Single-Event Upsets (SEUs) in the space radiation environment pose a serious threat to the reliability of satellite-borne intelligent systems. Traditional fault-tolerance methods such as Triple Modular Redundancy (TMR) and periodic scrubbing face challenges including excessive resource overhead and high power consumption. This paper presents a lightweight fault-tolerance method based on Adaptive Boosting-based Fault-Tolerance Method (AB-FTM) to address SEU vulnerabilities in convolutional neural networks. The proposed approach constructs a heterogeneous ensemble architecture comprising three weak models (ResNet20, ResNet32, ResNet44) and integrated with a dynamic weight adjustment mechanism. By integrating a dynamic weight adjustment mechanism, the method not only significantly reduces the parameter scale (achieving an 18.2% reduction compared to ResNet110) but also enhances classification accuracy, robustness, and fault tolerance. Experimental validation on datasets including CIFAR-10, MNIST, EuroSAT, and Galaxy10 DECals demonstrates that when 0.032‰ of parameters are affected by single-event upsets, the proposed method improves classification accuracy by 53.25%, 63.49%, 57.67%, and 47.43% respectively compared to the TMR-based ResNet110, significantly outperforming traditional triple modular redundancy solutions. This approach provides a novel solution for future space science satellites employing satellite-borne intelligent systems, balancing reliability, lightweight design, and computational efficiency.

In order to solve the problem of frequency stability and transmission accuracy decrease caused by temperature fluctuation during on-orbit operation of an extravehicular load equipment, a highly efficient thermal control scheme is proposed which bases on passive thermal control as the main approach and active thermal control as the auxiliary method. The scheme employs a single-phase liquid cold plate as the main heat dissipation surface, and 10-unit multi-layer insulation components for comprehensive encapsulation. High thermal conductivity materials are utilized to achieve efficient heat conduction. Simultaneously, heating sheets and TEC semiconductor ceramic sheets are used for precise temperature control. Through finite element simulation analysis of temperature distribution under both high and low temperature conditions and optimization design, the temperature variation of key components is controlled within ±0.5 K. The ground constant-temperature thermal balance experiment and the in-orbit data results indicate that this scheme effectively suppresses the interference of temperature fluctuations on the load equipment, significantly improves the overall temperature uniformity of the equipment. It enables photodiodes and other sensitive devices to operate within the optimal temperature range (25-50°C). The temperature change rate of temperature-sensitive devices on the extravehicular load equipment is better than 0.1 K·min^–1, meeting the on-orbit stability requirements of high-precision time-frequency transmission systems. It can provide an important reference for the thermal control design of similar space load equipment.

In response to the performance impact of long cable signal transmission between the search coil and the preamplifier circuit, this paper establishes for the first time a circuit equivalent model of Search Coil-Cable-Preamplifier Circuit. Through simulation analysis and experimental verification, the influence of cable length on the frequency distribution of sensor noise is revealed. Theoretical analysis indicates that cable length has limited impact on sensor sensitivity, but significantly increases the noise level in the high-frequency band (>1 kHz). Based on a prototype search coil magnetometer with a target specification of 10～1000 Hz bandwidth and 30 fT·Hz^1/2 (1 kHz) noise, the variation law of noise with cable length is validated. Experimental results show that as the cable length increases from 3 m to 39 m, the noise corner frequency shifts forward from 7.5 kHz to 2 kHz, while the high-frequency noise at 10 kHz increases by a factor of six. The study finds that an increase in cable length has a significant impact on the noise of inductive magnetometers, specifically manifested as a slight improvement in low-frequency noise and a sharp deterioration in high-frequency noise. Although cable length has a notable effect on inductive magnetometers, its influence can be predicted and mitigated through theoretical modeling incorporating cable parameters. This research provides critical parameter basis for the engineering implementation of search coil magnetometers in space exploration scenarios requiring long-cable applications.

