Estimation of electron density in plasma based on back propagation neural network algorithm

Abstract

The plasma sheath produced by high-speed aircraft flying in the atmosphere will form the blackout phenomenon. It can cause communication interruption and navigation failure. It is important to study the electron density distribution of plasma to improve the blackout phenomenon. Due to the complex calculation of plasma electromagnetic wave propagation theory, it is impossible to directly calculate the electron density through the inverse method. To solve this problem, this paper proposes a new algorithm for plasma electron density estimation based on back propagation neural network. Using a large number of data, the relationship model between reflection coefficient and electron density is established. The algorithm can simply obtain the electron density distribution of the plasma by measuring or calculating the electromagnetic wave reflection characteristics. The simulation experiment results show that the mean estimation error of plasma electron density is only 0.1%. This algorithm is very effective for the estimation of plasma electron density and greatly reduces computational complexity.