Methods for fault diagnosis based on metric learning, in which a query sample is classified by picking the closest prototype from the support set based on their feature similarities, have been the subject of many studies. In real-world applications of in-orbit products, such as circulating pumps, the computation of similarity between different pairs is prone to different degrees of inaccuracy, especially epistemic uncertainty. Knowing and considering the uncertainty of similarity may improve fault detection accuracy. This article provides a unique approach to fault diagnosis based on Prototypical Network (Pro-Net) and Uncertainty Theory. In particular, we use epistemic uncertainty by altering the representation of prototypes from a deterministic scalar to an uncertain representation. To assess the similarity between a query and the prototypes in a support set, we calculate the uncertain distance between the pairs using cross-entropy. Experiments with symmetrical structures reveal that our proposed method significantly enhances classification precision and achieves state-of-the-art performance. It improves the reliability of fault diagnosis and reduces the risk of making erroneous judgments in safety-critical systems, decreasing the possibility of adverse consequences.

An efficient and reliable spacecraft payload thermal management method is one of the key problems to build China Space Station (CSS). Aiming at the outstanding characteristics and difficulties of the payloads rack in the experimental lab module of CSS, in the aspects of thermal control interface, boundary constraint, scale quantity, space layout, diversity of temperature control, operation mode and working life, etc., a two-stage thermal management scheme based on single-phase fluid loop is proposed and constructed. A fluid drive unit is designed to drive the fluid circulation and health management of the module thermal control bus (TCB). In view of the different characteristics of the payload racks, three different types of second-stage thermal control systems are proposed, and different thermal control terminals are designed, in order to solve the problem of thermal control resource allocation and health management in payload racks. A method based on “pump bypass” and “valve resistance ratio” is used to control the flow rate of the thermal management system. Based on the overall scheme of the thermal management system and the developed flight products, on-orbit verification and data analysis are carried out. Flight verification shows that the key parameters of the fluid loop in the thermal management system on orbit are basically consistent with the ground data; the maximum deviation is 1.84%. The pressure control accuracy of the bypass system reaches ±2%, and the active heat transfer efficiency of the integrated thermal management system is over 83%, which can effectively meet the thermal control requirements of the two-stage payloads in the module and rack. After on-orbit flight verification, the designed thermal control system works well on orbit and exhibits good stability.

The supply and exhaust system of the experimental rack on the Chinese space station is a complex integrated system. In this paper, a multi-stage simulation, testing, and verification method is designed for a multi-team, multi-location, and multi-stage integrated gas system. This method is designed to solve the problem of missing input parameters between the gas supply system and the exhaust system. Preliminary tests and strategy verifications were carried out through theoretical simulation and semi-physical simulation, and good calculation results were obtained for the single-rack product. The external systems were tested using a simulation system, and a calculation method was designed to obtain relatively accurate parameters. In the early stage, the performance of the product was predicted using the parameter library of Flomaster and semi-physical simulation methods, but the error was large. In the middle and late stages of development, as some products became realistic, multi-stage testing was carried out using a vacuum simulator, simulated flow resistance, and other methods, achieving a performance prediction with an error of 12% before ground testing. The final ground test and on-orbit test showed that the design and calculation method of this paper is effective. The multi-stage design method proposed in this paper was successfully applied to the integrated gas system of the Chinese space station, which can provide a reference for the design of fluid components in long-term system engineering.