Numerical Simulation of Streamer Discharge Using Physics-Informed Neural Networks

Abstract

The streamer discharge physical process can be described by the coupled Poisson’s equation and the convection–diffusion equation. It is a multi-physical field problem involving electromagnetism and hydrodynamics. We propose a streamer discharge model based on physics-informed neural networks (PINNs) to improve the computational efficiency regarding the classical approach. Poisson’s equation and the convection–diffusion equation are trained to generate sufficient data and construct a deep operator network (DeepONet). The performance of the PINN, in terms of accuracy, is analyzed by applying it to different datasets (electron density and potential distribution) and comparing to a reference solution (spatial evolution of electrons). The simulation results show that the neural network (NN) has high accuracy in learning two types of equations.