IWOA-RBF Neural Network Ultra-Wideband Antenna Modeling Method

To solve the time-consuming problem of nonlinear data processing and modeling in the antenna structure, a modeling method based on the Improved Whale Optimization Algorithm (IWOA) to optimize the Radial Basis Function (RBF) neural network is proposed. The improved algorithm introduces nonlinear factors in the stage of hunting prey and introduces inertia weights when searching for positions to avoid being limited to local extreme values. In the stage of attacking the prey, the Cauchy strategy is used to improve the global search ability of individual whales and expands the solution space. Finally, the IWOA-RBF network proposed in this paper is simulated and predicted for the ultra-wideband stepped microstrip monopole antenna, and the accuracy of the model is observed by comparing the root mean square error (RMSE) of the return loss. The simulation results verify that compared with the WOA-RBF network and the RBF network, the accuracy of the IWOA-RBF neural network modeling method proposed in this paper is improved by 135.71% and 85.63%, respectively, and the time consumption is shorter, which proves the feasibility of this model.