Compositional Embedding Architecture for Physical Field Prediction in Componentized Aerospace Systems

2026-09-23

AI Summary

This study addresses the prohibitive computational cost of high-fidelity thermal simulations for spacecraft design and the limited predictive accuracy of existing surrogate models on variable configurations, which struggle to reuse local physical factors. To overcome these limitations, we propose the Tree-structured Factor Composition Network (TFCN). This method decomposes complex configurations into reusable local physical factors and explicitly models their cross-configurational reusability via a tree structure to learn global temperature field responses, thereby transcending the constraints of traditional holistic encoding. In two-dimensional steady-state thermal analysis experiments, TFCN reduces the RMSE by 65.6% and 33.6% compared to the strongest baseline on out-of-distribution tests with 16 and 25 components, respectively. These results demonstrate significantly enhanced generalization to unseen component quantities and combinations, effectively supporting the rapid screening of large-scale configurations.

Abstract

Spacecraft thermal design requires repeated evaluation of how variations in the number and spatial arrangement of heat-generating components and in thermal boundary conditions affect the temperature field. High-fidelity numerical simulations are computationally expensive and therefore difficult to use for large-scale design screening. Although surrogate models can accelerate temperature-field prediction, existing approaches generally encode each complete configuration as a whole and do not explicitly exploit the reusability of local physical constituents across configurations, which limits their accuracy for component counts and combinations not covered during training. To address this issue, we propose the Tree-Structured Factor Composition Network (TFCN), which decomposes complex spacecraft thermal configurations into reusable local physical factors and employs a tree-structured composition module to learn the global temperature-field response associated with different factor combinations. TFCN is evaluated on two-dimensional steady-state spacecraft thermal-analysis cases with prescribed-temperature and radiative-flux boundary conditions. The model is trained exclusively on configurations containing no more than 15 heat-generating components and evaluated on unseen configurations containing 16-25 components. For the prescribed-temperature and radiative-flux cases, TFCN achieves component-count out-of-distribution RMSE values of 4.21 K and 18.62 K, respectively, representing reductions of 65.6% and 33.6% relative to the strongest baseline. These results demonstrate that TFCN improves the reliability of temperature-field prediction under variations in component count and provides an efficient surrogate for rapid spacecraft thermal-design evaluation and large-scale configuration screening.

Problem

• Spacecraft thermal design
• Temperature field prediction
• Surrogate model
• Compositional embedding
• Out-of-distribution generalization

Innovation

• Tree-Structured Factor Composition Network
• Compositional Embedding
• Out-of-Distribution Generalization
• Surrogate Model
• Thermal Field Prediction