The Grid-based Energetic Neutral Atom imager (GENA) onboard the Queqiao-2 Satellite is the first lunar-based magnetospheric Energetic Neutral Atom (ENA) panoramic remote sensing imaging payload. It aims to provide scientific data support for the study of the injection triggering mechanism of the Earth’s magnetospheric substorm, the magnetotail energy conversion mechanism, and the generation process of the Earth’s magnetic storm through high temporal and spatial resolution observation data. The GENA innovatively combines the time-of-flight method with the two-dimensional coded modulation imaging technology to achieve accurate detection of single-particle events and large-scale, high-resolution magnetospheric neutral atom remote sensing imaging. It can be used to measure 4～200 keV H atoms and 8～250 keV O atoms, and the energy resolution is better than 1 keV @10 keV. At the same time, its instrument field of view is greater than 20°×45°, and the resolution of ENA remote sensing images is better than 0.5°. To accurately calibrate the above performance indicators of the GENA, a study on the calibration method of the GENA was carried out. The accelerator beam calibration test verifies the GENA’s ability to identify the type of neutral atoms and measure their energy, and the optical calibration test verifies the GENA’s field of view coverage and image resolution. The calibration test results show that the GENA’s ability to measure H and O atoms can cover the design indicators.

To address the critical need for efficient image storage and transmission in aerospace applications, this study presents a CCSDS 122.0-B-1-compliant compression core implemented on FPGA. The design incorporates innovative encoding control logic and optimized data organization through co-optimization of algorithmic features and hardware constraints. A segment-based architecture with 256-pixel blocks achieves superior compression efficiency among existing solutions, while effectively containing error propagation through segmented compression. The architecture further enables continuous quality adaptation and progressive image transmission. To resolve performance bottlenecks in scanning and encoding processes, fully parallelized scanning with adaptive parallel encoding was developed, and a 50% efficiency improvement was demonstrated in validation tests. Supporting images up to 4096×4096 pixel with 16-bit depth, the core delivers 90.64×10⁶ sample·s^–1 throughput, meeting operational requirements for diverse space missions.

The lunar surface becomes charged under the influence of solar wind and sunlight, and the potential differences on the surface form an electric field environment of various scales, which is the main mechanism for material transfer on the lunar surface. The electric field probe of Chang’E-7 will, for the first time, conduct in-situ electric field detection on the lunar surface. This paper introduces the design and results of the physical verification test of the probe in plasma environment based on the qualification model of the electric field probe. The basic principle of the electric field probe is the plasma electric probe principle. By clamping the probe current at a specific value, the potential of the probe can be determined according to the plasma V-I characteristic curve. The potential difference between different probes is the potential difference formed by the electric field environment on the plasma environment. Whether the probe can measure the plasma V-I characteristic curve is the key to the success of the probe design. With the help of a ground low-energy plasma simulation device, this test was carried out, and fixed current drive and scanning current measurement were conducted. The test results show that the electric field probe of Chang’E-7 can correctly reflect the plasma environment inside the simulation device and obtain a stable V-I characteristic curve. This proves that the electric field probe can achieve the function of obtaining the plasma potential by driving the probe current when working on the lunar surface, and the physical characteristics of the probe have been fully verified through the test.

The bidirectional Medium-Energy Proton Detector (MEPD) onboard the lunar surface exploration subsystem of the Chang’E-7 lander represents the first-ever implementation of dual-direction medium-energy proton measurements on the Moon. It is capable of providing spectral data of upward- and downward-directed medium-energy protons in the range of 0.03～30 MeV, offering crucial support for modeling the lunar particle radiation environment and for radiation protection in future crewed lunar missions. The unique challenges of ground calibration for the MEPD were addressed in this study. An electron accelerator was employed to achieve proton-equivalent energy calibration, while the full energy range was validated by analyzing the deposited energy of penetrating high-energy protons. In addition, the suppression capability against electron contamination was quantitatively evaluated through a combined approach of accelerator experiments and numerical simulations. The results show that the detector’s energy calibration deviation is better than 3%, its electron-rejection efficiency exceeds 94% for energies at or below 1.4 MeV, and the average geometric factors of the upward-and downward-facing detectors are 0.053 cm^–2·sr^–1 and 0.3041 cm^–2·sr^–1, respectively. These calibration results provide a reliable foundation for in-orbit data inversion. Furthermore, the established calibration and simulation framework offers valuable reference for the future calibration of lunar and deep-space charged-particle detectors.

