Session
Technical Poster Session 3
Location
Utah State University, Logan, UT
Abstract
High-power Small Satellites (SmallSats) have the potential to provide new and advanced capabilities for missions beyond low Earth orbit; however, certain challenges prevent wide-spread use. Of these, thermal management of high-heat loads is significant. Therefore, efficient, and effective thermal modeling and analysis methods are becoming increasingly more important, as reliable predictions of thermal behavior in SmallSat components and systems can significantly drive key design and mission-operation parameters. Thermal modeling and analysis efforts have traditionally relied on computer simulations that are often complex and computationally expensive. When properly developed, reduced-order models (ROMs) can overcome these challenges by providing a computationally efficient surrogate that accurately captures the behavior of an underlying high-fidelity thermal model (e.g., Thermal Desktop®) while significantly reducing computational expense. In response to this, a reduced-order modeling approach based on intelligent sampling and robust regression data-fitting methods was developed to predict a SmallSat thermal model’s behavior as a function of a user-defined set of input factors. Multiple use cases are provided, showing reduced-order model efficacy in SmallSat thermal sensitivity studies, thermal uncertainty quantification, and rapid thermal model correlation to thermal test data. Results of the reduced-order model created in these use cases are shown and compared strongly to the underlying Thermal Desktop® models, showing that ROMs are an innovative software technology that can provide significant advancement in Small Satellite missions.
Advanced SmallSat Thermal Analysis Using Reduced-Order Models
Utah State University, Logan, UT
High-power Small Satellites (SmallSats) have the potential to provide new and advanced capabilities for missions beyond low Earth orbit; however, certain challenges prevent wide-spread use. Of these, thermal management of high-heat loads is significant. Therefore, efficient, and effective thermal modeling and analysis methods are becoming increasingly more important, as reliable predictions of thermal behavior in SmallSat components and systems can significantly drive key design and mission-operation parameters. Thermal modeling and analysis efforts have traditionally relied on computer simulations that are often complex and computationally expensive. When properly developed, reduced-order models (ROMs) can overcome these challenges by providing a computationally efficient surrogate that accurately captures the behavior of an underlying high-fidelity thermal model (e.g., Thermal Desktop®) while significantly reducing computational expense. In response to this, a reduced-order modeling approach based on intelligent sampling and robust regression data-fitting methods was developed to predict a SmallSat thermal model’s behavior as a function of a user-defined set of input factors. Multiple use cases are provided, showing reduced-order model efficacy in SmallSat thermal sensitivity studies, thermal uncertainty quantification, and rapid thermal model correlation to thermal test data. Results of the reduced-order model created in these use cases are shown and compared strongly to the underlying Thermal Desktop® models, showing that ROMs are an innovative software technology that can provide significant advancement in Small Satellite missions.
