Particle identification is a critical component of space environment detection. In this study, we propose a CNN-LSTM based multi-class automatic particle classification network, which employs 1-D CNN to extract more discriminative local features of waveforms produced by particle detector, and utilizes the ability of LSTM to capture long-range dependencies for extracting global features with stronger generalization capability. The proposed network can effectively classify the waveforms from multi-type with multi-energy charged particles, achieving an accuracy of 91.72% in simultaneously classifying H, B, C, O, and Al ions across three distinct energy levels. Furthermore, the proposed network achieves an accuracy of 73.60% in classifying particles with unseen energy levels during training, significantly surpassing the accuracy of previous methods. Extensive analysis experiments demonstrate the robustness and effectiveness of the proposed network.

This article aims to conduct an in-depth investigation into the environmental impact of Hall thruster plumes on spacecraft surface charging. The non-uniform plasma plume generated by Hall thrusters may trigger charging and discharging effects, making the assessment of surface charging risks crucial. Through numerical simulations using SPIS system, this study evaluates the surface charging characteristics of a complex spacecraft in orbit, simulating the effects of turning on and off the thrusters, as well as varying distances between the thrusters and the spacecraft. The simulation demonstrates that turning on the thrusters significantly affects spacecraft charging, reducing the potential difference between spacecraft surfaces from 3740 V to 19.2 V, effectively alleviating electrostatic discharge on the spacecraft surface. The closer the thruster is to the spacecraft, the more CEX ions are collected on the surface, influenced by the beam ions, resulting in a surface potential change of 1.3 V, with minor effects on surface potential but contributing to increased deposition contamination on the spacecraft surface.

Space radiation particle detection plays a crucial role in scientific research and engineering practice, especially in particle species identification. Currently, commonly used in-orbit particle identification techniques include telescope methods, electrostatic analysis time of flight (ESA × TOF), time-of-flight energy (TOF × E), and pulse shape discrimination (PSD). However, these methods usually fail to utilize the full waveform information containing rich features, and their particle identification results may be affected by the random rise and fall of particle deposition and noise interference. In this study, a low-latency and lightweight onboard FPGA real-time particle identification platform based on full waveform information was developed utilizing the superior target classification, robustness, and generalization capabilities of convolutional neural networks (CNNs). The platform constructs diversified input datasets based on the physical features of waveforms and uses Optuna and Pytorch software architectures for model training. The hardware platform is responsible for the real-time inference of waveform data and the dynamic expansion of the dataset. The platform was utilized for deep learning training and the testing of the historical waveform data of neutron and gamma rays, and the inference time of a single waveform takes 4.9 microseconds, with an accuracy rate of over 97%. The classification expectation FOM (figure-of-merit) value of this CNN model is 133, which is better than the traditional pulse shape discrimination (PSD) algorithm’s FOM value of 0.8. The development of this platform not only improves the accuracy and efficiency of space particle discrimination but also provides an advanced tool for future space environment monitoring, which is of great value for engineering applications.