The Low-Energy Ion Analyzer (LEIA) and Low-Energy Electron Analyzer (LEEA), integral components of the Chang’E-7 lander’s lunar surface environment detection system, conduct in-situ measurements of low-energy charged particles (0.001～30 keV) to elucidate solar wind-regolith interaction mechanisms, investigate microstructure evolution in the lunar near-surface plasma environment, and support space-environment assessment for future lunar research stations. Employing identical hemispherical electrostatic analyzers with asymmetric electrostatic deflectors, both analyzers achieve wide-field detection (90°×360° FOV), broad energy coverage, and voltage-controlled variable geometric factors. Ground calibration using standard plasma beam sources confirmed compliance with mission requirements: energy resolution <15% (ΔE/E), dynamic flux range spanning seven orders of magnitude, and angular resolution <15°×22.5°, collectively enabling comprehensive characterization of lunar surface plasma phenomena.

The lunar pole’s water ice is essential for understanding the Moon’s evolution and building future lunar research station. Nevertheless, existing orbital remote-sensing missions and returned samples remain insufficient to resolve the key unknowns of polar water, including its occurrence modes (e.g., adsorbed water/hydroxyl, pore-filling ice, or ice-cemented regolith), its abundance and vertical variability, and its potential origin and evolutionary pathways. To address this gap, the Chang’E-7 mini-flying probe will carry the Lunar soil Water molecule Analyser (LUWA) to conduct in-situ detection of water ice at the permanently shadowed region for the first time. This paper describes the compositional structure of LUWA and the detection approach, comprising drilling, sampling, sealing, heating and analysis. We detail the calibration parameters, apparatus, and procedures for its three core analytical modules: the Tunable Diode Laser Absorption Spectrometer (TDLAS), the Time-of-Flight Mass Spectrometer (TOF-MS), and the Differential optical Absorption Spectrometer (DAS). The operational chain is designed to release water through thermal extraction, quantify water content with high sensitivity across a wide dynamic range, and enable the determination of D/H isotopic signatures. A systematic ground calibration methodology and a unified calibration framework are established for three functional modules dedicated to water measurement. The framework defines calibration objectives, procedures, and traceability pathways to characterize module response functions, assess background and temperature-dependent effects, verify detection limits and linearity, and evaluate accuracy and repeatability, thereby supporting robust conversion from raw observables to calibrated water-content and isotope products in flight. In parallel, a dedicated lunar in-situ exploration test platform is developed to replicate LUWA’s full operational workflow using lunar regolith simulants, including drilling, sampling, sealing, heating, and analysis. Integrated tests validate key performance metrics and the measurement workflow, providing crucial technical support for interpreting Chang’E-7 LUWA data and for assessing the abundance and occurrence of water ice in lunar soil within PSRs.

Due to the inherent limitations of current remote sensing techniques, the actual occurrence and accurate abundance of water ice in lunar regolith cannot be directly identified, making in-situ verification urgently necessary. The Chang’E-7 mission of China’s Lunar Exploration Program plans to conduct water ice detection at the lunar south pole, where an in-situ micro-sampling device mounted on the rover will perform quantitative collection of lunar regolith for the volatiles in-situ measurement instrument. However, the uncertainty of lunar surface conditions leads to considerable dispersion in sampling mass. Moreover, water ice sublimation loss caused by tool-soil temperature difference and mechanical interactions during sampling will further reduce the detection accuracy. To ensure the reliability of detection data from the volatiles measurement instrument, this paper proposes calibration methods for sampling mass and water ice loss. Icy lunar regolith simulant is prepared via vapor deposition during sampling, and sampling mass calibration experiments are carried out using a self-developed calibration device. Preliminary results show that sampling mass varies significantly with regolith particle size distribution. Further research will be conducted on the physical mechanism of icy lunar regolith sampling under multi-factor coupling conditions. This study aims to provide a high-confidence physical response parameter spectrum for accurate interpretation and scientific analysis of Chang’E-7 in-orbit data.