Study of Evaporation Duct Height Based on Adaptive Particle Swarm Optimization Algorithm and RBF Neural Network

Evaporation duct is a special atmospheric stratification that frequently appears on the sea surface, which causes over the horizon propagation of electromagnetic wave and electromagnetic blind zone. Accurately and effectively predicting the evaporation duct height (EDH) has important significance for radar detection and over-the-horizon communication on the sea surface. However, traditional methods for inversing EDH has the problems of difficult model building, slow inversion calculation speed and strong dependence on inversion data source. In this paper, the propagation loss data is firstly calculated through the parabolic equation (PE) method. The radial basis function (RBF) method is applied to inverse the EDH. And then the inversion model of RBF neural network based on adaptive particle swarm optimization (APSO) is established. Comparing with the prediction results of traditional RBF and back propagation (BP) algorithm, the predicted value of APSO-RBF model for inversing EDH is consistent with the actual value, and the root-mean-square deviation (RMSE) is very smaller. Moreover, the APSO-RBF proposed in this paper has higher prediction accuracy and lower time consumption, and the overall performance is improved.