Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Physics-Informed Machine Learning Towards A Real-Time Spacecraft Thermal Simulator

Manaswin Oddiraju, Zaki Hasnain, Saptarshi Bandyopadhyay, Eric Sunada, Souma Chowdhury

Abstract

Modeling thermal states for complex space missions, such as the surface exploration of airless bodies, requires high computation, whether used in ground-based analysis for spacecraft design or during onboard reasoning for autonomous operations. For example, a finite-element thermal model with hundreds of elements can take significant time to simulate, which makes it unsuitable for onboard reasoning during time-sensitive scenarios such as descent and landing, proximity operations, or in-space assembly. Further, the lack of fast and accurate thermal modeling drives thermal designs to be more conservative and leads to spacecraft with larger mass and higher power budgets. The emerging paradigm of physics-informed machine learning (PIML) presents a class of hybrid modeling architectures that address this challenge by combining simplified physics models with machine learning (ML) models resulting in models which maintain both interpretability and robustness. Such techniques enable designs with reduced mass and power through onboard thermal-state estimation and control and may lead to improved onboard handling of off-nominal states, including unplanned down-time. The PIML model or hybrid model presented here consists of a neural network which predicts reduced nodalizations (distribution and size of coarse mesh) given on-orbit thermal load conditions, and subsequently a (relatively coarse) finite-difference model operates on this mesh to predict thermal states. We compare the computational performance and accuracy of the hybrid model to a data-driven neural net model, and a high-fidelity finite-difference model of a prototype Earth-orbiting small spacecraft. The PIML based active nodalization approach provides significantly better generalization than the neural net model and coarse mesh model, while reducing computing cost by up to 1.7x compared to the high-fidelity model.

Physics-Informed Machine Learning Towards A Real-Time Spacecraft Thermal Simulator

Abstract

Modeling thermal states for complex space missions, such as the surface exploration of airless bodies, requires high computation, whether used in ground-based analysis for spacecraft design or during onboard reasoning for autonomous operations. For example, a finite-element thermal model with hundreds of elements can take significant time to simulate, which makes it unsuitable for onboard reasoning during time-sensitive scenarios such as descent and landing, proximity operations, or in-space assembly. Further, the lack of fast and accurate thermal modeling drives thermal designs to be more conservative and leads to spacecraft with larger mass and higher power budgets. The emerging paradigm of physics-informed machine learning (PIML) presents a class of hybrid modeling architectures that address this challenge by combining simplified physics models with machine learning (ML) models resulting in models which maintain both interpretability and robustness. Such techniques enable designs with reduced mass and power through onboard thermal-state estimation and control and may lead to improved onboard handling of off-nominal states, including unplanned down-time. The PIML model or hybrid model presented here consists of a neural network which predicts reduced nodalizations (distribution and size of coarse mesh) given on-orbit thermal load conditions, and subsequently a (relatively coarse) finite-difference model operates on this mesh to predict thermal states. We compare the computational performance and accuracy of the hybrid model to a data-driven neural net model, and a high-fidelity finite-difference model of a prototype Earth-orbiting small spacecraft. The PIML based active nodalization approach provides significantly better generalization than the neural net model and coarse mesh model, while reducing computing cost by up to 1.7x compared to the high-fidelity model.
Paper Structure (16 sections, 3 equations, 15 figures, 5 tables, 1 algorithm)

This paper contains 16 sections, 3 equations, 15 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Physics Informed Machine Learning (PIML) Methods
  3. Research Objectives:
  4. Differentiable Physics Simulators
  5. View Factor Lookup
  6. Down-sampling Thermal Loads
  7. Temperature Interpolation
  8. PIML Architectures
  9. Training Losses
  10. Case Study: Model Spacecraft
  11. Results
  12. Python Simulator Generated Dataset
  13. Thermal Desktop Generated Dataset
  14. Concluding Remarks
  15. Representative Examples of Predicted Temperature Contours (Python Sim Dataset)
  16. ...and 1 more sections

Figures (15)

  • Figure 1: Generic thermal modeling process for space applications
  • Figure 1: Loss Parameters
  • Figure 2: Architecture of the Python based auto-differentiable finite difference simulator
  • Figure 3: Model Architectures of : a) PIML model designed to be trained on Python simulator data b) PIML model with output shifting, designed to be trained on Thermal Desktop simulator data
  • Figure 4: Case Study Illustrations: a) Spacecraft Structure and Sub-Components b)Spacecraft in Orbit
  • ...and 10 more figures