Single-particle transient effect modeling method based on neural network regression

The invention provides a single-particle transient effect modeling method based on neural network regression, and solves the problems that existing single-particle transient modeling based on a physical mechanism is high in model complexity and is not suitable for circuit-level single-particle effect simulation analysis, and single-particle transient modeling based on specific function fitting is difficult to ensure the precision of the model. The method comprises the following steps: establishing a field effect transistor device structure model; carrying out TCAD simulation on the device structure model, and obtaining a drain transient current pulse curve generated by the single particle effect under different factor conditions; sampling the drain transient current pulse curve to obtain discrete data under different time and different LET conditions; randomly dividing discrete data into a training set, a verification set and a test set; constructing an SET model based on a neural network; and performing lear