Spacecraft Single-Event Effect Frequency Prediction Dataset

It is specifically designed for developing frequency prediction models for single-event effects (SEE) in spacecraft. Strictly adhering to the GB/T 44776-2024 industry standard, it employs Latin hypercube sampling for parameter space exploration, covering typical space radiation particles such as protons and heavy ions, as well as the actual radiation environment, shielding structures, and device response parameter ranges encountered by spacecraft in orbit.The dataset comprises 7 input feature categories (differential energy spectra for heavy ions/protons, equivalent shielding thickness, cross-section threshold, LET, etc.) and 5 output label categories (shielded energy spectrum/LET spectrum, cross-section curve, heavy ion/proton/total SEE frequency).

All data were obtained through engineering-grade precision simulations, ensuring strong physical consistency between inputs and outputs. After normalization during preprocessing, the dataset is divided into training, validation, and test sets in a 6:2:2 ratio. It supports training and validation of SEE prediction models such as physics-informed neural networks and deep learning, and can also be used for spacecraft radiation reliability analysis and space radiation effects research.