of structures. The harmony superposition method (HSM) is most widely used among all the simulation methods for civil engineering structures. Likewise, the time expense of the HSM can be significantly shortened by resorting to the FFT technique and various kinds of interpolation techniques without a significant loss of the accuracy. In the present paper, the RBF neural network interpolation method is introduced into to the traditional HSM (without introducing the interpolation technique), referred to as the RBF neural network based HSM. The RBF neural network based HSM is employed to simulate the wind speed time series at 10 points in a 100-metre building. Two indices, designated as the root mean square error ( ) and error factor ( ), are introduced to measure the accuracy of the proposed approach with respect to the traditional HSM and likewise their time expense is recorded. The numerical results show that the RBF neural network based HSM can render the satisfactory precision and great efficiency.