Prediction of radiation pattern of flexible antenna based on ANN

The electrical performance of flexible antennas changes when subjected to bending in practical applications. To calculate the radiation pattern, traditional methods require multiple simulations, which is time-consuming and laborious. Aiming at this problem, this paper proposes a flexible antenna radiation pattern prediction model based on artificial neural network (ANN). Based on the dataset which includes simulation results of radiation pattern at different bending degree of the antenna, a fully connected ANN framework was established. The model predicts the radiation pattern by using input parameters which are bending degree of the antenna, polar angle θ and azimuthal angle φ of radiation pattern. The prediction model can efficiently predict radiation pattern of flexible antennas with the loss value of 1.95×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-5</sup> and reduce the calculation time of one single state of bending from 1.5h to 12s, contributing to relevant research.